Package Details: backintime 1.1.12-4

Git Clone URL: https://aur.archlinux.org/backintime.git (read-only)
Package Base: backintime
Description: Simple backup system inspired from the Flyback Project and TimeVault. QT4 GUI version.
Upstream URL: https://github.com/bit-team/backintime
Licenses: GPL
Submitter: None
Maintainer: graysky
Last Packager: graysky
Votes: 253
Popularity: 2.710046
First Submitted: 2009-01-09 20:46
Last Updated: 2016-06-24 06:25

Latest Comments

graysky commented on 2016-08-04 16:38

No need for the dependency, see: https://wiki.archlinux.org/index.php/Arch_User_Repository#Prerequisites

alejandro.perez commented on 2016-08-04 16:02

Hi,

binutils should be included into the makedepends, as strip is required (otherwise it fails).

Thanks for maintaining this!

graysky commented on 2016-08-01 22:23

@willemw - At the moment, the build will file if you do that...

willemw commented on 2016-08-01 15:17

test() in PKGBUILD should be check().

graysky commented on 2016-06-24 18:45

Use reflector, adjust to your locale:

% which upp
upp () {
for i in 1 2 4 8
do
reflector -c US -a $i -f 5 -p http -p https -p ftp --sort rate --save /etc/pacman.d/mirrorlist.reflector
if [ $? -eq 0 ]
then
cat /etc/pacman.d/mirrorlist.reflector
sudo pacman -Syu
return 0
else
echo "something is fucked up."
fi
done
}

Maxiride commented on 2016-06-24 16:17

EDIT 2: my bad.. dammit still have to learn to run a -Syy regularly. Everything went fine.. I was just not in sync ahah.


EDIT: that's weird, clicking "Download from Mirror" in the package page downloads it without problems.


Sadly one of the mirrors for python-keyring-9.0-1 has been offline for days if anyone else is unable to download it. I'll try to contact his maintainer in the community repo here.

errore: impossibile scaricare il pacchetto 'python-keyring-9.0-1-any.pkg.tar.xz' da archlinux.students.cs.unibo.it : The requested URL returned error: 404
attenzione: impossibile scaricare alcuni file
errore: impossibile eseguire l'operazione richiesta (errore inaspettato)
Si sono verificati degli errori, nessun pacchetto è stato aggiornato.
==> ATTENZIONE: I pacchetti sono stati salvati in /tmp/yaourt-tmp-federico

cp: sovrascrivere '/tmp/yaourt-tmp-federico/backintime-cli-1.1.12-4-any.pkg.tar.xz'? y
[federico@arch-anywhere-federico]: ~>$

graysky commented on 2016-03-28 12:01

@MS and @Tids - Not the fault of this PKGBUILD; makepkg has not supported this feature for some months now: https://lists.archlinux.org/pipermail/pacman-dev/2015-September/020347.html

MaycneSonahoz commented on 2016-03-28 08:49

I also got the "makepkg: invalid option '--pkg'" error.

I got the message when i tried to update backintime and backintime-cli, with both Pamac and yaourt, with the following answers (French language, but the order of questions should be the same):
==> Éditer PKGBUILD ? [O/n] -> n (No)
==> Lancer la compilation de backintime ? [O/n] -> o (Yes)
==> Installer seulement backintime ? [O/n] -> o (Yes)
makepkg: option invalide '--pkg'

For the update to work, you need to say 'No (n)' when asked if you want to install only backintime (3rd question), and likewise for backintime-cli.

graysky commented on 2016-03-25 18:15

@Tids - Why are you passing the --pkg option to makepkg?

Tids commented on 2016-03-25 11:48

for me there is following error

makepkg: invalid option '--pkg'

so it wont build

jmx commented on 2016-03-02 16:25

@Wabuo: I see, thanks for the hint. So I will await the update for this package once upstream sorted out the last quirks...

Wabuo commented on 2016-03-02 16:18

@jmx it is a fixed bug.
For further informations see the issue:
https://github.com/bit-team/backintime/issues/549

jmx commented on 2016-03-02 10:54

Somehow the .desktop for the "root" version of Backintime does not work for me. When I click the symbol from the applications menu, the mouse changes to the busy-icon but after a while it returns to normal with no application window appearing. Not even the prompt to enter my password appears.

But when I enter `pkexec backintime-qt4` into a terminal (which is the Exec= line in the "root" version .desktop), I am instantly presented with the password prompt and Backintime starts fine afterwards.

Why does this work from the terminal, but not from the GUI applications menu? Am I missing some settings somewhere? I'm on GNOME 3.18 from the standard repos.

catalin.hritcu commented on 2016-02-15 18:57

Thanks a lot graysky. I can confirm that Aura works with 1.1.12-2.

graysky commented on 2016-02-15 18:46

@Simone - I honestly can't remember why I added the conflicts/provides arrays. Since they're so specific I'd like to think that I had a good reason. I just removed them and bumped to 1.1.12-2. Try your AUR wrapper.

catalin.hritcu commented on 2016-02-15 18:37

I can confirm what Simone98RC said, Aura doesn't work either.

Simone98RC commented on 2016-02-15 18:18

The required versioning equality between GUI and CLI version makes AUR wrappers like Pamac unable to do automatic updates (because a dependency is requiring the other one updated to same release).
Please prevent that by removing the versioning control, thanks much.

SajeOne commented on 2016-01-19 23:00

@DryEagle I experienced the same issues. It seems AUR helpers( yaourt in my case ) do not install the package properly. I downloaded the tarball and installed it manually just fine.

graysky commented on 2016-01-15 12:49

@aexoxea - Yes, good catch. But there is a problem currently with the test function, see: https://github.com/bit-team/backintime/issues/522

aexoxea commented on 2016-01-15 09:23

Hi, should "test()" actually be "check()" in the PKGBUILD? Everything otherwise working fine with 1.1.12-1; thanks for maintaining!

graysky commented on 2016-01-14 22:57

@DryEagle - Can't reproduce on my system. How are you building it? If you're not using makepkg, use makepkg.

DryEagle commented on 2016-01-14 17:06

won't let me update because it has a dependency of backintime-cli 1.1.8 but it wants to overwrite it to install 1.1.12 and can't. had the same issue at 1.1.10 as well. please fix the pkgbuild

catalin.hritcu commented on 2016-01-10 22:22

Downgrading python-keyring and python2-secretstorage as follows:
warning: downgrading package python-keyring (7.0.1-1 => 6.1.1-1)
warning: downgrading package python2-secretstorage (2.1.4-1 => 2.1.3-2)
makes the Gkr-Message go away.

catalin.hritcu commented on 2016-01-10 21:52

I'm getting the following error when trying to start backintime (both 1.1.10-1 and 1.1.8-4):

Gkr-Message: secret service operation failed: The name org.freedesktop.secrets was not provided by any .service files

Does anyone know what could be causing this problem?

graysky commented on 2015-12-16 19:51

Thanks, ainola. Updated in 1.1.8-4

ainola commented on 2015-12-16 18:02

homepage has moved: https://github.com/bit-team/backintime

dabbill commented on 2015-12-15 03:21

graysky - I know but I am to lazy to remember all the aur packages I have installed to update them manually. :)

graysky commented on 2015-12-14 19:43

@dabbill - Don't an an AUR helper... they are lame :p

dabbill commented on 2015-12-14 19:40

For some reason I keep getting prompted to update when running package manager. Then it just re-installs the same version. Does this no matter how many times I run apacman -Syu.

warning: backintime-cli-1.1.8-3 is up to date -- reinstalling
resolving dependencies...
looking for conflicting packages...

Package (1) Old Version New Version Net Change

backintime-cli 1.1.8-3 1.1.8-3 0.00 MiB

Total Installed Size: 1.09 MiB
Net Upgrade Size: 0.00 MiB

graysky commented on 2015-12-13 20:15

NP, sorry for the glitch with -2.

catalin.hritcu commented on 2015-12-13 20:06

Package 1.1.8-3 works well for me. Thank you.

graysky commented on 2015-12-13 20:02

Sorry about that... could not test on this machine. Please try 1.1.8-3 which has just been pushed.

Germar commented on 2015-12-13 19:41

@graysky please use this patch [1]

[1]: https://gist.github.com/Germar/481f2c201a26c5050697

catalin.hritcu commented on 2015-12-13 18:35

There is an indentation problem with the v1.1.8-2 package
(def __log_keyring_warning should not be indented but it is) that causes this failure:

[hritcu@detained ~]$ backintime
Traceback (most recent call last):
File "/usr/share/backintime/common/backintime.py", line 23, in <module>
import config
File "/usr/share/backintime/common/config.py", line 31, in <module>
import tools
File "/usr/share/backintime/common/tools.py", line 1173
from time import sleep
^
IndentationError: expected an indented block

graysky commented on 2015-12-13 18:24

Bump to v1.1.8-2
Changelog: Fix #473 ahead of next release.[1]

1. https://github.com/bit-team/backintime/issues/473

graysky commented on 2015-12-13 18:23

I had a typo. Please try 1.1.8-2 and verify that it fixes your problems.

catalin.hritcu commented on 2015-12-13 17:29

These are the 2 patches I applied *by hand*:
https://github.com/bit-team/backintime/commit/d7b28fa8f1f124e2e13ad9345b60c61110841485
https://github.com/bit-team/backintime/commit/9ae4960f4dca6092d3b636a3b6def68220b50a7e

graysky commented on 2015-12-13 17:25

@cat - Which patches did you apply and were they to 1.1.8? The two I found do not cleanly apply.

catalin.hritcu commented on 2015-12-13 16:39

Was getting following error on starting up as user:
[hritcu@detained ~]$ Traceback (most recent call last):
File "/usr/share/backintime/common/backintime.py", line 23, in <module>
import config
File "/usr/share/backintime/common/config.py", line 31, in <module>
import configfile
File "/usr/share/backintime/common/configfile.py", line 19, in <module>
import logger
File "/usr/share/backintime/common/logger.py", line 23, in <module>
import tools
File "/usr/share/backintime/common/tools.py", line 1173, in <module>
logger.warning('import keyring failed')
AttributeError: module 'logger' has no attribute 'warning'

Locally applying the 2 patches from issue #473 fixed it for me.
https://github.com/bit-team/backintime/issues/473

Wondering whether you can include those patches in your package until v1.1.10 comes out. Tool not starting at all seems like a big enough problem for regular users.

graysky commented on 2015-09-29 19:01

Bump to v1.1.8-1

Amber-Baum commented on 2015-09-13 10:44

Solides Programm, nutze es schon sehr lange und läuft sauber- Weiter so!

graysky commented on 2015-09-12 11:09

@tahayassen - validpgpkeys= is the Arch standard not just for the AUR but for all packages. Suggest you read the wiki: https://wiki.archlinux.org/index.php/Makepkg#Signature_checking

Germar commented on 2015-09-12 00:24

@tahayassen AFAIK this is a new directive for AUR Packages to make sure you won't install packages which where manipulated by a man in the middle. Maybe the Arch folks can explain this better. I'm just the Dev of BackInTime and have only little experience with Arch...

tahayassen commented on 2015-09-11 23:40

@Germar Why does this package require a separate key? If everyone did this, using the AUR would be quite inconvenient since we'd have to add keys for each package we want to install.

Germar commented on 2015-09-08 17:17

@nielsvd this was answered several times below. You need to run
gpg --keyserver pgp.mit.edu --recv-keys 615F366D944B4826
once to add my key.

nielsvd commented on 2015-09-07 20:22

Installer is not working:

==> Verifying source file signatures with gpg...
backintime-1.1.6.tar.gz ... FAILED (unknown public key 615F366D944B4826)
==> ERROR: One or more PGP signatures could not be verified!

graysky commented on 2015-09-02 18:10

gksu is no longer required for root access; the preferred method is using pkexec.

coderkun commented on 2015-09-02 12:05

Running backintime as root works under Gnome running
$ export QT_X11_NO_MITSHM=1
$ gksu backintime-qt4

see https://bugs.archlinux.org/task/42030

graysky commented on 2015-08-22 17:11

@willemw - So it has, thanks. Updated.

willemw commented on 2015-08-22 16:39

Package 'kdesdk-kompare' has been renamed to 'kompare'.

finswimmer commented on 2015-08-19 12:12

I have the same problem. Running backintime-qt4 as root only shows empty windows.

arm1e commented on 2015-08-06 08:27

I am also having these errors when trying to start the root version. Blank windows if done from terminal and nothing starts at all when using the menu icon in gnome. Are there any known fixes yet?

Wabuo commented on 2015-06-13 12:34

Yes I have, for testing I also installed polkit-qt4 polkit-qt5 polkit-gnome but nothing changed.

If I try to launch it with gksu I get these errors:
Qt: Session management error: None of the authentication protocols specified are supported
X Error: BadAccess (attempt to access private resource denied) 10
Extension: 130 (MIT-SHM)
Minor opcode: 1 (X_ShmAttach)
Resource id: 0x156
X Error: BadShmSeg (invalid shared segment parameter) 128
Extension: 130 (MIT-SHM)
Minor opcode: 5 (X_ShmCreatePixmap)
Resource id: 0xd9
X Error: BadDrawable (invalid Pixmap or Window parameter) 9
Major opcode: 62 (X_CopyArea)
Resource id: 0x220001b
X Error: BadDrawable (invalid Pixmap or Window parameter) 9
Major opcode: 62 (X_CopyArea)
Resource id: 0x220001b
X Error: BadDrawable (invalid Pixmap or Window parameter) 9
Major opcode: 62 (X_CopyArea)
Resource id: 0x220001b

graysky commented on 2015-06-12 18:56

Do you have polkit installed?

Wabuo commented on 2015-06-12 15:01

backintime-root wont start =(
the only thing I could find was this line in journalctl:
backintime-qt4-root.desktop[1303]: Refusing to render service to dead parents.

Where can I find the log files from backintime-root?

Someone some tips and tricks ?

Wabuo commented on 2015-06-12 15:01

backintime-root wont start =(the only thing I could find was this line in journalctl:
backintime-qt4-root.desktop[1303]: Refusing to render service to dead parents.

Where can I find the log files from backintime-root?

Someone some tips and tricks ?

Wabuo commented on 2015-06-12 14:58

backintime-root wont start =(
the only ting i could find was this line in journalctl:
backintime-qt4-root.desktop[1303]: Refusing to render service to dead parents.

Where can I find the log files from backintime-root?

Someone some tips and tricks ?

Rhinoceros commented on 2015-05-24 07:16

Does anyone know how to get the system tray icon working using Plasma 5? It worked at some stage in KDE 4, but I think stopped working at some point.

uffe commented on 2015-05-23 00:06

@Germar that could very well be the problem - will check that - thx

Germar commented on 2015-05-22 16:38

@uffe maybe you're bitten by this bug? https://bugs.launchpad.net/backintime/+bug/1437623

uffe commented on 2015-05-22 16:09

The last versions of backintime does not report done state to the status bar.
The last message ever written in the status bar is 'Saving permissions...' - the 'Done' message never appears.
Have anyone seen this problem ?

graysky commented on 2015-05-04 19:22

Yep, works without the larger package set. Updated and thanks.

graysky commented on 2015-05-04 19:04

Hmm... would need to spin up a VM to verify.

Germar commented on 2015-05-04 18:52

@CanalGuada @graysky AFAIK BIT only uses /usr/bin/xdpyinfo. On Debian this is in package x11-utils. But as Arch has separate packages for each, it should be enough to only use xorg-xdpyinfo instead of xorg-utils. Didn't test this yet as I have no working Arch at hand at the moment.

CanalGuada commented on 2015-04-30 16:21

Have got an issue with nautilus file manager (properties dialog box didn't show without clicking on the main window), under GNOME desktop (didn't check further) after installing this package.

Resolved building Qt4 GUI without xorg-utils dependency (on my own desktop, had only to install xorg-xlsatoms, xorg-xlsclients, xorg-xvinfo and xorg-xwininfo packages, that were missing in order to complete this meta package) that seems (at least wholly) unnecessary here.

10479 commented on 2015-03-26 08:09

@MightyPork - It was answered for you the last time you asked. You need to add the key server 'gpg --keyserver pgp.mit.edu --recv-key 615F366D944B4826'

MightyPork commented on 2015-03-25 14:04

I'm getting error "(unknown public key 615F366D944B4826)" when I try to do makepkg.

luuuciano commented on 2015-02-19 14:29

When I tried to install it fails with an error:
loading packages...
error: 'backintime-cli-1.1.2-2-any.pkg.tar.xz': duplicate target
Any idea??

PD: oooops, have to use regular makepkg & pacman instead packer :)
thanks a lot

luuuciano commented on 2015-02-19 14:21

When I tried to install it fails with an error:

loading packages...
error: 'backintime-cli-1.1.2-2-any.pkg.tar.xz': duplicate target

Any idea??

graysky commented on 2015-02-06 20:17

@wille - You are right... I missed the comment from before. Fixed in 1.1.2-2.

willemw commented on 2015-02-06 09:02

Typo: brz --> bzr. (Already mentioned by me: there is no need to list *-bzr.)

graysky commented on 2015-02-05 19:57

Bump to v1.1.2-1

Changelog: See CHANGES in source tarball. Main change is no more need for gksu, use pkexec.

Germar commented on 2015-01-31 19:54

Ah okay. Missed your second post before answering. So this would pin down the problem to Qt4. Could you reinstall that and all dependencies?

lockheed commented on 2015-01-31 19:50

Other than qtconfig-qt4 (which is affected by the same issue), I know of no other QT apps that I have.

Germar commented on 2015-01-31 19:49

Did you check if other Qt4 Apps run straight with root?

lockheed commented on 2015-01-31 19:48

I mean: the qtconfig-qt4 window displays no content. Vicious circle...

lockheed commented on 2015-01-31 19:47

@Germar,
What you wrote gave me an idea to change QT theme for root. So I run "sudo qtconfig-qt4". But... I run into exactly the same problem.

Germar commented on 2015-01-31 19:39

I can't explain nor reproduce it neither, sorry.
Last update came with huge changes especially for GUI (it's now pure Qt4).
I only have two thinks you could check:

1. on my Arch-VM BackInTime started as root shows up with this ugly, Win95-style default theme. Not the theme you selected. Maybe there is no such default theme on your machine, or your selected theme does make some troubles with this?

2. next version 1.1.2 (coming soon) will call 'pkexec backintime-qt4' for root. You could try out if this helps by coping http://bazaar.launchpad.net/~bit-team/backintime/trunk/view/head:/qt4/net.launchpad.backintime.policy to /usr/share/polkit-1/action/ and run 'pkexec backintime-qt4'

lockheed commented on 2015-01-30 20:38

Here are screenshots:
Asking whether I want to import old config: http://i.imgur.com/Ffpuuf7.png
Asking for location of the old config: http://i.imgur.com/ul5WDW0.png

graysky commented on 2015-01-30 20:35

I can't explain it nor can I reproduce it :/ Perhaps Germar [one of the BIT devs] has some ideas.

lockheed commented on 2015-01-30 20:32

@graysky, I created a new empty /root folder and run sudo backintime, but the problem persists. Up until the recent upgrade, I did not have such issues.

graysky commented on 2015-01-30 20:03

@lockheed - I found this if I didn't have a ttf installed which is why I added the meta packages ttf-font to the qt4 package. For me, I can call backintime as my user or root via sudo and it displays just fine. Not sure what's wrong with your system. Does your /root dir have anything in it that would override or ignore system fonts? Just for the hell of it, can you create a fresh /root dir with no profiles or the like and see if you get that same behavior?

Germar commented on 2015-01-30 12:15

@lockheed Could you please post a screenshot?

lockheed commented on 2015-01-30 12:08

@Germar, why would a missing font affect root, but not user? I installed this package but the problem persists.

Germar commented on 2015-01-30 11:57

@lockheed sounds like missing fonts. Did you install 'ttf-font' from dependencies?

lockheed commented on 2015-01-30 10:57

I can run backintime as user fine, but if I run it as root, all the qt4 dialog boxes are empty gray slabs of space in the shape and size of windows of user-run backintime.

optomus commented on 2015-01-25 23:52

Just missed your comment. Thank you!

optomus commented on 2015-01-25 23:52

This rectified my situation.
gpg --keyserver pgp.mit.edu --recv-keys 615F366D944B4826
I am assuming it was due to the lack of a keyserver in the gpg.conf file though it is unfortunate that https://wiki.archlinux.org/index.php/Makepkg#Signature_checking does not have a more thorough description or a link to https://wiki.archlinux.org/index.php/GnuPG as it made this process confusing.
Thank you for maintaining!

huw commented on 2015-01-25 23:49

Optomus (and MightyPork), you need to specify a keyserver, like this:

gpg --keyserver pgp.mit.edu --recv-key 944B4826

optomus commented on 2015-01-25 23:24

I am having the same problem as MightyPork. I am on a fresh install and having read the link posted, I found that I did not have a ~/.gnupg/pubring.gpg.
I tried to then add the key with"gpg --recv-keys 615F366D944B4826" and recieve "gpg: keyserver receive failed: No keyserver available".
I know this is not a help board though I have spent a few hours now trying to rectify the problem and this is likely the most direct route.

graysky commented on 2015-01-24 14:30

https://wiki.archlinux.org/index.php/Makepkg#Signature_checking

MightyPork commented on 2015-01-24 12:25

Not sure if I'm doing something wrong, but it fails for me with
backintime-1.1.0.tar.gz ... FAILED (unknown public key 615F366D944B4826)

Germar commented on 2015-01-19 20:22

Qt4 GUI in 1.1.0 will run on Gnome, KDE and all other DE's. No need for two separate GUI's any more \o/

graysky commented on 2015-01-19 19:56

No, with version 1.1.0 it is qt4 only.

gatonero commented on 2015-01-19 18:24

Isn't there a gtk or gnome version?

graysky commented on 2015-01-17 15:19

Sorry, that should have been ttf-font which I have now corrected.

adrianx commented on 2015-01-17 15:00

warning: cannot resolve "ttf", a dependency of "backintime"

It works when I remove "ttf" as a dependency.

adrianx commented on 2015-01-17 14:56

warning: cannot resolve "ttf", a dependency of "backintime"

Not sure if there is a package called "ttf". I use infinality-bundle, if that makes any difference.

willemw commented on 2015-01-17 12:57

The 'depends' and 'conflicts' lists are lists of "provide" names, not package names. Each package implicitly provides 'pkgname'.

If correctly defined in *.bzr packages:
- depends=(name of the non-devel default provide name)
- conflicts=(name of the non-devel default provide name)
then your point 2) will not occur, because the package manager will see a package conflict between, for example, 'backintime' and 'backintime-bzr'.

graysky commented on 2015-01-17 12:47

OK..
1) removed python from deps since python-keyring pulls it.
2) are you sure that your brz packages don't try to overwrite key files this pkg provides?

willemw commented on 2015-01-17 12:42

'python2-keyring' and 'python2-secretstorage' already pull in 'python2'.

All the *-bzr packages here already 'provide' and 'conflict' the non-bzr package names. A normal package should usually not know about the related devel package.

For another thread maybe:

I have now a problem with the new backintime-bzr (QT4) / backintime-cli-bzr: cannot use version numbers in 'depends' as you do, because they are *-bzr packages, so when upgrading backintime-bzr (qt4), there is no message that says to also upgrade backintime-cli-bzr, in case that is installed. Note: backintime-cli maybe be installed instead of backintime-cli-bzr. Actually, I tried 'depends=("backintime-cli-bzr=$pkgver" ...) in 'backintime-bzr' and it seems to work, but I am not sure it is allowed (in the AUR).

graysky commented on 2015-01-17 12:10

@wllemw - Yes, trying to see how to allow the PKGBUILD to show a set of packages has optdeps... might need to open a forum thread. Also, why is python not needed as a dep and why shouldn't I conflict with the vcs packages?

willemw commented on 2015-01-17 08:24

Some feedback on the PKGBUILD file itself (hope you don't mind): the dependencies in the PKGBUILD file are are not equivalent to the dependencies in the backintime README file, however maybe you are still working on that. Also, you should not have to specify 'python' and the '*-bzr' packages.

graysky commented on 2015-01-16 21:05

OK gang! Germar has released version 1.1.0 which a qt4 gui. I have a draft PKGBUILD that I would like you guys to test out. The conflicts/replaces and deps give me working packages on my system running xfce4 and on another system running lxde. I would like to hear some feedback before I push it to the AUR.

https://gist.github.com/graysky2/4ffb10c81801960518bf

huw commented on 2015-01-11 21:14

@gatonero

That worked. Thank you very much.

gatonero commented on 2015-01-11 21:09

@huw
See the link to https://bbs.archlinux.org/viewtopic.php?pid=1360283#p1360283 !

1) Download the tarball
2) Extract it
3) cd into it
4) makepkg -src
5) Manually install the needed packages using pacman -U ...

for example: sudo pacman -U backintime-1.0.34-4-any.pkg.tar.xz
sudo pacman -U backintime-gtk-1.0.34-4-any.pkg.tar.xz

huw commented on 2015-01-11 19:30

When trying to install this, the script downloads backintime-gnome and backintime-gtk then complains that these two conflict. The packages are only optional dependencies of backintime so why are they being downloaded at all?

petaramesh commented on 2014-12-31 07:38

1. == ERROR: One or more PGP signatures could not be verified! :-(

graysky commented on 2014-12-30 10:25

Bump to v1.0.40-3
Changelog: PKGBUILD now to fully supports new version of makepkg. If the build ends in an error[1], users need only need to grab Germar's key[2] and run makepkg again. Do not sign or trust the key. Do not attempt to add the key to pacman's keyring.

1. ==> ERROR: One or more PGP signatures could not be verified!
2. gpg --keyserver pgp.mit.edu --recv-key 944B4826

graysky commented on 2014-12-29 14:29

The new pacman update looks for the key to be in your keyring and fails if it isn't.

makepkg --skippgpcheck

kartapux commented on 2014-12-29 12:38

Same as chem.tand : cannot build (One or more PGP signatures could not be verified!)

noplomplom commented on 2014-12-29 09:07

Cannot build: fails to verify PGP signature (unknown public key)

graysky commented on 2014-12-28 19:22

Bump to 1.0.40-2

Changelog: Updated source and url.

graysky commented on 2014-11-25 21:47

Scroll down in the comments. In short, do not build with yaourt until upstream switches over to qt.

Profpatsch commented on 2014-11-25 21:44

Cannot install with yaourt, always get „:: backintime-gnome and backintime-gtk are in conflict“ and have to install the .tar.gz afterwards.

graysky commented on 2014-11-13 20:09

Bump to 1.0.40-1

Changelog: http://backintime.le-web.org/change-log

catalin.hritcu commented on 2014-09-29 06:32

@naraesk: I've already filed this bug against aura: https://github.com/aurapm/aura/issues/218
They say it will be only fixed in Aura2 though.

graysky commented on 2014-09-28 21:13

Use makepkg.

naraesk commented on 2014-09-28 20:39

sudo aura -A backintime
aura >>= Determining dependencies...

It got stuck there, determining dependencies for over an hour yet. Is this a problem caused by aura or are there any problems with this package?

graysky commented on 2014-09-28 19:51

I had a look at that bug report but don't know that I can add anything to the content since I have no experience using BIT to read or write to removable media... :/

Germar commented on 2014-09-28 19:19

@graysky could you please join the discussion on https://bugs.launchpad.net/backintime/+bug/1374343 ?

Germar commented on 2014-08-18 19:45

@graysky - No, Qt4 GUI will come with 1.1 release. I am keeping 1.0 releases for bug fixing and add new features only to 1.1. But I'm still working on it as there are still some problems.

graysky commented on 2014-08-18 19:42

@turorg - Changed in 1.0.36-3.

tuborg commented on 2014-08-18 19:12

While you're at it the kde4/backintime-kde4-root.desktop command should be kdesu instead of kdesudo. Thank you both for the great work.

graysky commented on 2014-08-18 18:59

@Germar - I sure did thinking you fixed it upstream. Replaced now in 1.0.36-2 with a comment for myself that you won't be changing it.

While you're here reading this, didn't you plan to remove support for gtk entirely with the 1.0.36 release?

Germar commented on 2014-08-18 13:03

@graysky did you accidentally remove 's/gksu/gksudo/g' patch from new PKGBUILD (take a look at my comment from 2014-03-03)? This should stay as I will not change this upstream. Other distries work fine with gksu.

catalin.hritcu commented on 2014-06-22 14:34

Trying to delete a snapshot ate up all memory causing my machine to become super unresponsive without actually deleting anything. Filed a new bug about this: https://bugs.launchpad.net/backintime/+bug/1332969

catalin.hritcu commented on 2014-06-22 08:49

Trying to delete a snapshot ate up all memory causing my machine to become super unresponsive without actually deleting anything. It seems related to a bug reported since 2012: https://bugs.launchpad.net/backintime/+bug/992026

Jristz commented on 2014-06-09 22:19

You can add a systemd.timer (and therefor cronie make optdep)
I mean like the one in prey-tracker (user need to activate it to work, like a normal systemd.service) so those like me that want Not have cronie in they system can use it without edit the pkgbuild every time is updated

willemw commented on 2014-06-04 06:50

build() always builds all make targets. For example, it builds 'gnome' and 'kde4' make targets for the 'backintime' package.

graysky commented on 2014-06-03 21:26

disable_keyring_for_root.patch is included in the source tarball... not sure what your problem is with this.

gatonero commented on 2014-06-03 21:09

It had been very nice to insert the link to https://bbs.archlinux.org/viewtopic.php?pid=1360283#p1360283 here!

1) Download the tarball
2) Extract it
3) cd into it
4) makepkg -src
5) Manually install the needed packages using pacman -U ...

for example: sudo pacman -U backintime-1.0.34-4-any.pkg.tar.xz
sudo pacman -U backintime-gtk-1.0.34-4-any.pkg.tar.xz

Germar commented on 2014-06-03 20:50

Click on 'Download tarball' and extract it...

gatonero commented on 2014-06-03 20:45

Where do I get the source disable_keyring_for_root.patch?

gatonero commented on 2014-06-02 17:55

I downloaded PKGBUILD. When executing makepkg -src in the same directory I got
==> FEHLER: disable_keyring_for_root.patch wurde nicht im build Verzeichnis gefunden und ist keine URL.

graysky commented on 2014-06-01 20:45

Don't use a wrapper script when building split packages that provide subpackages that conflict with eachother. Just use `makepkg -src` and then `pacman -U ...`

gatonero commented on 2014-06-01 19:43

Please give some detaoled explanations how to update this package.

==> Continue installing backintime ? [Y/n]
==> [v]iew package contents [c]heck package with namcap
==> ---------------------------------------------------
==>

Lade Pakete ...
Löse Abhängigkeiten auf...
:: Es stehen 2 Provider für kactivities zur Verfügung:
:: Repositorium extra
1) kactivities-frameworks 2) kactivities4

Geben Sie eine Zahl ein (Voreinstellung=1):
Suche nach Zwischenkonflikten...
Fehler: Nicht lösbare Paketkonflikte gefunden
Fehler: Konnte den Vorgang nicht vorbereiten (In Konflikt stehende Abhängigkeiten)
:: backintime-gnome und backintime-gtk stehen miteinander in Konflikt
==> WARNING: Your packages are saved in /tmp/yaourt-tmp-cs
cp: „/tmp/yaourt-tmp-cs/backintime-1.0.34-4-any.pkg.tar.xz“ überschreiben? n
cp: „/tmp/yaourt-tmp-cs/backintime-gnome-1.0.34-4-any.pkg.tar.xz“ überschreiben? n
cp: „/tmp/yaourt-tmp-cs/backintime-gtk-1.0.34-4-any.pkg.tar.xz“ überschreiben? n
cp: „/tmp/yaourt-tmp-cs/backintime-kde4-1.0.34-4-any.pkg.tar.xz“ überschreiben? n
==> ERROR: unable to update

graysky commented on 2014-05-30 20:54

1.0.34-4 is just a proper split package as the AUR now supports them; no other difference from the -3 release.

graysky commented on 2014-04-28 19:00

@catalin - backintime is the base package, the other three flavors provide GUIs without forcing users to pull down deps if they don't want it. The next release of backintime will be only qt so the gnome and gtk packages will disappear when it is released. Until then, you will have to use makepkg and manually install what you want.

catalin.hritcu commented on 2014-04-28 17:23

About the aura problem, it's quite justified given that this package depends on itself. WTF?

catalin.hritcu commented on 2014-04-28 17:18

> Build the packages with $ makepkg and install only either -gnome or -gtk but not both.

This works, but it's a strange work-flow, and it confuses both yaourt (the conflict willll reported) and aura (Determining dependencies... forever).

Why not split this into 3 separate packages?

orschiro commented on 2014-04-26 18:28

@willll

Build the packages with $ makepkg and install only either -gnome or -gtk but not both.

willll commented on 2014-04-26 15:50

I have the following issue :

looking for inter-conflicts...
error: unresolvable package conflicts detected
error: failed to prepare transaction (conflicting dependencies)
:: backintime-gnome and backintime-gtk are in conflict

snassar commented on 2014-04-15 07:59

After finding dependencies in AUR, backintime 1.0.34-3 built cleanly and installed cleanly.

snassar commented on 2014-04-15 07:47

Build failure: http://pastie.org/9081518

python2-secretstorage in particular can't be fixed either in from Arch repositories or AUR.

DaveCode commented on 2014-04-09 07:23

Build failure report - all deps installed OK prior to extract:
http://pastebin.archlinux.fr/499483

neitsab commented on 2014-04-08 22:03

Do you think it would be problematic to replace gksu with polkit/pkexec? I have written polkit actions and authorization rules for a couple of graphical applications that needed privilege escalation (nautilus, gedit, meld, baobab, bleachbit) mostly to open and edit stuff as root, and it all works great. I did so because I was wary of gksu deprecation, but it is required by BIT installation and now I'm wondering whether I could replace it as well. Any thoughts from the more knowledgeable?

orschiro commented on 2014-04-04 11:42

@Germar

Because I do not want to shutdown but hibernate. But I don't think this feature needs to be added to BIT. It's just a special use case for me. :)

Germar commented on 2014-04-04 09:03

@orschiro
why don't you use BIT's new shutdown button? It will automatically shutdown your system after new snapshot is done.

orschiro commented on 2014-04-04 05:43

@Germar

Thank you. I deleted the message file and the error disappeard. Can very well be that it was a result of a failed backup (I usually backup and set a hibernation timer which sometimes could have been to short to complete the backup).

Germar commented on 2014-04-03 21:17

@orschiro
I guess this was just some leftovers from a previous failed backup cron job. Status bar will display messages from ~/.local/share/backintime/worker.message. It may happen that BIT failed (e.g. because you shutdown your system during the cron job was starting) and left the old worker.message which will now spam your status bar. Just remove worker.message will solve this.

orschiro commented on 2014-04-03 06:42

@Germar

I encountered an interesting issue.

In the bottom status bar, BIT reports that it cannot find the snapshot folder.

However, in the sidebar all snapshots are correctly displayed.

See this screenshot: http://i.imgur.com/LY4voDm.png

A known bug?

Germar commented on 2014-03-11 21:30

@jeroenMathon python2-secretstorage is in depends in PKGBUILD. But for what ever reason it's not shown on this site. Take a look at grayskys comment from 2014-01-28 20:44.

JeroenMathon commented on 2014-03-11 20:10

@graysky
It seems you forgot to add Python2-secretstorage to the dependencies list.

Heres a url ^_^ https://aur.archlinux.org/packages/python2-secretstorage

JeroenMathon commented on 2014-03-11 20:09

It seems that the package Python2-Secretstorage is not in the Dependencies list.

Here is the url for that package.
https://aur.archlinux.org/packages/python2-secretstorage

orschiro commented on 2014-03-06 07:29

@Germar

Thanks! Renaming the folder did it. Now everything seems to be fine, including the udev rule:

~ cat /etc/udev/rules.d/99-backintime-root.rules
ACTION=="add", ENV{ID_FS_UUID}=="2aac80d3-3f44-41dc-9f99-18f251d518dd", RUN+="/usr/bin/su - 'root' -c 'test -x /usr/bin/anacron && test -e /root/.config/backintime/anacrontab-2aac80d3-3f44-41dc-9f99-18f251d518dd && /usr/bin/anacron -s -t /root/.config/backintime/anacrontab-2aac80d3-3f44-41dc-9f99-18f251d518dd -S /root/.local/share/backintime/anacron' &"

Germar commented on 2014-03-04 14:08

@ottomops and @orschiro
this is because BIT is using 'Auto Host/User/ProfileID' (General Settings below Advanced) by default and 'User' now changed to 'root' (how it should be). But your existing profile was configured with User 'orschiro' or 'ottomops' or whatever your normal user is called.
I'd suggest to keep 'gksudo' but manually deactivate 'Auto Host/User/ProfileID' and change 'User' back to your normal users name. Or alternative rename the snapshot folder '/path/to/backintime/<HOST>/<USER>' into '/path/to/backintime/<HOST>/root' if there are only snapshots for root inside.

orschiro commented on 2014-03-04 12:57

@Germar

I can confirm what ottomops says. My previously made backups are not longer recognised. Though now I can successfully create the udev rule.

So what is the optimal solution? gksu or gksudo?

ottomops commented on 2014-03-04 09:21

Since using gksudo my backup folder is not recognized anymore by backintime. Changing the starter back to gksu fixed it for me. So I would vote for going back to gksu.

graysky commented on 2014-03-03 21:10

Bump to v1.0.34-3

Changelog: Per Garmar's request, modified gnome/gtk .desktop file to use gksudo rather than gksu.

Germar commented on 2014-03-03 13:03

@orschiro
You are already root because of "gksu backintime-gnome". That's why you are not asked for your root pw again.

I just found out that Arch gksu behave different than Debian gksu. "gksu env | grep -e '^USER=\|^LOGNAME='" should print root for both but on Arch it will print orschiro which causes your problems. Using "gksudo backintime-gnome" instead should solve this.

@graysky
Could you please change 'gnome/backintime-gnome-root.desktop' to use 'gksudo'?

orschiro commented on 2014-03-03 05:52

@Germar

When you said that something is wrong with my user, I deleted the udev rule and started from scratch. I launched `gksu backintime-gnome` and tried to recreate the udev rule. I selected Schedule: When drive gets connected.

But then I get the error: `Failed to create udev rules`

The CLI output:

Profile: Arch Linux System
Automatic backup: When drive get connected (udev)
Command "gksudo -m "Please provide your sudo password to install the udev rule." -- cp "/tmp/tmpFn8HNV" "/etc/udev/rules.d/99-backintime-orschiro.rules"" returns 256

That is, I am not asked for my root password.

qqqqqqqqq9 commented on 2014-03-03 00:32

Thanks for clearing it up that it's all about teaching us not to use AUR-Helpers and run makepkg by hand. An Optdepend wouldn't suck at all, the only "harm" it would do is an additional line being printed out during installation.

Germar commented on 2014-03-02 17:04

@orschiro
There is something wrong with the user in that udev rule (/usr/bin/su - 'orschiro' [...] vs. /root/.config[...]). It should be either 'orschiro' OR 'root'. But not mixed.
How did you start BIT? With sudo or gksu?

orschiro commented on 2014-03-02 08:28

@mojtabazn

You have to build the package manually and then selectively install only backintime-gnome or backintime-gtk but not both

mojtabazn commented on 2014-03-02 05:04

i have this problem for install :

error: unresolvable package conflicts detected
error: failed to prepare transaction (conflicting dependencies)
:: backintime-gnome and backintime-gtk are in conflict

orschiro commented on 2014-03-01 07:50

I have problem with the automatic udev rule. Back In Time created the following rule in /etc/udev/rules.d but whenever I plug in my device, I am not prompted for any password, nor is the backup process started automatically:

99-backintime-orschiro.rules:

ACTION=="add", ENV{ID_FS_UUID}=="2aac80d3-3f44-41dc-9f99-18f251d518dd", RUN+="/usr/bin/su - 'orschiro' -c 'test -x /usr/bin/anacron && test -e /root/.config/backintime/anacrontab-2aac80d3-3f44-41dc-9f99-18f251d518dd && /usr/bin/anacron -s -t /root/.config/backintime/anacrontab-2aac80d3-3f44-41dc-9f99-18f251d518dd -S /root/.local/share/backintime/anacron' &"

orschiro commented on 2014-02-26 15:23

@Germar

Thanks! I was so free to add this information to the wiki page:

https://wiki.archlinux.org/index.php/Back_In_Time

Germar commented on 2014-02-26 13:50

@orschiro you can use backintime-gnome as normal user even if you want to use udev schedule. BIT will ask for your sudo password to create udev rules.
You only need to run 'gksu backintime-gnome' if you want to backup/restore files which are only rw by root.

orschiro commented on 2014-02-26 10:53

Since the disabling of the keyring I am a bit confused how to launch backintime. For instance for some settings (such as creating the auto-backup udev rule) I need root permissions.What do you run?

~ backintime
~ backintime-gnome
~ gksudo backintime-gnome
~ gksu backintime-gnome

Thanks for shedding light!

orschiro commented on 2014-02-21 11:46

@zapataz

Then build the package manually and only install backintime-gtk/gnome

zapataz commented on 2014-02-21 11:39

@Orschiro

With Yaourt

orschiro commented on 2014-02-21 08:14

@zapataz

How did you build and try to install backintime?

Using an AUR helper?

zapataz commented on 2014-02-21 07:58

Hi,

Got package conflict error during build (below). I removed backintime-gnome references in PKGBUILD and it worked.

loading packages...
resolving dependencies...
looking for inter-conflicts...
error: unresolvable package conflicts detected
error: failed to prepare transaction (conflicting dependencies)
:: backintime-gnome and backintime-gtk are in conflict

graysky commented on 2014-02-17 20:17

@qqqq - Extra dependencies suck is the point. Germar mentioned that the next major release will drop the gtk/gnome flavor in favor of a qt GUI only. See his comment from 2013-12-19 20:20.

qqqqqqqqq9 commented on 2014-02-17 10:58

Hi,
what's the point of the backintime-gnome version (beside making it uninstallable by aur-helpers)? The only difference between backintime-gnome and backintime-gtk is gnome-session, which is not even installed during the build.
Couldn't you just make gnome-session an optdepend of backintime-gtk and drop backintime-gnome?

graysky commented on 2014-01-28 20:44

It is in the PKGBUILD... not sure why the AUR isn't listing it:

depends=('openssh' 'python2' 'rsync' 'cron' 'python2-keyring' 'python2-secretstorage')

ryuzy commented on 2014-01-28 19:05

Dependency python2-secretstorage is not listed

graysky commented on 2014-01-24 19:39

Bump to v1.0.34-2

Changelog: fix slow startup/unresponsiveness with BIT running as root

graysky commented on 2014-01-24 19:39

Sorry, missed your note about the patch.

Germar commented on 2014-01-12 19:18

@graysky please use this patch to fix slow startup/unresponsiveness with BIT running as root: http://pastebin.com/znjrQKtT

It will disable keyring for root completely as there is no use of it and it just causes annoying troubles.

orschiro commented on 2014-01-09 06:44

@Germar

Unfortunately the unresponsiveness (maybe related to what petaramesh reported about slow startup) returned. The workaround of creating /root/.cache does not help to reduce this lagging. This is the content of /root/.cache

[root@thinkpad orschiro]# ls -la /root/.cache
total 36
drwx------ 6 root root 4096 Jan 8 10:43 .
drwxr-x--- 19 root root 4096 Jan 9 06:43 ..
drwx------ 2 root root 4096 Nov 11 00:12 dconf
-rw-r--r-- 1 root root 12288 Dec 1 09:12 event-sound-cache.tdb.thinkpad.x86_64-unknown-linux-gnu
drwxr-xr-x 2 root root 4096 Apr 7 2013 gstreamer-1.0
drwx------ 2 root root 4096 Nov 11 00:12 keyring-SeKFqn
drwx------ 3 root root 4096 Apr 7 2013 thumbnails

Germar commented on 2014-01-06 20:57

@petaramesh this is only with BIT running as root, correct? If so this is caused by https://bugzilla.gnome.org/show_bug.cgi?id=711810.
For workaround you can either create '/root/.cache' or set environ 'BIT_USE_KEYRING=false' before starting BackInTime. E.g. 'gksu env BIT_USE_KEYRING=false backintime-gnome'

petaramesh commented on 2014-01-06 09:43

Latest update causes backintime to now take about ONE FULL MINUTE at startup, before opening (waiting while nothing seems to happen, assuming the program startup failed, but well... it didn't...), or before reacting when shifting profile.
Happens on 2 different machines here.
Besides that... It works.

graysky commented on 2013-12-21 17:49

Bump to v1.0.34-1

Changelog: http://backintime.le-web.org/change-log

Germar commented on 2013-12-21 17:28

We released 1.0.34 today.
Regards,
Germar

blackshark commented on 2013-12-19 20:22

Yea ofc. Thanks for your efforts guys!

Germar commented on 2013-12-19 20:20

We are currently working on a new major version 1.1.0 which will come with only one Qt4 GUI. It will run on all DEs so there is no need for splitting anymore. So maybe you could live with the current situation for a couple more month until we are done with 1.1.0?

blackshark commented on 2013-12-19 19:29

Imho the simplest would be to split into different packages with dependencies. But actually I do not master this kind of stuff, so I can't help you on this... :/

graysky commented on 2013-12-19 19:24

Perhaps you are referring to the group options? I'm not sure how to deal with that for this split PKGBUILD. Basically, I can remove the -gtk option but then non-gnome users will have extra deps to use BIT which is something I don't want to force on people. I am open to suggestions to make this easier.

blackshark commented on 2013-12-19 19:21

Sure, I understand, and thanks for your job! If I remember when you install Gnome (all the environment) that ask you which package you want to install.

graysky commented on 2013-12-19 19:15

I'm doing my best to keep the PKGBUILD as close to the Distro Standards as possible. Adding a selector to it would be over the top indeed. Which gnome packages offer this?

blackshark commented on 2013-12-19 19:10

@graysky: Why do not implement a selector (like for gnome packages) ? It's pretty impratical to update/install your package. It's a pity because your work is really good!

graysky commented on 2013-12-19 12:18

Just build first, and install second.

makepkg -src
makepkg -U ...

fleaz commented on 2013-12-19 12:17

@graysky: How do I select if I want bit-gnome or bit-gtk? I cant't find the part in the PKGBUILD where I can select if -gnome or -gtk should be installed

graysky commented on 2013-12-14 23:31

You got it... plz post here when you want it bumped.

Germar commented on 2013-12-14 23:26

Please hold back new release 1.0.32 for couple days. There is a second serious issue in this release (BIT root will shutdown system). We will release a new version soon.

graysky commented on 2013-12-14 10:34

You can only install one of them... if running gnome use -gnome, if not, use -gtk. Look at the PKGBUILD, -gnome has more dependencies.

piedro commented on 2013-12-14 09:31

I get the message that dependencies can't be resolved: backintime-gnome and backintime-gtk ...

None of these is installed. What am I missing?

p.

orschiro commented on 2013-11-10 23:14

@Germar

~ $ sudo /usr/bin/gnome-keyring-daemon --start --foreground --components=secrets
GNOME_KEYRING_CONTROL=/root/.cache/keyring-SeKFqn
GNOME_KEYRING_PID=20470

Germar commented on 2013-11-10 20:40

The patch is still necessary cause there are other problems with keyring, too. Please post the output of the Exec command in '/usr/share/dbus-1/services/org.freedesktop.secrets.service' running with sudo.
This should be 'sudo /usr/bin/gnome-keyring-daemon --start --foreground --components=secrets'

orschiro commented on 2013-11-10 10:48

I can confirm what graysky reports. `gksu backintime-gnome` causes a timeout of ~20 seconds although /root/.cache exists.

graysky commented on 2013-11-10 10:45

So on my system, I do have /root/.cache but running `gksu backintime-gnome` causes a delay of around 24 sec before the GUI appears.

graysky commented on 2013-11-10 10:32

...is the patch still needed if /root/.cache is indeed present?

graysky commented on 2013-11-10 00:44

I can add a check for this to the readme.install if needed...

Germar commented on 2013-11-09 23:18

Found the reason for that timeout on 'gksu backintime-gnome':
/root/.cache is missing :D

Just run 'sudo mkdir /root/.cache' and all should be fine again.
Take a look at https://bugs.launchpad.net/ubuntu/+source/gnome-keyring/+bug/1181344

orschiro commented on 2013-11-05 12:53

@Germar

It raises the exception:

~ $ gksu -- python2 -c "import keyring"
[...]
dbus.exceptions.DBusException: org.freedesktop.DBus.Error.NoReply: Did not receive a reply. Possible causes include: the remote application did not send a reply, the message bus security policy blocked the reply, the reply timeout expired, or the network connection was broken.

I have no idea what caused the delay in the first place, but I think it is working now responsively. In case I have this issue again, I will report again.

Thanks!

Germar commented on 2013-11-04 22:18

Hmm, strange. My last patch should set 'BIT_USE_KEYRING=false' automatically after 'import keyring' first time raised an exception and block subsequent 'import keyring'. Looks like this doesn't work on your system.

Please post the output of 'gksu -- python2 -c "import keyring" '
This should normally raise dbus.exceptions.DBusException

orschiro commented on 2013-11-04 21:45

@Germar

I did not patch it manually but installed the updated PKGBUILD.

Yet I can confirm that your first hack is working like a charm. The GUI is responsive and startup fast as hell. :)

~ $ BIT_USE_KEYRING=false gksu backintime-gnome
gksu-run: :0
gksu-run: gksu/backintime-gnome/7765-0-thinkpad_TIME0
gksu-run: d2939297000f65e4dbf60514332d3568

WARNING: import keyring failed

Germar commented on 2013-11-04 20:32

@orschiro did you manually patch /usr/lib/python2.7/site-packages/keyring/backends/SecretService.py to not throw an exception anymore?

Please try to run 'BIT_USE_KEYRING=false gksu backintime-gnome' (this should start BIT immediately without timeout on start). If that will still give you an unresponsive GUI please comment out lines 28-32 in /usr/share/backintime/common/tools.py (starting from 'try:' to 'os.putenv(...)' including those lines)

If this both doesn't fix it I'd say there is an other problem on your system ;-)

orschiro commented on 2013-11-04 07:16

@Germar

Thanks for the patch. However, the GUI is still very unresponsive. One example is profile switching which takes ~45 seconds.

But never mind. Can be also a problem on my system only.

graysky commented on 2013-11-04 00:58

Bump to v1.0.28-6

Changelog: Applied upstream patch #2 to fix gksu issues.
Commit: http://pkgbuild.com/git/aur-mirror.git/commit/backintime?id=e83d55b1cdf14fd1d0a361c9d9956c2709a74001

graysky commented on 2013-11-03 23:21

Bump to v1.0.28-6

Changelog: Applied upstream patch #2 to fix gksu issues.

Germar commented on 2013-11-03 20:06

@graysky please replace gksu_backintime-gnome.patch with the one in http://pastebin.com/hPAk2JXt

This will fix the unresponsive GUI reported by orschiro.

Germar commented on 2013-10-31 20:40

@orschiro I can confirm the unresponsive GUI now. There is something repeatedly trying to 'import keyring' again and again and always running into the same timeout. That shouldn't happen. I'll have a look.

orschiro commented on 2013-10-31 20:08

@Germar

Sorry for having been sloppy. The ~30sec startup time for the GUI to come up is totally fine for me. And I thought your workaround referred only to this.

Yet the unresponsive GUI is slightly annoying since I can never be sure if my button clicks where already recognised and simply take time to be processed.

Thus I wanted to make sure if it is related to the workaround as well or another problem on my system.

Germar commented on 2013-10-31 14:24

@orschiro do you remember your post from 22. Oct? "I have no problem to wait ~30sec". It takes a while because I added the workaround for the bug in python-secretstorage which will always run into a timeout on root account.
I'm not sure about if it's generally slower and unresponsive. Maybe this is just because you are getting bored during waiting for GUI to come up?

orschiro commented on 2013-10-31 06:34

Do you encounter the same that when opening Backintime (root), it takes a while for the interface to show up and the program behaves quite slowly and feels unresponsive?

Gruentee commented on 2013-10-30 17:43

Thanks, coderkun, for offering your repo!

orschiro commented on 2013-10-25 09:16

@coderkun

Thanks! The key is now signed. I must have overlooked this part.

coderkun commented on 2013-10-25 08:47

@orschiro: As it’s written in the Wiki¹: “After verifying the fingerprint, you need to locally sign this key:
# pacman-key --lsign-key keyid”


¹ https://wiki.archlinux.org/index.php/Pacman-key#Adding_unofficial_keys

orschiro commented on 2013-10-25 08:44

@coderkun

[orschiro@thinkpad Downloads]$ sudo pacman-key -a coderkun.asc
==> Updating trust database...
gpg: next trustdb check due at 2014-01-22

But then after running pacman -Syyu:

coderkun-aur 149,3 KiB 26,9K/s 00:06 [############################] 100%
coderkun-aur.sig 72,0 B 0,00B/s 00:00 [############################] 100%
error: coderkun-aur: signature from "coderkun <olli@coderkun.de>" is unknown trust
error: failed to update coderkun-aur (invalid or corrupted database (PGP signature))

coderkun commented on 2013-10-25 08:22

@orschiro: You can download the key file¹ and use “pacman-key -a” to import it. I think it is time for me to upload the key.

¹ http://arch.coderkun.de/coderkun.asc

orschiro commented on 2013-10-25 08:14

@coderkun

How do I add your key and sign the packages?

[orschiro@thinkpad ~]$ sudo pacman-key -r A6BEE374
gpg: requesting key A6BEE374 from hkp server pool.sks-keyservers.net
gpgkeys: key A6BEE374 not found on keyserver
gpg: no valid OpenPGP data found.
gpg: Total number processed: 0
==> ERROR: Remote key not fetched correctly from keyserver.

coderkun commented on 2013-10-25 06:22

My repo with a few [AUR] packages I regularly build: http://arch.coderkun.de

orschiro commented on 2013-10-25 06:13

@coderkun

Thanks. I would be fine with your repo. And why not using what you already have. Can you post a link to your repo?

coderkun commented on 2013-10-25 06:13

I have a small private repo with a few packages I build from [AUR]—including backintime. But since I’m not a trusted user or something like that, you are probably more comfortable with graysky’s suggestion.

coderkun commented on 2013-10-25 06:11

I have a small private repo with a few packages I build from [AUR]—including backintime. But since I’m not a trusted user or something like that, maybe you are more comfortable with graysky’s suggestion.

orschiro commented on 2013-10-25 06:00

@graysky

Personally I think it would be great if you can set up a repo for precompiled packages!

Thanks ahead

graysky commented on 2013-10-24 20:35

The split package is IMO the way to package Backintime since there are common components and specific components with unique dependencies. I understand that it breaks the functionality of AUR helper scripts. I don't want to break it into 4 different PKGBUILD since all are created from the same source tarball. That would require users to download a minimum of two different PKGBUILD files (common + whatever GUI).

Is there interest in an unofficial repo to build precompiled packages for backintime+deps? I am reluctant to just throw them into repo-ck but perhaps I can spin up another sub-repo specifically for the BIT packages.

blackshark commented on 2013-10-24 20:14

That's sux. Why don't split it in distinct packages ?

orschiro commented on 2013-10-24 20:10

@blackshark

Unfortunately yes.

blackshark commented on 2013-10-24 20:09

Ok.. but I have to do manual updating ? If yes this isn't a good new.

orschiro commented on 2013-10-24 20:05

@blackshark

Yes. See the comments below. The latest big update split backintime into different packages.

blackshark commented on 2013-10-24 19:42

Is it normal than I can't install BIT with Yaourt ? I need to perform update by pacman -U...

petaramesh commented on 2013-10-23 09:25

It works now \o/ Thanks !

orschiro commented on 2013-10-23 06:20

@Germar

Thanks so much! I am finally again able to take a (root) backup. Keep on your work on this excellent piece of software!

Germar commented on 2013-10-22 21:50

Thanks!
You didn't test the new udev schedule yet. So there is no udev rule to delete. If you do I'd appreciate some feedback.

graysky commented on 2013-10-22 21:35

...and welcome to Arch!

graysky commented on 2013-10-22 21:35

On my system, only the first line in the readme.install applied; the other two returned nulls (fine with the -f switch). Thanks again for your active development of this excellent util.

Germar commented on 2013-10-22 20:44

@graysky I was uncomfortable with fiddling around in home, too. But with 'purge' people expect to have a complete clean start again and where complaining that the config is still in there (most times their problems where somewhere else than config)
Anyways, removing the udev rules and .pyc should be enough.

graysky commented on 2013-10-22 20:28

Bump to v1.0.28-5

Changelog: Applied upstream patch to fix gksu issues.

graysky commented on 2013-10-22 20:28

@Germar - Done and done. Your question about a purge function is, I think, answered in the readme.install file which I included. I kept a few lines of the script you referenced, but I think it's an Arch no-no to modify any file in a user homedir so I omitted that bit. Look in the readme.install to see.

graysky commented on 2013-10-22 20:26

@hal - Read down a few comments; AUR helper scripts do not seem to handle split packages with the dep structure well.

Germar commented on 2013-10-22 18:46

@graysky please append this patch[1]. It will fix both the 'gksu backintime-gnome' problem and the 'backintime-askpass' problem. 'python2-gnomekeyring' is now save to be removed completely but keep 'python2-secretstorage' in deps.

I'm not sure if there is something similar to 'apt-get purge' in Arch (I'm quite new to Arch). But if it is, it would be wise to implement the same as we do for Debian/Ubuntu in 'common/debian_specific/postrm'[2]. This will purge all config, udev rules and other leftovers.

[1] http://pastebin.com/PgijHfkL
[2] http://bazaar.launchpad.net/~bit-team/backintime/trunk/view/head:/common/debian_specific/postrm

iromeister commented on 2013-10-22 15:06

I had the same problem as hal and managed to install backintime-kde4 by running

pacman -U /tmp/yaourt-tmp-hal/backintime-1.0.28-4-any.pkg.tar.xz
pacman -U /tmp/yaourt-tmp-hal/backintime-kde4-1.0.28-4-any.pkg.tar.xz

(adapted to your username, hal)

iromeister commented on 2013-10-22 15:04

I had the same problem as hal and managed to install backintime-kde4 by running

pacman -U /.../backintime-1.0.28-4-any.pkg.tar.xz
pacman -U /.../backintime-kde4-1.0.28-4-any.pkg.tar.xz

[...] censored ;-)

hal commented on 2013-10-22 13:44

Hello!

I've got the following:

loading packages...
resolving dependencies...
looking for inter-conflicts...
error: unresolvable package conflicts detected
error: failed to prepare transaction (conflicting dependencies)
:: backintime-gnome and backintime-gtk are in conflict
==> WARNING: Your packages are saved in /tmp/yaourt-tmp-hal

(x86-64)

hal commented on 2013-10-22 13:43

Hello!

I've got the following:

loading packages...
resolving dependencies...
looking for inter-conflicts...
error: unresolvable package conflicts detected
error: failed to prepare transaction (conflicting dependencies)
:: backintime-gnome and backintime-gtk are in conflict
==> WARNING: Your packages are saved in /tmp/yaourt-tmp-hal

orschiro commented on 2013-10-22 05:15

@Germar

I vote for the second option. I have no problem to wait ~30sec. I need a stable and reliable Backup software so the first option is no real option. I don't want the program to crash during backup.

Germar commented on 2013-10-21 23:08

Nope, it's not fixed yet. I can make a patch for you guy's but I'm still waiting on comments which solution you want (take a look at my comments below):

1. turn back like it was done in the fix for 1.0.26 and risk a segfault or
2. catch the exception thrown by 'import keyring' and deactivate storing to keyring in that case. This will probably add a ~30sec timeout on every 'gksu backintime-gnome'

The 3rd option to port BIT from gtk 2.x to Gtk+ 3.x is not an option as it would take several weeks or months.

So please let me know what you want.

graysky commented on 2013-10-21 22:27

It should not. Germar is one of the upstream devs. Perhaps he can answer your question.

orschiro commented on 2013-10-21 21:17

@graysky

Thanks for your effort! I was able to install -any and -gtk.

Yet 1.0.28-4 does not fix the gksu problem yet, correct?

graysky commented on 2013-10-21 20:41

Bump to v1.0.28-4
HOW TO INSTALL:

1) For this version bump you do actually have to uninstall the previous version of the package.
2) Install the common package at the same at as you optionally install a GUI like this: sudo pacman -U backintime-gtk-1.0.28-4-any.pkg.tar.xz backintime-1.0.28-4-any.pkg.tar.xz

Changes to PKGBUILD:
* name change /qt4/kde4/
* broke out the 'backintime' package which is common to all GUIs
* added openssh as a dep per Germar's comment
* added sshfs and encfs as deps per Germar's comment (note that 'cron' included anacron, at least it does with cronie the Arch default)
* builds backintime-notify now (long standing gap)
* changed backintime-kde4 deps per Germar's comments

Germar commented on 2013-10-21 20:03

@graysky I had a look on PKGBUILD and would like to add some notes:
- openssh should be in depends array as it contain ssh-agent which is started with every BIT session because of a bug in ssh-add
- optdepends should have sshfs, encfs and maybe anacron if that isn't installed by default on Arch.
- you don't build backintime-notify at all (notify folder in tarball). This is a BIT-plugin for notifying user about errors.
- backintime-qt4 (or backintime-kde4) doesn't need python2-gnomekeyring. python2-secretstorage or python-keyring-kwallet (don't know which package this is in Arch) will do the job on KDE
- backintime-qt4 (or backintime-kde4) should have 'kompare' at least in optdepends

Kind regards,
Germar

graysky commented on 2013-10-21 09:00

@germar - I will change the qt4 suffix to kde4.
@petara - I will break out a common package from the gutless, gnome, and kde4 variants as well as fix the kde4/gtk/gnome simultaneous deps. I won't be able to do it until later today though.

petaramesh commented on 2013-10-21 08:04

@graysky I'm not using makepkg/pacman, I'm using the excellent "yaourt" tool (check https://wiki.archlinux.org/index.php/yaourt ) that includes the full functionalities of pacman, plus automatizes compilation/installation of AUR packages.

yaourt builds the packages, and installs every produced package it finds. If the build produced several final packages, it installs them all. If the build produced mutually-incompatible packages, then it breaks.

Also, backintime-kde4 and backintime-gnome were *not* incompatible in previous packages versions, and I'd like to have both installed, because I have both environments on my systems, and different users use different environments, so having both backintime versions installed is great.

What used to work 3 days ago, doesn't work anymore, that's why I say it's now broken.

Germar commented on 2013-10-20 22:13

Oh, I forgot an other option:
I could just catch the exception that is thrown on 'import keyring' and deactivate storing passwords in keyring in that case (there would still be a ~30sec timeout). While root normally doesn't have a Keyring (and so 'gksu backintime' neither) this wouldn't make any difference.

I think I would preferred this option ;)

Germar commented on 2013-10-20 22:05

@graysky you're right. This has nothing to do with the package split. It is a bug in python2-secretstorage (upstream, not in Runiq's AUR package).

I don't know how to fix this by now. But I would like to discus the options:
With python2-keyring 3.X the python2-gnomekeyring fail silently and refuse to work. This is because BIT statically import gtk with 'import gtk' (old variant with gtk 2.0) and python2-gnomekeyring dynamically import Gtk with 'from gi.repository import Gtk' (new variant with Gtk+ 3.0).

As a workaround in the fix I provided for 1.0.26 I imported 'keyring' before 'gtk'. So gnomekeyring was working again but there was 'Warning: g_boxed_type_register_static: assertion 'g_type_from_name (name) == 0' failed' during every start of BIT and I was afraid this could cause a segfault in BIT.

So I decide to remove 'import keyring' before 'import gtk' again and let gnomekeyring fail silently. python2-secretstorage does the same and should replace python2-gnomekeyring. If all would have went well you could remove python2-gnomekeyring.

As long as the bug in python2-secretstorage isn't fixed we could go back like it was done in the fix (just add 'import keyring' at the start of gnome/app.py, remove python2-secretstorage but keep python2-gnomekeyring) and risk a segfault.

An other fix would be to port BIT from gtk 2.X to Gtk+ 3.0 which would take enormous effort. And in long terms I'm actually thinking about removing backintime-gnome completely and only provide a pure backintime-qt4 (didn't discus this with anyone yet). The current backintime-kde4 uses lots of KDE bindings but I think they could be transformed to pure Qt4 with relative low effort. This would make future dev lot easier because we don't need to handle two GUI's anymore. The Gnome GUI is quite a mess in code anyways. KDE GUI was rewritten and is more straight.

BTW: the 'backintime-qt4' package name is misleading. As I stated above it uses KDE4 not pure Qt4.

graysky commented on 2013-10-20 20:31

@Germar - Please advice on a solution since this seems to be an upstream bug. Also, you wrote, "python-keyring changed from 1.6.1 to 3.X and we added python2-secretstorage dependencies as replacement for python-gnomekeyring."

That is not totally accurate since the pkg depends on both:
* python2-secretstorage
* python2-gnomekeyring

Is this correct or is python2-secretstorage meant to actually replace python2-gnomekeyring? In other words, should I modify the dependency array?

graysky commented on 2013-10-20 20:28

@orschiro - I don't think that statement is accurate, in other words, the split package has nothing to do with this error. I have old versions of my packages (or you too may download one from the aur-mirror.git[1] for this package). I can verify that indeed 1.0.26-6 can reproduce this bug:

% gksu backintime-gnome
Traceback (most recent call last):
File "/usr/share/backintime/gnome/app.py", line 23, in <module>
import keyring
File "/usr/lib/python2.7/site-packages/keyring/__init__.py", line 12, in <module>
from .core import (set_keyring, get_keyring, set_password, get_password,
File "/usr/lib/python2.7/site-packages/keyring/core.py", line 141, in <module>
init_backend()
File "/usr/lib/python2.7/site-packages/keyring/core.py", line 63, in init_backend
keyrings = backend.get_all_keyring()
File "/usr/lib/python2.7/site-packages/keyring/util/__init__.py", line 24, in wrapper
func.always_returns = func(*args, **kwargs)
File "/usr/lib/python2.7/site-packages/keyring/backend.py", line 127, in get_all_keyring
exceptions=TypeError))
File "/usr/lib/python2.7/site-packages/keyring/util/__init__.py", line 35, in suppress_exceptions
for callable in callables:
File "/usr/lib/python2.7/site-packages/keyring/backend.py", line 119, in is_class_viable
keyring_cls.priority
File "/usr/lib/python2.7/site-packages/keyring/util/properties.py", line 20, in __get__
return self.fget.__get__(None, owner)()
File "/usr/lib/python2.7/site-packages/keyring/backends/SecretService.py", line 28, in priority
secretstorage.Collection(bus)
File "/usr/lib/python2.7/site-packages/secretstorage/collection.py", line 35, in __init__
collection_obj = bus_get_object(bus, SECRETS, collection_path)
File "/usr/lib/python2.7/site-packages/secretstorage/util.py", line 42, in bus_get_object
return bus.get_object(name, object_path)
File "/usr/lib/python2.7/site-packages/dbus/bus.py", line 241, in get_object
follow_name_owner_changes=follow_name_owner_changes)
File "/usr/lib/python2.7/site-packages/dbus/proxies.py", line 248, in __init__
self._named_service = conn.activate_name_owner(bus_name)
File "/usr/lib/python2.7/site-packages/dbus/bus.py", line 180, in activate_name_owner
self.start_service_by_name(bus_name)
File "/usr/lib/python2.7/site-packages/dbus/bus.py", line 278, in start_service_by_name
'su', (bus_name, flags)))
File "/usr/lib/python2.7/site-packages/dbus/connection.py", line 651, in call_blocking
message, timeout)
dbus.exceptions.DBusException: org.freedesktop.DBus.Error.NoReply: Did not receive a reply. Possible causes include: the remote application did not send a reply, the message bus security policy blocked the reply, the reply timeout expired, or the network connection was broken.

1. http://pkgbuild.com/git/aur-mirror.git/log/backintime

Germar commented on 2013-10-20 19:44

python-keyring changed from 1.6.1 to 3.X and we added python2-secretstorage dependencies as replacement for python-gnomekeyring

orschiro commented on 2013-10-20 19:38

@Germar

The gksu bug is interesting because it worked flawlessly before the package splitting. Now idea what changed since then though.

Germar commented on 2013-10-20 19:25

Regarding backintime-askpass:
this is bad coding by me (shame on me ;) ). backintime-askpass is missing environs ASKPASS_PROFILE_ID and ASKPASS_MODE. But you don't need to start backintime-askpass at all. It is used by ssh-add and encfs and in that time Backintime set those environs by it self. You don't need to report this upstream. I'm aware of now and will take care.

regarding gksu backintime-gnome:
I got a hint about this bug some month ago on Xubuntu and reported that on https://bugs.launchpad.net/ubuntu/+source/python-keyring/+bug/1181344. No response yet. Maybe that wasn't the right place for the bug. I'll try to bump that bug.

Kind regards,
Germar, BIT-Team

PS: I'd appreciate some feedback about new udev schedule.

graysky commented on 2013-10-20 18:33

Nothing is badly broken; stop using AUR helper scripts if you don't understand the basics of makepkg/pacman.

1) makepkg -src
2) pacman -U name.pkg.tar.xz

Where name is one of the following:
backintime = cli only
backintime-gtk = gtk client+cli
backintime-gnome = gnome client+cli (only difference here is more deps)
backintime-qt4 = kde4 client+cli

petaramesh commented on 2013-10-20 16:20

Latest version is badly broken:
- Cannot install using "yaourt" as it generates 4 mutually-incompatible packages (which wasn't previously the case)
- Now impossible to run and backup as root, both the -kde4 version with kdesudo, or gnome version with gksu, bail out in error complaining about a Dbus issue. ...And I badly need to be able to continue performing my system's backups as root...

gatonero commented on 2013-10-20 15:53

Please submit a step by step instruction how to install this package now. I don't get a question for one of the four possible selections but an error:

Fehler: Nicht lösbare Paketkonflikte gefunden
Fehler: Konnte den Vorgang nicht vorbereiten (In Konflikt stehende Abhängigkeiten)
:: backintime-gnome und backintime-gtk stehen miteinander in Konflikt

graysky commented on 2013-10-20 11:24

@farnsworth - You are doing nothing wrong; this is the expected behavior. Once you have built the four, you need to select which of them is right for you.

backintime = cli only
backintime-gtk = gtk client+cli
backintime-gnome = gnome client+cli (only difference here is more deps)
backintime-qt4 = kde4 client+cli

blackshark commented on 2013-10-20 10:24

@orschiro - The same for me.(1.0.28-3)

farnsworth commented on 2013-10-20 10:09

Same problem as orschiro.
Also, the PKGBUILD creates 4 packages,
backintime-1.0.28-3-any.pkg.tar.xz, backintime-gnome-1.0.28-3-any.pkg.tar.xz, backintime-gtk-1.0.28-3-any.pkg.tar.xz, backintime-qt4-1.0.28-3-any.pkg.tar.xz, and fails installing.
What did I do wrong ?

graysky commented on 2013-10-20 10:05

@orschiro - I get the same with 1.0.28-3 or with 1.0.26-6. Please open a ticket upstream.

orschiro commented on 2013-10-20 07:41

Does someone else have a problem with running `gksu backintime-gnome`?

I get the following dbus error:

http://pastebin.com/eS7kTHm5

Aside from Cinnamon I have installed the following packages:

gnome-keyring 3.10.1-1
libgnome-keyring 3.10.1-1
python2-gnomekeyring 2.32.0-14
python2-keyring 3.0.5-1

graysky commented on 2013-10-19 23:16

@blackshark - Well, I just downgraded to 1.0.26-6 and I get the same error when attempting to run backintime-askpass so I am concluding that this is something upstream or I am missing a dependency in the PKGBUILD. If we cannot solve this internally, please open a bug report upstream.

Using 1.0.26:
% backintime-askpass
Traceback (most recent call last):
File "/usr/share/backintime/common/askpass.py", line 43, in <module>
print(pw.get_password(None, profile_id, mode))
NameError: name 'profile_id' is not defined

graysky commented on 2013-10-19 23:10

@blackshark - I get a similar error but admittedly, I have never run this before (I run /usr/bin/backintime-gnome). I am looking into what if anything my modification has done to cause this. I will post back shortly.

I assume that v1.0.26 gave you no issues when running this?

% backintime-askpass
Traceback (most recent call last):
File "/usr/share/backintime/common/askpass.py", line 43, in <module>
print(pw.get_password(None, profile_id, mode))
NameError: name 'profile_id' is not defined

blackshark commented on 2013-10-19 22:51

And I have an error when I launch backintime-askpass :

Traceback (most recent call last):
File "/usr/share/backintime/common/askpass.py", line 43, in <module>
print(pw.get_password(None, profile_id, mode))
NameError: name 'profile_id' is not defined

blackshark commented on 2013-10-19 22:48

I have to run "pacman -U" manually too.

graysky commented on 2013-10-19 19:13

@zozo - Yes, see my previous comment about AUR helper scripts.

noplomplom commented on 2013-10-19 19:06

@graysky: tried to remove backintime first, then reinstall it with the new PKGBUILD, to no avail, same error.
But I tried to install it with packer. That would be the explanation, then?
Anyway, packer builds the packages alright, I just have to "cd" into the build directory and "pacman -U" does the trick.

noplomplom commented on 2013-10-19 19:03

@graysky: tried to remove backintime first, then reinstall it with the new PKGBUILD, to no avail, same error. But of course, "pacman -U" works fine and that's what I actually did.

graysky commented on 2013-10-19 19:01

@orschiro - I don't use helper scripts believing that they are all evil ;p The split PKG is designed to do just this type of work. Making separate PKGBUILDs for the various flavors would be inefficient and a pain to maintain.

orschiro commented on 2013-10-19 18:58

@graysky

Thanks for clarifying. The drawback with this approach is that it cannot be automatised with helpers such as Aura.

What about splitting backintime into separate AUR packages?

graysky commented on 2013-10-19 18:50

@orschiro - When you `makepkg -src` it will build all four packages. You simply need to `pacman -U backintime-foo...` to select the one you want.

orschiro commented on 2013-10-19 18:49

@gravsky

So how can I explicitly install the -gtk version instead of the -gnome one?

I am on Cinnamon 2.0.2 which now no longer depends on Gnome. Thus I would like to install the gtk version only.


graysky commented on 2013-10-19 18:48

@zozo - For me:
% sudo pacman -U backintime-qt4-1.0.28-2-any.pkg.tar.xz
loading packages...
resolving dependencies...
:: There are 2 providers available for phonon-backend:
:: Repository extra
1) phonon-gstreamer 2) phonon-vlc

Enter a number (default=1):
looking for inter-conflicts...
:: backintime-qt4 and backintime are in conflict. Remove backintime? [y/N] y

Packages (69): akonadi-1.10.3-1 aspell-0.60.6.1-1 attica-0.4.2-1 backintime-1.0.28-2 [removal] cdparanoia-10.2-5
convertlit-1.8-6 docbook-xml-4.5-5 docbook-xsl-1.78.1-1 ebook-tools-0.2.2-1 enchant-1.6.0-4 exiv2-0.23-1
ffmpeg-1:2.0.2-2 grantlee-0.3.0-2 gstreamer0.10-base-plugins-0.10.36-1 hspell-1.2-1 ilmbase-2.0.1-1
kactivities-4.11.2-1 kdebase-runtime-4.11.2-1 kdebindings-python-common-4.11.2-1 kdebindings-python2-4.11.2-1
kdelibs-4.11.2-1 kdepim-runtime-4.11.2-1 kdepimlibs-4.11.2-1 libdbusmenu-qt-0.9.2-2 libdmtx-0.7.4-4
libical-1.0-2 libiodbc-3.52.8-2 libkfbapi-1.0-1 libkgapi-2.0.1-1 libkolab-0.4.2-1 libkolabxml-0.8.4-2
libqzeitgeist-0.8.0-3 libutempter-1.1.6-1 libvisual-0.4.0-4 libzip-0.11.1-1 mariadb-5.5.33.a-1
mariadb-clients-5.5.33.a-1 media-player-info-19-1 nepomuk-core-4.11.2-1 openexr-2.0.1-1
oxygen-icons-4.11.2-1 phonon-1:4.6.0-8 phonon-gstreamer-4.6.3-1 polkit-qt-0.103.0-2 poppler-0.24.2-1
poppler-qt4-0.24.2-1 prison-1.0-2 qca-2.0.3-4 qjson-0.8.1-2 qrencode-3.4.3-1 qscintilla-2.7.2-3
raptor-2.0.10-3 rasqal-1:0.9.30-1 redland-1:1.0.16-2 shared-desktop-ontologies-0.11.0-1 soprano-2.9.4-1
strigi-0.7.8-4 udisks2-2.1.1-3 virtuoso-base-1:6.1.7-1 xerces-c-3.1.1-4 xorg-utils-7.6-8
xorg-xdriinfo-1.0.4-3 xorg-xev-1.2.1-1 xorg-xlsatoms-1.1.1-1 xorg-xlsclients-1.1.3-1 xorg-xprop-1.2.2-1
xorg-xvinfo-1.1.2-1 xorg-xwininfo-1.1.3-1 backintime-qt4-1.0.28-2

Total Installed Size: 420.06 MiB
Net Upgrade Size: 419.13 MiB

:: Proceed with installation? [Y/n]

graysky commented on 2013-10-19 18:45

@zozo - Works for me... can you first remove backintime and then try to install the qt4 version?
@coderkun - The -gtk version does not depend on gnome-session whereas the -gnome version does.

coderkun commented on 2013-10-19 18:36

@graysky: What is the difference between backintime-gnome and backintime-gtk?

noplomplom commented on 2013-10-19 18:33

PKGBUILD v1.0.28-2 doesn't work on my KDE: the script creates the four packages (backintime, backintime-gnome, backintime-gtk and backintime-qt4) then cannot install anything because they all conflict with each other.
Here's part of my output:

warning: removing 'backintime' from target list because it conflicts with 'backintime-gnome'
error: unresolvable package conflicts detected
error: failed to prepare transaction (conflicting dependencies)
:: backintime-gnome and backintime-gtk are in conflict

graysky commented on 2013-10-19 18:15

Bump to v1.0.28-2

Changelog: http://backintime.le-web.org/change-log
PKG Changelog: The tricks in the PKBUILD that fukawi created are ingenious but cause the AUR to get confused over the deps and also limit the scope of the package to what is installed on the machine that builds it. I took a stab at a split pkg here that should satisfy all concerned. Please report any problems.

blackshark commented on 2013-10-14 20:24

Ok it's now working ! Thx

Germar commented on 2013-10-14 19:58

@blackshark please run:
'sed -i 1i"PATH=$PATH\nHOME=$HOME\nDISPLAY=$DISPLAY\n\n" /root/.config/backintime/anacrontab; rm /root/.local/share/backintime/anacron/1_Profil_principal'

and watch out if Backintime runs at the next quarter hour. This are modifications I already added for next version.

graysky commented on 2013-10-13 23:45

Not sure what to say about it... suspect something is jacked-up on your system.

blackshark commented on 2013-10-13 21:05

I think the problem is more with cron or anacron in fact... Because if I run the backup manually there is no problem, except an assertion with Python.

$ sudo /usr/bin/backintime --backup-job

Back In Time
Version: 1.0.26

Back In Time comes with ABSOLUTELY NO WARRANTY.
This is free software, and you are welcome to redistribute it
under certain conditions; type `backintime --license' for details.

/usr/lib/python2.7/site-packages/gobject/constants.py:24: Warning: g_boxed_type_register_static: assertion 'g_type_from_name (name) == 0' failed
import gobject._gobject
INFO: Lock
INFO: on process begins
INFO: [GnomePlugin.Systray.run]
INFO: [GnomePlugin.Systray.run] begin loop
INFO: Profile_id: 1
INFO: Call rsync to take the snapshot

graysky commented on 2013-10-13 20:47

It looks sane. Can you manually run that and see anything from the output (omit the >/dev/null redirect to see the output).

Run this as root and post any unusual output: /usr/bin/backintime --backup-job

blackshark commented on 2013-10-13 20:45

sudo cat /root/.config/backintime/anacrontab
1 0 1_Profil_principal nice -n 19 ionice -c2 -n7 /usr/bin/backintime --backup-job >/dev/null 2>&1

graysky commented on 2013-10-13 20:45

Post the output of:
cat /root/.config/backintime/anacrontab

blackshark commented on 2013-10-13 20:45

[[ -x /usr/bin/anacron ]] && echo ok
ok

graysky commented on 2013-10-13 20:43

That looks right... do you have /usr/bin/anacron on your system? Post the output of: [[ -x /usr/bin/anacron ]] && echo ok

blackshark commented on 2013-10-13 20:43

Yep

graysky commented on 2013-10-13 20:43

That looks right... do you have /usr/bin/anacron installed?

blackshark commented on 2013-10-13 20:42

My mistake.

#Back In Time system entry, this will be edited by the gui:
*/15 * * * * test -x /bin/anacron && test -e /root/.config/backintime/anacrontab && /bin/anacron -s -t /root/.config/backintime/anacrontab -S /root/.local/share/backintime/anacron

graysky commented on 2013-10-13 20:40

Is that for your root user or your unprivileged user? Standard notation is to prefix root command with a # and either an $ or % for non-root users.

blackshark commented on 2013-10-13 20:33

Exactly.

$ crontab -l
#Back In Time system entry, this will be edited by the gui:
#Please don't delete these two lines, or all custom backintime entries are going to be deleted next time you call the gui options!

graysky commented on 2013-10-13 20:31

If I understand you correctly, you ran BIT as root and enabled an anacron task but it is not firing. Post the output of:

# crontab -l

blackshark commented on 2013-10-13 18:58

Hi,

I have a problem with the anacron task. It's looks like there is no task schedule. I don't understand because I've enabled a daily task in the GUI (in root).
Did you have an idea ?

graysky commented on 2013-10-10 12:15

Bump to v1.0.26-6

Changelog: Fix from upstream for python2-keyring v3.x.
Commit: http://pkgbuild.com/git/aur-mirror.git/commit/backintime?id=6c9de8c803141c5186b8f5bac4e2aec8ff7d2c5f

fettouhi commented on 2013-10-09 19:22

Seems like a fix has been committed:

https://bugs.launchpad.net/backintime/+bug/1234024

blackshark commented on 2013-10-08 19:47

@graysky: It actually works like a charm ! Thank you so much !

graysky commented on 2013-10-07 20:10

@blackshark - Yes, you have to edit the PKGBUILD and change the pkgver to provide 1.6.1, update the checksum array and then build it.

blackshark commented on 2013-10-07 20:00

Unfortunatly I have removed the old package (1.6.1-1) from my system and python2-keyring=1.6.1-1 isn't avaible on AUR. Any ideas ?

firun commented on 2013-10-06 22:07

@graysky: Thanks, everything working great now :-)

graysky commented on 2013-10-06 17:09

@firun - I updated the PKGBUILD. Plz verify that it works for your KDE system.
@blackshark - Yes, just downgrade until upstream fixes.

blackshark commented on 2013-10-06 15:03

Can I go back after updating python2-keyring (and remove the old package. Yes I know..) ?

firun commented on 2013-10-06 11:56

@graysky: My version of python2-keyring: 1.6.1-1 (not the evil last update ;-)

I suppose the requirement "python-kde4" from the readme means a python2 binding for kde, which is kdebindings-python2 (as there are no python-kde packages in arch). For example in Debian, python-kde4 is a package which depends on python2 (http://packages.debian.org/jessie/python-kde4).

And again, I've tried both, only kdebindings-python2 made backintime work :-)

I think it could be possible that the readme in backintime is outdated, because the version 4.1 mentioned in the readme is quite old, and python3 might not have been an issue at that time...

firun commented on 2013-10-06 11:55

@graysky: My version of python2-keyring: 1.6.1-1 (not the evil last update ;-)

I suppose the requirement "python-kde4" from the readme means a python2 binding for kde, which is kdebindings-python2 (as there are no python-kde packages in arch). For example in Debian, python-kde4 is a package with depends on python2 (http://packages.debian.org/jessie/python-kde4).

And again, I've tried both, only kdebindings-python2 made backintime work :-)

I think it could be possible that the readme in backintime is outdated, because the version 4.1 mentioned in the readme is quite old, and python3 might not have been an issue at that time...

graysky commented on 2013-10-06 10:21

@firun - What version of python2-keyring are you using? Also, according to the README, the only deps from sources:

Common (dependencies: python, rsync, cron)
KDE4 (dependencies: x11-utils, python-kde4 (>= 4.1), kompare)

graysky commented on 2013-10-06 10:19

@firun - What version of python2-keyring are you using?

firun commented on 2013-10-06 10:11

@graysky: Is it possible that backintime with KDE requires kdebindings-python2 instead of kdebindings-python? Because as far as I know, backintime uses python2, and I needed to install kdebindings-python2 in order to avoid this error on startup:

$ backintime-kde4 Traceback (most recent call last):
File "/usr/share/backintime/kde4/app.py", line 44, in <module>
from PyKDE4.kdecore import *
ImportError: No module named kdecore

Thanks a lot for checking :-)

graysky commented on 2013-10-05 19:57

Bump to v1.0.26-4

Changelog: Added a versioned requirement to python2-keyring to keep users from updating to 3.0.5 which FUBR's backintime. A bug report is active upstream concerning this issue.[1]
Commit: http://pkgbuild.com/git/aur-mirror.git/commit/backintime?id=0f8175e5181ead95137030a7d26fe00c5ec6062d

1. https://bugs.launchpad.net/backintime/+bug/1234024

graysky commented on 2013-10-05 14:51

Yes, the update of python2-keyring from 1.6.1-1 -> 3.0.5-1 has fubr'ed backintime. I can repeat this on my gtk setup after updated that package.

There is an open bug report against this upstream.[1]

1. https://bugs.launchpad.net/backintime/+bug/1234024

% backintime-gnome

Back In Time
Version: 1.0.26

Back In Time comes with ABSOLUTELY NO WARRANTY.
This is free software, and you are welcome to redistribute it
under certain conditions; type `backintime-gnome --license' for details.

Traceback (most recent call last):
File "/usr/share/backintime/gnome/app.py", line 1280, in <module>
main_window = MainWindow( cfg, app_instance )
File "/usr/share/backintime/gnome/app.py", line 333, in __init__
mnt = mount.Mount(cfg = self.config, profile_id = profile_id, parent = self.window)
File "/usr/share/backintime/common/mount.py", line 52, in __init__
pw_cache = password.Password_Cache(self.config)
File "/usr/share/backintime/common/password.py", line 265, in __init__
self.keyring_supported = tools.set_keyring(backend)
File "/usr/share/backintime/common/tools.py", line 505, in set_keyring
backends = (keyring.backend.GnomeKeyring(), keyring.backend.KDEKWallet())
AttributeError: 'module' object has no attribute 'GnomeKeyring'

graysky commented on 2013-10-05 14:41

Yes, the update of python2-keyring from 1.6.1-1 -> 3.0.5-1 has fubr'ed backintime. I can repeat this on my gtk setup after updated that package.

% backintime-gnome

Back In Time
Version: 1.0.26

Back In Time comes with ABSOLUTELY NO WARRANTY.
This is free software, and you are welcome to redistribute it
under certain conditions; type `backintime-gnome --license' for details.

which: no gloobus-preview in (/usr/local/sbin:/usr/local/bin:/usr/bin:/usr/bin/vendor_perl:/usr/bin/core_perl:/home/squishy/bin:/home/squishy/bin/browsers:/home/squishy/bin/makepkg:/home/squishy/bin/mounts:/home/squishy/bin/repo:/home/squishy/bin/benchmarking:/home/squishy/bin/chroots:/home/squishy/bin/backup)
Traceback (most recent call last):
File "/usr/share/backintime/gnome/app.py", line 1280, in <module>
main_window = MainWindow( cfg, app_instance )
File "/usr/share/backintime/gnome/app.py", line 333, in __init__
mnt = mount.Mount(cfg = self.config, profile_id = profile_id, parent = self.window)
File "/usr/share/backintime/common/mount.py", line 52, in __init__
pw_cache = password.Password_Cache(self.config)
File "/usr/share/backintime/common/password.py", line 265, in __init__
self.keyring_supported = tools.set_keyring(backend)
File "/usr/share/backintime/common/tools.py", line 505, in set_keyring
backends = (keyring.backend.GnomeKeyring(), keyring.backend.KDEKWallet())
AttributeError: 'module' object has no attribute 'GnomeKeyring'

graysky commented on 2013-10-05 14:40

Yes, the update of gnome-keyring from 1.6.1-1 -> 3.0.5-1 has fubr'ed backintime. I can repeat this on my gtk setup after updated that package.

% backintime-gnome

Back In Time
Version: 1.0.26

Back In Time comes with ABSOLUTELY NO WARRANTY.
This is free software, and you are welcome to redistribute it
under certain conditions; type `backintime-gnome --license' for details.

which: no gloobus-preview in (/usr/local/sbin:/usr/local/bin:/usr/bin:/usr/bin/vendor_perl:/usr/bin/core_perl:/home/squishy/bin:/home/squishy/bin/browsers:/home/squishy/bin/makepkg:/home/squishy/bin/mounts:/home/squishy/bin/repo:/home/squishy/bin/benchmarking:/home/squishy/bin/chroots:/home/squishy/bin/backup)
Traceback (most recent call last):
File "/usr/share/backintime/gnome/app.py", line 1280, in <module>
main_window = MainWindow( cfg, app_instance )
File "/usr/share/backintime/gnome/app.py", line 333, in __init__
mnt = mount.Mount(cfg = self.config, profile_id = profile_id, parent = self.window)
File "/usr/share/backintime/common/mount.py", line 52, in __init__
pw_cache = password.Password_Cache(self.config)
File "/usr/share/backintime/common/password.py", line 265, in __init__
self.keyring_supported = tools.set_keyring(backend)
File "/usr/share/backintime/common/tools.py", line 505, in set_keyring
backends = (keyring.backend.GnomeKeyring(), keyring.backend.KDEKWallet())
AttributeError: 'module' object has no attribute 'GnomeKeyring'

ArchGiac commented on 2013-10-05 13:43

@orschiro
Thanks, but it did not work!
Gnome Keyring is started:
###> ps ax | grep gnome
660 ? Sl 0:00 /usr/bin/gnome-keyring-daemon --start --components=pkcs11
1351 pts/1 S+ 0:00 grep gnome

orschiro commented on 2013-10-05 11:31

@ArchGiac

Could it be that also on KDE you need the Gnome Keyring?

Does that help?

http://www.jfdesignnet.com/?p=2731

ArchGiac commented on 2013-10-05 11:28

After last update, I have this bug:

backintime -b

Back In Time
Version: 1.0.26

Back In Time comes with ABSOLUTELY NO WARRANTY.
This is free software, and you are welcome to redistribute it
under certain conditions; type `backintime --license' for details.

INFO: Lock
Traceback (most recent call last):
File "/usr/share/backintime/common/backintime.py", line 388, in <module>
start_app()
File "/usr/share/backintime/common/backintime.py", line 171, in start_app
take_snapshot( cfg, True )
File "/usr/share/backintime/common/backintime.py", line 53, in take_snapshot
snapshots.Snapshots( cfg ).take_snapshot( force )
File "/usr/share/backintime/common/snapshots.py", line 887, in take_snapshot
hash_id = mount.Mount(cfg = self.config).mount()
File "/usr/share/backintime/common/mount.py", line 52, in __init__
pw_cache = password.Password_Cache(self.config)
File "/usr/share/backintime/common/password.py", line 265, in __init__
self.keyring_supported = tools.set_keyring(backend)
File "/usr/share/backintime/common/tools.py", line 505, in set_keyring
backends = (keyring.backend.GnomeKeyring(), keyring.backend.KDEKWallet())
AttributeError: 'module' object has no attribute 'GnomeKeyring'

Info:
Arch i686
Kernel 3.11.3
Kde 4.11.2

Danyright commented on 2013-10-03 14:05

Thanks for fixing the issue :-)
Works fine for me too (kde4).

graysky commented on 2013-10-02 23:17

Bump to v1.0.26-3

Changelog: Added kde4 fix.
Commit: http://pkgbuild.com/git/aur-mirror.git/commit/backintime?id=bead457945e3ad95bfd6a334f3178375fa1678bd

graysky commented on 2013-10-02 21:19

Bump to v1.0.26-3

Changelog: Added kde4 fix.
Commit:

fettouhi commented on 2013-10-02 16:26

Yes, that patch fixes the kde4 gui.

graysky commented on 2013-10-02 08:51

Can you verify that it fixes your problem? Again, I cannot verify until I get some time to try in a VM.

graysky commented on 2013-10-02 08:49

Can you verify that it fixes your problem? Again, I cannot verify until I get some time to try in a VM.

fettouhi commented on 2013-10-02 08:16

There is a patch for the kde4 gui error
I'm getting here:

https://bugs.launchpad.net/backintime/+bug/1232694

will be fixed in 1.0.28 it seems. Would you mind adding the patch until then?

fettouhi commented on 2013-10-02 06:52

The gnome gui works, so I have that for now.

graysky commented on 2013-10-01 22:16

I actually don't run kde so testing it is impossible... Might spin up a VM and give it a try. Any other kde users out there have this problem?

fettouhi commented on 2013-10-01 22:01

crontab -l is empty

and I don't have a anacrontab file. I only have a config file in ~/.config/backintime/. I run backintime manually when I need it and when I try to start backintime-kde4 I get:

[af@andre ~]$ backintime-kde4

Back In Time
Version: 1.0.26

Back In Time comes with ABSOLUTELY NO WARRANTY.
This is free software, and you are welcome to redistribute it
under certain conditions; type `backintime-kde4 --license' for details.

Fontconfig warning: "/etc/fonts/conf.d/50-user.conf", line 14: reading configurations from ~/.fonts.conf is deprecated. please move it to /home/af/.config/fontconfig/fonts.conf manually
Traceback (most recent call last):
File "/usr/share/backintime/kde4/app.py", line 1201, in <module>
main_window = MainWindow( cfg, app_instance, kapp, kaboutdata )
File "/usr/share/backintime/kde4/app.py", line 259, in __init__
self.list_files_view_model.removeColumns( 3, 2 )
TypeError: KDirModel.removeColumns() is a private method

graysky commented on 2013-10-01 21:26

@fettouhi - Post the output of:

1) crontab -l
2) cat ~/.config/backintime/anacrontab

fettouhi commented on 2013-10-01 20:34

Even with the patch added I'm getting the same error as before.

fukawi2 commented on 2013-10-01 09:23

Patch added in -2

graysky commented on 2013-09-28 14:30

Please include the patch to fix bug #1224620 upstream.[1]
Here is a patched PKGBUILD and the requisite patch.[2] Thanks for maintaining!

1. https://bugs.launchpad.net/backintime/+bug/1224620
2. https://gist.github.com/graysky2/6742633

Germar commented on 2013-09-26 00:38

@Polly you're right. KDirModel.removeColumns() seems to have no affect. Even if I comment out line 273-274 (where unnecessary columns get hidden) removeColumns doesn't change the amount of columns. So I'll append the fix upstream. Thanks

Polly commented on 2013-09-25 09:01

If I understood correctly line 259 in app.py is part of the GUI, in particular controls parts of the list files view. Thus, this is just a graphical problem. I commented the line (as usual with # in front) and the GUI became available. There should be some columns more visible somewhere but since I just started using the application I have no idea.
If this workaround leads to any problem I will post it here. Cheers guys.

agustin commented on 2013-09-11 14:24

Now, after updating kde4 and python2-sip I can confirm the same error reported by @fettouhi

Germar commented on 2013-09-10 22:38

@agustin take a look at https://bugs.launchpad.net/ubuntu/+source/python-kde4/+bug/1218473
It's a bug in python-kde4

fettouhi commented on 2013-09-10 18:04

After updating python2-sip and kde4 to 4.11.1 I get this error running backintime-kde4

[af@andre ~]$ backintime-kde4

Back In Time
Version: 1.0.26

Back In Time comes with ABSOLUTELY NO WARRANTY.
This is free software, and you are welcome to redistribute it
under certain conditions; type `backintime-kde4 --license' for details.

Fontconfig warning: "/etc/fonts/conf.d/50-user.conf", line 14: reading configurations from ~/.fonts.conf is deprecated. please move it to /home/af/.config/fontconfig/fonts.conf manually
Traceback (most recent call last):
File "/usr/share/backintime/kde4/app.py", line 1201, in <module>
main_window = MainWindow( cfg, app_instance, kapp, kaboutdata )
File "/usr/share/backintime/kde4/app.py", line 259, in __init__
self.list_files_view_model.removeColumns( 3, 2 )
TypeError: KDirModel.removeColumns() is a private method

agustin commented on 2013-09-09 22:43

backintime-kde 1.0.26 also crashes under KDE, same error output. Any hint to solve it?

fukawi2 commented on 2013-09-09 01:51

Bump to 1.0.26

agustin commented on 2013-09-09 00:19

A newer versión (1.0.26) is already out

rectec commented on 2013-08-30 01:54

I've made a huge mistake. BIT actually doesn't start without openssh. I should've tested it without having openssh installed. Let's just add openssh to the depends and forget all about this, ok? :)

agustin commented on 2013-08-27 11:25

Same here. This is the KDE Crash handler report http://pastebin.com/Kv9Gdwxj

Anonymous comment on 2013-08-23 19:36

Crashes on KDE.
Back In Time
Version: 1.0.23
Back In Time comes with ABSOLUTELY NO WARRANTY.
This is free software, and you are welcome to redistribute it
under certain conditions; type `backintime-kde4 --license' for details.
KCrash: Application 'app.py' crashing...
KCrash: Attempting to start /usr/lib/kde4/libexec/drkonqi from kdeinit

Germar commented on 2013-08-15 17:04

openssh is indeed only needed for SSH-Mode. But as a workaround of https://bugs.launchpad.net/ubuntu/+source/openssh/+bug/841672
BIT always starts a new ssh-agent at startup. Even if you don't use SSH-Mode. If this bug does not occur on Arch-Linux (didn't test that) you can patch remove ssh-agent in common/backintime and gnome/backintime-gnome so you don't need openssh at all if you don't use SSH-Mode.
Germar, BIT-Team

rectec commented on 2013-08-14 15:46

Remove 'ssh-agent' from depends and add 'openssh' to optdepends; it's only needed for the remote (SSH File Transfer) backup feature. The program runs fine without it. I'll email you the fixed tarball/PKGBUILD. Thanks.

Corubba commented on 2013-08-14 09:52

Confirmed, build and runs fine for me with installed openssh package from [core]. /usr/bin/ssh-agent is owned by that package.

noplomplom commented on 2013-08-14 09:47

Apparently the package "ssh-agent" does not exist in the repos nor in the AUR. I think ssh-agent is part of "openssh".

fukawi2 commented on 2013-08-14 08:16

ssh-agent added as dependency

morodan commented on 2013-08-09 15:02

Was having a "/usr/bin/backintime: line 30: ssh-agent: command not found" problem, fixed with this:
https://answers.launchpad.net/backintime/+question/232771
(solution: install package "ssh-agent")
Could it be set as dependency or set aside from the requirements?

morodan commented on 2013-08-09 13:56

Was having a "/usr/bin/backintime: line 30: ssh-agent: command not found" problem, fixed with this:
https://answers.launchpad.net/backintime/+question/232771
(solution: install package "ssh-agent")
Could it be set as dependency or set aside from the requirements?

zoe commented on 2013-06-19 09:16

Sorry for having flaagged out of date. camarao has the same issue.
To try to reproduce it, have you upgrade your sip files ?
I thought it was because a problem of backintime version, I did not check on backintime-le-web.org.
Sorry, again, and thank you for maintening this package.

Edit : since update today of python2-poppler-qt, backintime does work again.

zoe commented on 2013-06-19 08:42

Sorry for having flaagged out of date. camarao has the same issue.
To try to reproduce it, have you upgrade your sip files ?
I thought it was because a problem of backintime version, I did not check on backintime-le-web.org.
I dropped backintime from my system, I will miss it, for I used it for a long time, and I was used to like it.
Sorry, again, and thank you for maintening this package.

zoe commented on 2013-06-19 08:39

@fukawi2 :
Sorry for having flaagged out of date. camarao has the same issue.
I thought it was because a problem of version, I did not check on backintime-le-web.org.
I dropped it from my system, I will miss it, for I used it for a long time, and I was used to like it.
Sorry, again, and thank you for maintening this package.

fukawi2 commented on 2013-06-18 10:05

@zoe I am unable to reproduce your issue on my system.

fukawi2 commented on 2013-06-18 09:58

There has not been a new release upstream. Please do not flag out of date.

zoe commented on 2013-06-18 09:47

Does not work anymore :
Traceback (most recent call last):
File "/usr/share/backintime/kde4/app.py", line 44, in <module>
from PyKDE4.kdecore import *
RuntimeError: the sip module implements API v10.0 but the PyKDE4.kdecore module requires API v9.2

camarao commented on 2013-06-17 19:10

After upgrading my sip files I got the following messages:
backintime
Traceback (most recent call last):
File "/usr/share/backintime/common/backintime.py", line 25, in <module>
import config
File "/usr/share/backintime/common/config.py", line 32, in <module>
import password
File "/usr/share/backintime/common/password.py", line 25, in <module>
import keyring
File "/usr/lib/python2.7/site-packages/keyring/__init__.py", line 9, in <module>
from core import (set_keyring, get_keyring, set_password, get_password,
File "/usr/lib/python2.7/site-packages/keyring/core.py", line 13, in <module>
from keyring import backend
File "/usr/lib/python2.7/site-packages/keyring/backend.py", line 101, in <module>
from keyring.backends.kwallet import Keyring as KDEKWallet
File "/usr/lib/python2.7/site-packages/keyring/backends/kwallet.py", line 8, in <module>
from PyKDE4.kdeui import KWallet
RuntimeError: the sip module implements API v10.0 but the PyKDE4.kdeui module requires API v9.2

Any workaround this problem?

Before backintime I had backintime-kde4 installed. i uninstalled thi version, made the upgrade e reinlled backintime (without "-kde").

Corubba commented on 2013-06-15 10:54

Build and install fine for me. Check with `pacman -Qo *path*` if any package owns this files and resolv the conflict. If nothings owns it, remove them manually in filesystem.

Profpatsch commented on 2013-06-15 10:31

error: failed to commit transaction (conflicting files)
backintime: /usr/share/backintime/common/dummytools.py exists in filesystem
backintime: /usr/share/backintime/common/mount.py exists in filesystem
backintime: /usr/share/backintime/common/sshtools.py exists in filesystem
Errors occurred, no packages were upgraded.

Got this with the last update and the one before it. Some files are not right.

orschiro commented on 2013-06-09 05:10

@Vaios

You do not need to edit PKGBUILD if it builds and installs fine for you.

PKGBUILD is just meant to be edited if you want to customise the package.

Anonymous comment on 2013-06-08 18:59

When i try to install the package aur/backintime 1.0.24-2 i got this:

backintime 1.0.24-2 (Fri Jan 9 22:46:15 EET 2009)
( Unsupported package: Potentially dangerous ! )
==> Edit PKGBUILD ? [Y/n] ("A" to abort) :

Noob question, but i don't know what was that suppose to mean.
Do i need to edit like this file PKGBUILD?
Do i need to copy the contents of PKGBUILD that is located on the site here to the file that it prompt's me to edit??

altkrall commented on 2013-05-11 00:26

is gnome-python still needed? For me it works without

fukawi2 commented on 2013-05-09 02:14

Ooops, I did miss that. But I did do a test build that didn't fail (obviously, or I wouldn't have upload it :))

Sorry about that, I'll fix now. PKGBUILD'ing before coffee is a bad idea.

esponki commented on 2013-05-09 02:13

Since the package is not being downloaded from trunk the PKGBUILD is broken. Just delete every instance of ~bit-team/$pkgname/trunk/ and everything will be fine.

fukawi2 commented on 2013-05-08 23:21

Thanks @Germar :)
PKGBUILD updated!

Germar commented on 2013-05-08 17:55

We released a new BIT version 1.0.24 today. You'll find a tarball on http://backintime.le-web.org/download_page/ again.

Germar commented on 2013-05-08 17:54

We released a new BIT version 1.0.24 today. You'll find a tarball on http://backintime.le-web.org/download_page/ again.

orschiro commented on 2013-05-05 06:02

Please view the comments below.

It is a well-known problem that the md5sums are constantly changing.

You need to download the tar.gz and then edit the md5sum in the PKGBUILD manually.

gumper commented on 2013-05-04 20:17

When I try to build this the source package does not pass the md5sum check.

ArchGiac commented on 2013-05-04 20:01

@gumper: same problem!

gumper commented on 2013-05-04 19:27

When I try to build this the source package does not pass the md5sum check.

fukawi2 commented on 2013-05-02 22:52

Updated PKGBUILD with corrected logic for building the GUI (or not)

daniel_shub commented on 2013-05-02 13:46

@k0Do You need to check GUI_ENFORCE when you are checking for KDE and Gnome, just like you do when they are not found. Alternatively, you could just wrap all the checks in a

if [[ ${GUI_ENFORCE} == 1 ]] ; then

If you just unset _type, you end up with a bunch of unneeded dependencies. Further, I think python2-gnomekeyring and python2-keyring are only needed if you have a gui, so should only be added to $depends if $_type is not empty.

k0Do commented on 2013-05-02 13:11

@danial_shub I did the GUI_ENFORCE patch cause it won't work else for Xfce, Enlightenment etc. If you unset _type and set GUI_ENFORCE=0, there shouldn't be a GUI at all. Alternatively, you could remove the package checks completely.

k0Do commented on 2013-05-02 13:05

@danial_shub I did the GUI_ENFORCE patch cause it won't work else for Xfce, Enlightenment etc. If you unset _type and set GUI_ENFORCE=0, there shouldn't be a GUI at all. Alternatively, you could remove the package checks completely.

daniel_shub commented on 2013-05-02 10:30

@orschiro While the source code is split into common, KDE, and Gnome branches I don't think it makes much sense to split the package. I think just making the PKGBUILD respect the GUI_ENFORCE option would be enough.

daniel_shub commented on 2013-05-02 10:00

@orschiro While the source code is split into common, KDE, and Gnome branches I don't think it makes much sense to split the package. I think just making the PKGBUILD respect the GUI_ENFORCE option would be enough.

orschiro commented on 2013-05-02 09:30

@daniel_shub

I assume you are right that the keyring is only needed for the GUI/Gnome version. Maybe it does make sense to split the package into a GUI and non-GUI version?

daniel_shub commented on 2013-05-02 09:27

It seems if you set GUI_ENFORCE=0, but have KDE/Gnome installed that you will still get the gui. Further, if GUI_ENFORCE=0 and you don't have KDE/Gnome installed, I am not sure that either python2-gnomekeyring or python2-keyring are dependencies. Can someone confirm this?

fukawi2 commented on 2013-04-25 22:48

@fawn See previous comments about md5sums changing.

fawn commented on 2013-04-25 18:47

New MD5SUM: bb948491fba13c069ea4e7b9d330f3b1

fukawi2 commented on 2013-04-14 23:51

Thx, fixed.

willemw commented on 2013-04-14 06:55

Typo in PKGBUILD: "conflict=" should be "conflicts=".

k0Do commented on 2013-04-13 15:21

update md5 in PKGBUILD with makepkg -g
https://bugs.archlinux.org/task/15051

or use makepkg --skipchecksums (not recommended)

willemw commented on 2013-04-13 14:42

Typo in PKGBUILD: "conflict=" should be "conflicts=".

(Also, package backintime-bzr should provide "backintime" in its PKGBUILD, then this package can drop 'backintime-bzr' from "conflicts=").

joetotale commented on 2013-04-13 14:09

"Validating source files with md5sums..." fails, returning "One or more files did not pass the validity check!" Could someone fix the md5 pls?

joetotale commented on 2013-04-13 14:00

"Validating source files with md5sums..." fails, returning "One or more files did not pass the validity check!" Could someone fix the md5 pls?

fukawi2 commented on 2013-04-11 06:43

Done. md5sum is still borked tho.

orschiro commented on 2013-04-11 04:59

Refering to the error: "Can't connect to gnomekeyring to save password"

This is a bug in backintime and will be fixed with the next version: https://bugs.launchpad.net/backintime/+bug/1162299

Please update the PKGBUILD and add the python keyring dependencies.

http://pastebin.com/j1MRDqFd

orschiro commented on 2013-04-10 18:43

What else is needed to connect to gnome-keyring?

When changing the settings I always get the error:

"Can't connect to gnomekeyring to save password"

I have installed python2-keyring and python2-gnomekeyring.

fukawi2 commented on 2013-04-08 22:52

@k0Do I don't have the inclination to host their source downloads for them if they can't even be bothered releasing proper tarballs. Would you like to adopt the package?

k0Do commented on 2013-04-08 16:50

Proposal: Download the tarball once from Launchpad, host it on a custom space. Then you can go with the same checksum till a new release is available.

k0Do commented on 2013-04-08 16:47

@fukawi2 Do you read the official changelog? IMHO 1.0.22 is the official release.
http://backintime.le-web.org/change-log/
http://bazaar.launchpad.net/~bit-team/backintime/trunk/view/head:/VERSION

k0Do commented on 2013-04-08 16:45

@fukawi2 Do you read the official changelog? IMHO 1.0.22 is the official release.
http://backintime.le-web.org/change-log/

sirkingchase commented on 2013-04-08 02:09

Im not sure how to use "SKIP" so after checksum failed I went into the tmp directory and got the "md5sum" of the file then I edited the pkgbuild with that value.

After the install it was not executing, terminal gave me error-
"ImportError: No module named keyring"

I install "mercurial_keyring" from AUR and now it works. Not sure if this needs to be a dependecy. After reading http://stackoverflow.com/questions/5173197/python-import-seems-to-behave-differently-in-mercurial-keyring-py-file

"Most likely, hg is running using the system python (2.6) rather than the copy of 2.7 you have installed.

Try installing mercurial_keyring and keyring under 2.6, and see if that gets things working as expected."

fukawi2 commented on 2013-04-07 23:15

@test0 See my earlier comment about Launchpad returning different tarballs each download when downloading from the BZR tree tag.

I'm tempted to roll this back to 1.0.8 since the new Launchpad development doesn't seem to have formally made any releases. There's tags in the BZR tree, but no downloads available.

test0 commented on 2013-04-07 12:32

security check fails.

orschiro commented on 2013-04-07 09:57

So, what is the solution to cope with the python2 - python3 conflict?

[orschiro@thinkpad ~]$ backintime
File "/usr/share/backintime/common/backintime.py", line 73
print ''
^
SyntaxError: invalid syntax

cleger commented on 2013-04-07 02:35

The sed invocation in the PKGBUILD needs to be changed to something like:
sed -e 's|^\(ssh-agent \)\?python |\1python2 |g' -i $pkgdir/usr/bin/*

Since 1.0.22, /usr/bin/backintime and /usr/bin/backintime-gnome invoke python through ssh-agent to work around this bug:
https://bugs.launchpad.net/ubuntu/+source/openssh/+bug/841672

cleger commented on 2013-04-07 02:23

The sed expression in the PKGBUILD needs to be changed to something like:
sed -e 's|^\(ssh-agent \)\?python |\1python2 |g' -i $pkgdir/usr/bin/*

Otherwise, /usr/bin/backintime and /usr/bin/backintime-gnome will try to run with "ssh-agent python" instead of "ssh-agent python2". They made that change in 1.0.22 to workaround this bug:
https://bugs.launchpad.net/ubuntu/+source/openssh/+bug/841672

cleger commented on 2013-04-07 02:21

The sed expression in the PKGBUILD needs to be changed to something like:
sed -e 's|^\(ssh-agent \)\?python |\1python2 |g' -i $pkgdir/usr/bin/*

Otherwise, /usr/bin/backintime and /usr/bin/backintime-gnome will try to run with "ssh-agent python" instead of "ssh-agent python2". They made that change to workaround this bug:
https://bugs.launchpad.net/ubuntu/+source/openssh/+bug/841672

cleger commented on 2013-04-07 02:17

The sed expression in the PKGBUILD needs to be changed to something like:
sed -e 's|^\(ssh-agent \)\?python |\1python2 |g' -i $pkgdir/usr/bin/*

Otherwise, /usr/bin/backintime and /usr/bin/backintime-gnome will try to run with "ssh-agent python" instead of "ssh-agent python2". They made that change in those files to workaround [url=https://bugs.launchpad.net/ubuntu/+source/openssh/+bug/841672]this bug[/url].

coderkun commented on 2013-04-06 08:22

@Corubba:
midori-git [AUR] also needs python 2 (for waf) and has the following in the PKGBUILD:
$ msg2 "Python2 fix..."
$ sed -i '0,/on/s//on2/' waf wscript

maybe fukawi2 can add something like that?

kar commented on 2013-04-05 15:54

Ok, thank you.

After I installed "python2-keyring" I can start "backintime-kde4" (as root with kdesu), but if I try to change some settings, there comes a message:

*******************************************
Can't connect to KWallet to save password. On Debian you need to install "apt-get install python-keyring-kwallet".
*******************************************

The terminal output is (as normal user, there are no wrong outputs as root with kdesu):

http://pastebin.com/nzVXRphM




Corubba commented on 2013-04-05 15:26

backintime is using `python` to run, which is a symlink to python3 in arch, but seems to need python2. We need a patch to replace the python calls in the /usr/bin scripts.
The missing keyring module is in the aur/python2-keyring package. This need to be a dep.

kar commented on 2013-04-05 14:53

Ok, with "SKIP" I could install this version of BackInTime, but I can't start it and I don't know why?!

My desktop is KDE Plasma 4.10.2

Here is the terminal output:

http://pastebin.com/AAELeSZn

What's going wrong? Can anybody help?

Thank you.

PS.:
The package "backintime-kde4" (from AUR) works fine, but I would like try the new BackInTime version.



kar commented on 2013-04-05 14:41

Ok, with "SKIP" I could install this version of BackInTime, but I can't start it and I don't know why?!

My desktop is KDE Plasma 4.10.2

Here is the terminal output:

http://pastebin.com/AAELeSZn

What's going wrong? Can anybody help?

Thank you.

PS.:
The package "backintime-kde4" (from AUR) works fine, but I would like try the new BackInTime version.



xpumax commented on 2013-04-05 12:34

@fukawi2:

Since pacman 4.1.0 you can now use SKIP in checksums. That should solve the problem of launchpad generating different tarballs.

See Changelog:
https://projects.archlinux.org/pacman.git/tree/NEWS

fukawi2 commented on 2013-04-04 23:33

Did you delete your download each time?

makepkg -g && rm -f backintime-1.0.22.tar.gz

ValHue commented on 2013-04-04 07:59

Running "makepkg-g" always gives me this value:
md5sums = ('3671e8123fe69dc7d13f81ed9d768b14')

I put it in PKGBUILD and all ok

fukawi2 commented on 2013-04-03 22:22

OK, it appears Launchpad gives a slightly different tarball each time you download, so the md5sums are always going to fail.

Anyone know how to deal with that?

fukawi2 commented on 2013-04-03 22:11

Well it appears upstream have adopted the Ubuntu fork, so here's a new PKGBUILD for 1.0.22 :)

kar commented on 2013-02-12 01:23

Ok, I think it's not really necessary, because version 1.0.8 (I use the KDE4 version) is very stable and for me it runs, runs, runs and runs. ;-)

But I don't know, if there are some new features or if it's only for stability- and security-updates.

fukawi2 commented on 2013-02-11 22:45

It seems Ubuntu have forked BIT since upstream dev seems to have come to a halt. I'm unsure if we should switch to track the Ubuntu version or not...

kar commented on 2013-02-11 15:09

I'm not sure, but I think the package is (maybe) outdated?!

On the website http://backintime.le-web.org/download_page/ is only version 1.0.8 available, but on Launchpad ( https://launchpad.net/~bit-team/+archive/stable ) is for Ubuntu version 1.0.20 available.

fukawi2 commented on 2013-01-01 06:28

@coderkun Because I'm not a TU, and obviously none of the current TU's have an interest in picking it up.

coderkun commented on 2012-12-31 12:05

Why is BackInTime not in the official repos?—all dependencies are.

capoeira commented on 2012-11-04 08:44

anybody got backups working with a systemd timer (without cron)?

fukawi2 commented on 2012-10-28 22:14

Added in pkgrel 5
https://github.com/fukawi2/aur-packages/commit/06212232a9e16e1de323218fa1e5a51508e45293

mrbit commented on 2012-10-28 18:53

please add conflict backintime-kde4

Henry78 commented on 2012-10-20 11:23

@graysky: I should also have read the older comments, as the python move was already mentioned. Sorry.
@fukawi2: Please bump the pkgrel

fukawi2 commented on 2012-10-20 11:09

What is Manjaro?
python-notify is no longer in in the repositories, so there is no point having a dependency for it, optional or otherwise.

k0Do commented on 2012-10-20 07:52

Manjaro still uses python-notify, so please keep also the other dependency (optdepends maybe).

graysky commented on 2012-10-20 00:06

@Henry78 - Dude... Updated your package and read fukawi2's comment about the flag out-of-date abuse... although in your defense, fukawi2 should have bumped the pkgrel to 5 to avoid confusion.

Henry78 commented on 2012-10-19 23:41

:: backintime: requires python-notify
which was recently replaced by extra/python2-notify

Henry78 commented on 2012-10-19 23:41

:: backintime: requires python-notify
which was recently replaced by extra/python2-notify?

fukawi2 commented on 2012-10-18 00:57

I understand now; my mirror has not got the pacman db with that change. I will update now.

Anonymous comment on 2012-10-18 00:30

I think the issue xpumax was bringing up is that python-notify has been removed from the official repositories and replaced with python2-notify. The backintime package is current in that 1.0.8 is the latest version, but the dependencies now include a package that is no longer an officially supported Arch package. This is a problem for someone who always updates to the most current packages as the official package list now tells pacman to replace python-notify with python2-notify, but it can't because of the dependency in this package.

fukawi2 commented on 2012-10-17 23:10

@xpumax Please do not flag packages out of date because of PKGBUILD errors; they are errors, not out of date. My testing shows that it works fine with python-notify rather than python2-notify. Can you provide more information what you've found that doesn't work?

graysky commented on 2012-10-17 14:52

...why is this out-of-date? 1.0.8 is current.

xpumax commented on 2012-10-17 13:13

Dependencies need to be updated:
python-notify -> python2-notify

capoeira commented on 2012-10-08 15:05

yea, one doesn't seam to be able to configure the backups at CLI. but it seams, once you configured the profile in the GUI one can reinstall ist without gui and use the profile created before to backup with a cron

k0Do commented on 2012-10-08 10:03

For the CLI options, try "man backintime" or "backintime --help".

k0Do commented on 2012-10-08 09:39

For the CLI options, try "man backintime" or "backintime --help".

capoeira commented on 2012-10-04 16:56

CLI version has all the options the GUI has?

k0Do commented on 2012-10-04 07:43

You can modify PKGBUILD to build only the cli version. Therefore you won't have any GUI and specific dependencies. Maybe you need just the cron entry.

fukawi2 commented on 2012-10-04 01:27

No problem; and that's why it defaults to Gnome instead of KDE if you're running E17 or whatever... Gnome tends to pull less deps than KDE.

capoeira commented on 2012-10-04 01:27

thanks, seams like kdebindings would install even more packages.
I wont try it.

Thanks for the AUR, this is the only backup software that suites my needs

fukawi2 commented on 2012-10-04 01:18

Because the software is only written with KDE (qt) and Gnome (gtk) bindings; hence you have to have one and/or the other for things to work since the author hasn't written an E17, Xfce etc only version.

You can mod the PKGBUILD to force the KDE version if you prefer KDE/QT bindings.

capoeira commented on 2012-10-04 01:11

just tryig to understand a little of this all. If I use E17, why do I need gnome-python?

fukawi2 commented on 2012-08-29 00:34

Updated:
https://github.com/fukawi2/aur-packages/commit/ea3eff8cc32724bd295ffd21ca8c6467cf56c80e

k0Do commented on 2012-08-28 14:47

I guess gnome-session is only needed for GnomePlugin which can give you an optional tray icon while a snapshot is working.
So, gnome-session should be a suggestion, no hard dependency.

k0Do commented on 2012-08-28 14:38

Request for out of date: There's version 1.0.10 tagged in bzr, though no offical tarball available.

The configure script for the gnome GUI does not work for --no-check, meaning gnome-session is always enforced. Not good for Xfce, Enlightenment etc.

Here my proposal for the PKGBUILd part:

# Build a custom array of depends based on what DE is installed
GUI_ENFORCE=1 # set to 0 if you don't want a GUI at all
if [ -e '/usr/lib/libkonq.so' ]; then
# KDE is installed
depends=( ${depends[@]} xorg-utils python2-pyqt kdebindings-python)
_type=( ${_type[@]} kde4 )
fi
if [ -n "`which gnome-session 2> /dev/null`" ]; then
# GNOME really
depends=( ${depends[@]} pygtk python-notify gnome-python meld gksu gnome-session )
_type=( ${_type[@]} gnome )
fi
if [ ${GUI_ENFORCE} == 1 -a ${#_type[@]} == 0 ]; then
# other DE like Xfce, Enlightenment etc.
depends=( ${depends[@]} pygtk python-notify gnome-python meld gksu )
_type=( ${_type[@]} gnome )
fi

Replace the line "./configure" with "cp Makefile.template Makefile" as a work around for the gnome section.
The check for gnome-session is useless here.
http://bazaar.launchpad.net/~bit-team/backintime/trunk/view/head:/gnome/configure

capoeira commented on 2012-07-14 14:45

is it neccessary that gvfs runs proccesses the whole time? I'm not using gnome

fukawi2 commented on 2011-12-28 02:02

Very weird... Makes me think even more it's an issue further down the dep chain (ie, somewhere in python) rather than backintime itself. Still keen to fix in the PKGBUILD if possible though :)

graysky commented on 2011-12-28 01:47

Odd thing is that this is only happening on one machine. I have your PKG running just fine on 3 other x86_64 boxes... of course the machine that's causing a problem resides at my parent's house. Gotta coordinate a time with them to ssh in... they power down EVERYTHING including the modem and router when the machine is not in use :p

fukawi2 commented on 2011-12-27 22:44

Hmm, same... Appears to have last been updated 6+ months ago:

$ grep lib32-gnutls /var/log/pacman.log
[2010-04-27 20:04] installed lib32-gnutls (2.8.6-1)
[2010-09-05 09:40] upgraded lib32-gnutls (2.8.6-1 -> 2.8.6-3)
[2010-10-10 17:40] upgraded lib32-gnutls (2.8.6-3 -> 2.8.6-4)
[2011-01-18 18:16] upgraded lib32-gnutls (2.8.6-4 -> 2.10.4-1)
[2011-05-08 11:45] upgraded lib32-gnutls (2.10.4-1 -> 2.12.2-1)

I wonder if it's a python version issue; I see you're running it with 2.7.. There is python-gnutls in the repos, and it depends on python26. Could you try downgrading to 2.6 and see if it works?

graysky commented on 2011-12-27 22:26

@fukaw - wish I knew.

# pacman -S lib32-gnutls
error: target not found: lib32-gnutls

I have multilib enabled. Really odd!

:: Synchronizing package databases...
core is up to date
extra is up to date
community is up to date
multilib is up to date

fukawi2 commented on 2011-12-27 22:14

Hmmm, well I obviously haven't got it as a dep, it just already exists on my system. What do you think is the best solution here?

graysky commented on 2011-12-27 16:23

That's even more fucked up... it's x86_64.

fukawi2 commented on 2011-12-27 10:53

@graysky;
32 or 64 bit system?

$ pacman -Qo /usr/lib32/libgnutls.so.26
/usr/lib32/libgnutls.so.26 is owned by lib32-gnutls 2.12.2-1

graysky commented on 2011-12-26 12:36

$ gksu backintime-gnome
Traceback (most recent call last):
File "/usr/share/backintime/gnome/app.py", line 33, in <module>
import gnome
File "/usr/lib/python2.7/site-packages/gtk-2.0/gnome/__init__.py", line 13, in <module>
from _gnome import *
ImportError: libgnutls.so.26: cannot open shared object file: No such file or directory

fukawi2 commented on 2011-12-08 08:20

Thanks graysky; patch applied to 1.0.8-3 :)

graysky commented on 2011-12-07 23:24

@fukawi2 - here is a patch that'll fix the annoying incorrect version for backintime. Applies cleanly to 1.0.8-2

http://pastebin.com/J6Z1anGV

graysky commented on 2011-07-29 00:12

Thanks leonardof - looks like it's in [extra now] and everything works for me.

leonardof commented on 2011-07-24 01:04

As commented on the bug report by graysky, the issue is in gnome-vfs, which doesn't support libcrypt 1.5. There is a bug report on this [https://bugs.archlinux.org/task/25225] and, as a workaround, a patched gnome-vfs in AUR [http://aur.archlinux.org/packages.php?ID=50918].

altkrall commented on 2011-07-22 22:20

and yes, I use gnome.

altkrall commented on 2011-07-22 22:09

ah, thanks.

fukawi2 commented on 2011-07-21 22:55

@graysky I'm running xfce now, but previously Gnome. There is perhaps a missing dependency that I'm not aware of. If you find it, let me know and I'll add it.

fukawi2 commented on 2011-07-21 22:54

@altkrall: I'm running xfce now, but previously Gnome. There is perhaps a missing dependency that I'm not aware of. If you find it, let me know and I'll add it.

graysky commented on 2011-07-21 20:30

@fukawi2 - Are you running gnome as your DE? I tried this under several users and I get the same error in each case. I just tried it on another machine running just lxde and it ends in the same error.

Opened a bug report with upstream since it seems to be more than just my single machine: https://bugs.launchpad.net/backintime/+bug/814271

graysky commented on 2011-07-21 20:10

@fukawi2 - Are you running gnome as your DE? I tried this under several users and I get the same error in each case :(

altkrall commented on 2011-07-21 18:51

I also get the error of graysky. How can I repair this?

fukawi2 commented on 2011-07-21 10:31

Nope, WFM:

$ gksu backintime-gnomeINFO: Update to config version 4: other snapshot locations
INFO: Random tag for profile 1: 506
INFO: Update to config version 5: other snapshot locations

Back In Time
Version: 1.0.7

Back In Time comes with ABSOLUTELY NO WARRANTY.
This is free software, and you are welcome to redistribute it
under certain conditions; type `backintime-gnome --license' for details.

which: no gloobus-preview in (/usr/lib/cw:/usr/local/bin:/usr/bin:/bin:/usr/local/sbin:/usr/sbin:/sbin:/opt/java/bin:/opt/java/db/bin:/opt/java/jre/bin:/usr/bin/site_perl:/usr/bin/vendor_perl:/usr/lib/perl5/vendor_perl/bin:/usr/bin/core_perl:~/bin/:/usr/local/bin:/usr/local/sbin:/usr/local/usr/bin/)

graysky commented on 2011-07-21 08:31

fukawi2 - seems broken for me, can you confirm:

$ gksu backintime-gnome
Traceback (most recent call last):
File "/usr/share/backintime/gnome/app.py", line 33, in <module>
import gnome
File "/usr/lib/python2.7/site-packages/gtk-2.0/gnome/__init__.py", line 13, in <module>
from _gnome import *
ImportError: /usr/lib/libgnomevfs-2.so.0: undefined symbol: gcry_control

fukawi2 commented on 2011-07-11 03:39

Whoops, you are correct. My apologies, and fixed in pkgrel 2:
https://github.com/fukawi2/aur-packages/commit/a6be1e3d086a019a260de37d796bd315360489f2

leonardof commented on 2011-07-11 03:08

@fukawi2: Indeed, having the dep set tp 'cron' would allow us to use any cron package, but it is set to 'dcron'. Would you please set it to 'cron' instead? Thanks!

fukawi2 commented on 2011-06-26 23:12

Bump to 1.0.8
* Fix bug: https://bugs.launchpad.net/backintime/+bug/723545
* Fix bug: https://bugs.launchpad.net/backintime/+bug/705237
* Fix bug: https://bugs.launchpad.net/backintime/+bug/696663
* Fix bug: https://bugs.launchpad.net/backintime/+bug/671946

fukawi2 commented on 2011-05-16 22:59

'cronie' provides 'cron' so having the dep set to 'cron' allows you to use any cron packages (assuming the packager has correctly set the 'provides' array to include 'cron')

jaclar commented on 2011-05-16 14:13

Could you change the dependencies to:
depends=( 'python2' 'rsync' 'cron' )
because cronie is now in core and i wouldn't like to install dcron. Or is there any reason why backintime won't work with cronie?

fukawi2 commented on 2011-02-11 11:40

Indeed it appears it should... Fixed:
https://github.com/fukawi2/aur-packages/commit/003076fd72a28bacb9676f78fa75d591e9fa8f66

Linas commented on 2011-02-07 16:05

This should be an 'any' package.

fukawi2 commented on 2011-02-01 21:49

Probably not -- this should build the GTK and/or the QT version depending on which you have installed.

schivmeister commented on 2011-02-01 17:05

There is no need for the -kde4 package right?

fukawi2 commented on 2011-01-23 03:21

Thx, fixed :)

ptb commented on 2011-01-21 08:12

In the depends field pyqt should be replaced by python2-qt.

fukawi2 commented on 2011-01-02 22:41

Update to 1.0.6

fukawi2 commented on 2010-12-19 12:51

Gnome should be the default if KDE isn't found (ie, for Gnome and "other" users)

graysky commented on 2010-12-19 12:34

@fukawi2 - Can you modify the PKGBUILD to work with other DEs like lxde? I can verify that the gnome dependency set will build a functional package for lxde users.

fukawi2 commented on 2010-10-28 23:52

Bump to version 1.0.4

fukawi2 commented on 2010-10-27 22:07

Right -- it hasn't been packaged and released on the main website, so since this isn't a -bzr package, it can't be updated until that happens.

mr_hangman commented on 2010-10-27 22:05

The developer has released 1.0.4 on launchpad. https://code.launchpad.net/backintime
With a critical bug fixed, I thought it should be updated here. Sorry if I misunderstood it.

mr_hangman commented on 2010-10-27 21:59

The developer has released 1.0.4 on launchpad. https://code.launchpad.net/backintime
With a critical bug fixed, I thought it should be updated here. Sorry if I misunderstood it.

fukawi2 commented on 2010-10-27 21:43

Again, why is this marked out of date?? There is NO NEW VERSION on the developers website. Please do NOT mark out of date if it isn't.

fukawi2 commented on 2010-10-24 22:45

Fixed a bash syntax error that would have broken Gnome and *box installations. Please rebuild using the new PKGBUILD if you DO NOT use KDE.

fukawi2 commented on 2010-10-20 21:58

Why is this out of date? There's no new release on the developers website?

PKGBUILD updated to patch the /usr/bin/ scripts to use python2

domanov commented on 2010-10-20 14:51

Turns out, python2 needs to be called as interpreter in the bash scripts. The workaround is to simply change:

from:
python ${APP_PATH}/app.py "$@"

to:
python2 ${APP_PATH}/app.py "$@"

in /usr/bin/backintime, -gnome and -kde4. Works flawlessy :)
Of course python2 has to be installed, but I think it is the default for arch. PKGBUILD (dependencies and so on) needs to be changed.

graysky commented on 2010-10-19 20:20

Yep, same here

$ backintime-gnome
File "/usr/share/backintime/gnome/app.py", line 308
if self.snapshots.has_old_snapshots():
^
TabError: inconsistent use of tabs and spaces in indentation

domanov commented on 2010-10-19 11:43

After rebuild:

$ backintime-gnome
File "/usr/share/backintime/gnome/app.py", line 308
if self.snapshots.has_old_snapshots():
^
TabError: inconsistent use of tabs and spaces in indentation

Anonymous comment on 2010-10-19 09:26

[lorenzo@nemesis ~]$ backintime
File "/usr/share/backintime/common/backintime.py", line 47
print ''
^
SyntaxError: invalid syntax

Guess python3 related...

graysky commented on 2010-10-17 10:39

Looks nice, fukawi2. The "|| return 1" statements are no longer needed with the current makepkg.

fukawi2 commented on 2010-10-17 05:14

Bump to 1.0.2

fukawi2 commented on 2010-10-15 05:59

kdebase-libs creates /usr/lib/libkonq.so so I've changed the PKGBUILD to look for this file.

Anonymous comment on 2010-10-14 21:09

To detect KDE, what about using a pacman query in your if statement? Here's a pastebin diff of your pkgbuild that detects if kdebase-lib is installed. Kdebase-lib is only required by a handful of KDE applications (dolphin, kdepasswd, kfind, plasma, kget, and ark). Odds are that if those are installed so is the rest of KDE.

http://aur.pastebin.com/z1vDyczV

fukawi2 commented on 2010-09-15 22:41

k3b depends on kdelibs, which provides kdeinit4. Unless you have a better way to determine if KDE is installed as a DE, then there's nothing I can do.

graysky commented on 2010-09-15 21:12

@fukawi2 - it's k3b!

fukawi2 commented on 2010-09-14 23:30

If kdeinit is still present, then KDE wasn't removed completely. I'm not aware of any better way to determine if kde is installed.

You should probably remove whatever package it is that owns kdeinit4:
pacman -Qo /usr/bin/kdeinit4
pacman -Rns <output of above cmd>

graysky commented on 2010-09-14 20:16

Thanks... I had to remove the first bit of the deps line (kde detector). I installed KDE on this box a while back, but have since removed it (back to Gnome), but the PKGBUILD wanted to pull down the KDE deps.

$ makepkg -s
==> Making package: backintime 1.0-1 (Tue Sep 14 16:09:54 EDT 2010)
==> Checking Runtime Dependencies...
==> Installing missing dependencies...
resolving dependencies...
looking for inter-conflicts...

Targets (10): sip-4.10.5-1 pyqt-4.7.4-2 boost-libs-1.43.0-1 mysql-clients-5.1.50-1 mysql-5.1.50-1
akonadi-1.4.0-1 kdepimlibs-4.5.1-1 kdepim-runtime-4.4.6-1 qscintilla-2.4.4-2
kdebindings-python-4.5.1-1

Total Download Size: 21.10 MB
Total Installed Size: 161.64 MB

Proceed with installation? [Y/n] n

---

Here is the output of the detect commands:

$ which kdeinit4
/usr/bin/kdeinit4

$ which gnome-session
/usr/bin/gnome-session

I should say that I only have k3b installed now so perhaps that's the problem?

graysky commented on 2010-09-14 20:12

Thanks... I had to remove the first bit of the deps line (kde detector). I installed KDE on this box a while back, but have since removed it (back to Gnome), but the PKGBUILD wanted to pull down the KDE deps.

$ makepkg -s
==> Making package: backintime 1.0-1 (Tue Sep 14 16:09:54 EDT 2010)
==> Checking Runtime Dependencies...
==> Installing missing dependencies...
resolving dependencies...
looking for inter-conflicts...

Targets (10): sip-4.10.5-1 pyqt-4.7.4-2 boost-libs-1.43.0-1 mysql-clients-5.1.50-1 mysql-5.1.50-1
akonadi-1.4.0-1 kdepimlibs-4.5.1-1 kdepim-runtime-4.4.6-1 qscintilla-2.4.4-2
kdebindings-python-4.5.1-1

Total Download Size: 21.10 MB
Total Installed Size: 161.64 MB

Proceed with installation? [Y/n] n

---

Here is the output of the detect commands:

$ which kdeinit4
/usr/bin/kdeinit4

$ which gnome-session
/usr/bin/gnome-session

graysky commented on 2010-09-14 20:12

Thanks... I had to remove the first bit of the deps line (kde detector). I installed KDE on this box a while back, but have since removed it (back to Gnome), but the PKGBUILD wanted to pull down the KDE deps.

$ makepkg -s
==> Making package: backintime 1.0-1 (Tue Sep 14 16:09:54 EDT 2010)
==> Checking Runtime Dependencies...
==> Installing missing dependencies...
resolving dependencies...
looking for inter-conflicts...

Targets (10): sip-4.10.5-1 pyqt-4.7.4-2 boost-libs-1.43.0-1 mysql-clients-5.1.50-1 mysql-5.1.50-1
akonadi-1.4.0-1 kdepimlibs-4.5.1-1 kdepim-runtime-4.4.6-1 qscintilla-2.4.4-2
kdebindings-python-4.5.1-1

Total Download Size: 21.10 MB
Total Installed Size: 161.64 MB

Proceed with installation? [Y/n] n

fukawi2 commented on 2010-09-13 22:51

Bump to v1.0

The changes are as follows:
- Multiple profiles support.
- Expert options: copy unsafe links and copy links.
- Custom backup hour (for daily backups or mode).
- Many bugs have been fixed.
- ionice support for the user/cron backup process.
- The backintime snapshot folder has been restructured to ../backintime/machine/user/profile_id/

graysky commented on 2010-09-02 22:54

s'ok, thanks for trying. I asked on the home page:

https://answers.launchpad.net/backintime/+question/123873

graysky commented on 2010-09-02 22:49

s'ok, thanks for trying.

fukawi2 commented on 2010-09-02 22:41

That I don't know how to fix -- I'm guessing a missing Gnome dep, but I don't know what :-/

graysky commented on 2010-09-02 21:23

@fukawi2 - it builds just fine now under lxde but it will not run. I get the following when I try to do it with or without gksu:

$ gksu backintime
** Message: secret service operation failed: The name
org.freedesktop.secrets was not provided by any .service files
Back In Time
Version: 0.9.26

Back In Time comes with ABSOLUTELY NO WARRANTY.
This is free software, and you are welcome to redistribute it
under certain conditions; type `backintime --license' for details.

# backintime
Back In Time
Version: 0.9.26

Back In Time comes with ABSOLUTELY NO WARRANTY.
This is free software, and you are welcome to redistribute it
under certain conditions; type `backintime --license' for details.

graysky commented on 2010-09-02 21:20

@fukawi2 - it builds just fine now under lxde but it will not run. I get the following when I try to do it

$ gksu backintime
** Message: secret service operation failed: The name
org.freedesktop.secrets was not provided by any .service files
Back In Time
Version: 0.9.26

Back In Time comes with ABSOLUTELY NO WARRANTY.
This is free software, and you are welcome to redistribute it
under certain conditions; type `backintime --license' for details.

fukawi2 commented on 2010-08-30 22:54

Wow, I didn't know so many people used the KDE *and* Gnome UI's at the same time. Sorry folks.

Another new version that restores the old behaviour *and* should default to Gnome for those who don't use either DE.

adrianx commented on 2010-08-30 20:25

I use both KDE and Gnome, and backintime-gnome was missing because of it. This is what I did (perhaps not very elegant):

http://aur.pastebin.com/hby9yT5b

adrianx commented on 2010-08-30 20:16

I run both KDE and Gnome, and backintime-gnome was missing because of it. This is what I did (perhaps not very elegant):

## Build a custom array of depends based on what DE is installed
if [ `which kdeinit4 2> /dev/null` ]; then
# KDE
depends=( ${depends[@]} 'xorg-utils' 'pyqt' 'kdebindings-python')
_type=( ${_type[@]} kde4 )
fi

if [ `which gnome-session 2> /dev/null` ]; then
# Gnome
depends=( ${depends[@]} 'pygtk' 'python-notify' 'gnome-python' 'meld' )
_type=( ${_type[@]} gnome )
else
# Default to Gnome (less deps for users without KDE or Gnome)
depends=( ${depends[@]} 'pygtk' 'python-notify' 'gnome-python' 'meld' )
_type=( ${_type[@]} gnome )
fi

adrianx commented on 2010-08-30 08:49

backintime-gnome is now completely missing. (I use both KDE and Gnome)

fukawi2 commented on 2010-08-30 01:35

I've updated the PKGBUILD to default to Gnome, so LXDE/XFCE/not-gnome-or-kde users should work now.

Please test and let me know if it still doesn't work.

graysky commented on 2010-08-30 00:50

Not very descriptive error I'm afraid...

$ makepkg -s

My box only has lxde on it so no kde/gnome stuff at all.

graysky commented on 2010-08-30 00:49

Not very descriptive error I'm afraid...

$ makepkg -s

==> ERROR: An unknown error has occurred. Exiting...

fukawi2 commented on 2010-08-06 08:32

Do you have 'gnome-session' or 'kdeinit4' binaries? What errors do you get?

graysky commented on 2010-08-06 08:03

+1 can't build with LXDE

Anonymous comment on 2010-06-16 20:05

Out of GNOME/KDE ambient, makepkg doesn't work.
Editing makepkg and leaving just GTK (GNOME) dependencies, it bulds, and run, but no interface works.

I'm on LXDE.

Aeon_Tux commented on 2010-05-07 15:50

If large Files (>3 GB) are backed up starting the Gnome GUI of backintime might result in an error.

This is a known bug: https://bugs.launchpad.net/backintime/+bug/409130 which can easily be fixed by changing "int" to "float" in /usr/share/backintime/gnome/app.py Line 212


211 # display name, relative path, icon_name, type (0 - directory, 1 - file), size (str), date, size (int)
212 self.store_folder_view = gtk.ListStore( str, str, str, int, str, str, float )

fukawi2 commented on 2010-04-27 03:54

That will happen if you have both Gnome and KDE on your system, as it installs a .desktop file for each GUI (gtk and qt). You can force it to only build a GUI for your preferred environment (and therefore only install that .desktop file) by removing one of the IF blocks above build(). Remove the appropriate kdeinit4 or gnome-session block.

joelsc commented on 2010-04-25 15:41

Minor problem: each backintime icon shows up twice in the gnome applications menu. As a result, I have 4 icons in Applications>System tools :

-2 Back in time icons
-2 Back in time (root) icons