Package Details: logmein-hamachi

Git Clone URL: (read-only)
Package Base: logmein-hamachi
Description: A zero-configuration VPN service
Upstream URL:
Licenses: custom
Conflicts: hamachi
Replaces: hamachi
Submitter: Xavion
Maintainer: Xavion (RazZziel)
Last Packager: Xavion
Votes: 218
Popularity: 1.894057
First Submitted: 2010-07-10 01:04
Last Updated: 2017-09-18 03:50

Latest Comments

Xavion commented on 2017-09-18 03:51

Thanks for pointing that out. I've had a go at making the PKGBUILD work for everyone again.

kanachan commented on 2017-09-17 11:30

I had some issues compiling it for RaspberryPi (armv6h), here is the fix I found:

logmein-hamachi source arm works well on armv6h:

[ "${CARCH}" == "armv6h" ] && _filearch_=armel && sha1sums=('328ea87a2f371163a90b9d7518417e5acbcb2da5' 'c0d5b67ba0a938cef9742568fd09bb2a4b85b513')

but glibc-2.25-7 depends of the arquitecture, so here is the source I used (Its the only archlinuxarm archive I found, if there is an official, please correct me):


Also added this line to package() because is a reference to a file inside the package which caused troubles at packaging.

rm "${srcdir}"/usr/lib/

Here is the full pkgbuild:

I know a better solution would be to make a pkgbuild that get the source according to the arquitecture but my knowledge doesnt go that far..

ztefn commented on 2017-09-15 18:26

Glibc 2.26 is currently in the Ubuntu 17.10 proposed repository:

I guess LogMeIn won't be able to play the unsupported distro card for too long on this matter.

RazZziel commented on 2017-09-13 10:00

:) great, thanks!

Xavion commented on 2017-09-12 23:36

Yes, we can't keep everyone stranded at 'glibc' v2.25 forever. Your latest build files work well here, so I have set them as the default.

Furthermore, I have added you as a co-maintainer of this package for future updates. Thanks again for your assistance. The Arch community is lucky to have people like you.

RazZziel commented on 2017-09-12 17:04

This is the official position of LogMeIn about this issue:

> Hamachi currently is only supported in these two Linux distros (Ubuntu and CentOS) per our system requirements page (

RazZziel commented on 2017-09-12 13:34

Oops, last diff was wrong, missed some libraries:

RazZziel commented on 2017-09-12 12:46

In case it's acceptable, I took the liberty to patch the PKGBUILD:

Works For Me(TM) :)

RazZziel commented on 2017-09-12 11:19

Meanwhile, would it be acceptable to patch it inside the PKGBUILD, adding the trick?

I've notified LogMeIn about the issue, but I haven't received a reply yet, and I suspect it may take long

ztefn commented on 2017-09-12 10:11

This PKGBUILD won't work if you (already) have glibc version 2.26 installed:
Failed to prepare transaction:
could not satisfy dependencies:
unable to satisfy dependency 'glibc=2.25' required by logmein-hamachi

In this case you have to downgrade binutils to version 2.28 and then glibc to version 2.25 and probably keep them there until LogMeIn release a new Hamachi version that is compatible with glibc 2.26.

Xavion commented on 2017-09-12 07:20

Yes, you're right about that. Thanks for the heads-up. I've modified the PKGBUILD accordingly.

RazZziel commented on 2017-09-12 06:05

Worked around the issue for now by downloading glibc-2.25 and running hamachid with it:

mkdir glibc-2.25-7-x86_64
cd glibc-2.25-7-x86_64
tar -xf ../glibc-2.25-7-x86_64.pkg.tar.xz
sudo LD_LIBRARY_PATH=$PWD/usr/lib/ usr/lib/ /opt/logmein-hamachi/bin/hamachid debug

With that the hamachid daemon runs fine, and then I can run the hamachi client normally (with glibc 2.26)

RazZziel commented on 2017-09-12 04:48

This package is broken since the upgrade from glibc 2.25-7 to 2.26-3:

06:46:14 ~$ sudo /opt/logmein-hamachi/bin/hamachid debug
09.12 06:46:53.934 cfg: load()
09.12 06:46:53.934 dbg: deploy_id '' parse failed
09.12 06:46:53.934 cfg: hkc set: [2048] [2048]
09.12 06:46:53.934 cfg: save()
09.12 06:46:54.477 ses: resolving ..
hamachid: relocation error: hamachid: symbol __res_maybe_init, version GLIBC_PRIVATE not defined in file with link time reference

RazZziel commented on 2017-01-16 08:25

Hamachi is available at

Zekario commented on 2016-01-18 14:08

I installed this, and when I run `hamachi`, it says `Run '/etc/init.d/logmein-hamachi start' to start daemon.` but /etc/init.d doesn't exist.

EDIT: Just run `systemctl enable logmein-hamachi.service`

Xavion commented on 2015-09-07 07:35

Sorry, I'm not having that problem here. LogMeIn-Hamachi starts on boot without prompting for a password. I last upgraded my system packages two days ago.

PhotonX commented on 2015-09-07 06:44

Following problem: Reinstalling didn't help.

Xavion commented on 2014-08-16 06:29

Thanks for letting me know about that problem. The 'systemd' file was missing because the ultra-tight "mkaurball" script - that packagers are now forced to use - omits anything not listed in the "source=()" array. I've fixed the problem and uploaded the resultant PKGBUILD.

Zeben commented on 2014-08-15 23:13

I have a troubles with building Hamachi with this new PKGBUILD:
What can I do to solve this?

Xavion commented on 2014-06-01 00:12

You'll need to create the "/var/lib/logmein-hamachi/h2-engine-override.cfg" file manually. Then, restart the LMI Hamachi daemon via "sudo systemctl restart logmein-hamachi.service".

inukaze commented on 2014-05-31 21:23

How i can fix the follow ? :

$ hamachi
You do not have permission to control the hamachid daemon.
You can run 'hamachi' as root, or you can add your login name to the file
and restart the daemon with
sudo /etc/init.d/logmein-hamachi restart
Ipc.User <login name>

because the files
/etc/init.d/logmein-hamachi restart

dont exist in ArchLinux x86_64

coopstah13 commented on 2014-05-05 20:47

The source link is broken, please fix

Xavion commented on 2014-04-25 03:56

Yep, not a problem; I've updated the PKGBUILD.

alevalv commented on 2014-04-25 02:15

Hey, can you update the package to give support to the armv7h architecture? it uses the armhf file on the logmein server, here is the pkgbuild with the modifications to give support to that.


ztefn commented on 2014-02-09 21:18

No problem =)

karol_007 commented on 2014-02-09 21:04

I've just seen the edits, thanks a lot for your contributions, ztefn :-)

ztefn commented on 2014-02-09 20:35

karol_007: I'm fairly new to pure Arch (just did my first install this week), but I do have quite some experience with Hamachi. So I've done some cleanup on the article today; the installation section has been corrected, all information regarding Hamachi 1 has been removed and the packages in the GUI section now link to the corresponding AUR pages.

All the information should now be correct and up-to-date. I would like to invite anyone to help bringing the article further up to par and meet the style conventions.

karol_007 commented on 2014-02-01 18:48

Could somebody please have a look at the poor wiki article ? It says "Best way to install hamachi2 is to download the .deb package (...)" but you seem to be doing fine w/o it.

Xavion commented on 2013-06-21 23:28

I thought he might have meant that the daemon scripts should "keep track" of the processes they spawn.

This was already happening in the case of 'sysvinit', while it's taken care of automatically by 'systemd'.

strugee commented on 2013-06-21 01:36

(Without looking at the actual state of sysvinit and systemd in this package,) maybe he meant traction?

Xavion commented on 2013-06-07 22:31

Sorry, I'm having trouble understanding you. Perhaps someone else can explain what you mean ...

sl1pkn07 commented on 2013-06-07 15:22

ow, yes fail me, sorry

another question, why no track in pkgbuild sysvinit (deprecated) and systemd service?

Xavion commented on 2013-06-01 01:57

This package doesn't have any files in either of the 'sbin' directories.

sl1pkn07 commented on 2013-05-31 10:50

need change bin path from /usr/sbin /sbin to /usr/bin

Xavion commented on 2013-03-03 20:48

Thanks for adding ARM support; I have uploaded your PKGBUILD.

Anonymous comment on 2013-03-03 03:52

With the help of WarheadsSE (Mostly him. Alright basically only him. BUT I BUGGED HIM TO HELP) this PKGBUILD allows for hamachi installation on arm device.

ztefn commented on 2013-01-12 14:59

Thanks for your feedback. I have made a revision to Haguichi that removes any dependencies on the scripts' output (

If your not afraid of Mono you should try running Haguichi some day! With these changes the next release should work seamlessly on Arch too. :D

Xavion commented on 2013-01-10 23:15

I was referring to you, not your Haguichi application. *People* using 'systemd' should only control daemons via 'systemctl'.

Quamachi has been around for over five years, which means I introduced that Bash script long before 'systemd' came into existence.

The other 'sysvinit' scripts output the same errors when using 'systemd'. For that reason, I don't want to change this one.

By the way, I'd like to mention that I think Haguichi looks pretty impressive from the screenshots. Keep up your good work!

ztefn commented on 2013-01-10 21:52

Yes, that's obvious. The reasoning for doing so was not obvious to me because you stated earlier on that one should _only_ attempt to control the daemon via systemctl.

I think it's perfectly fine if packagers do it their own way and alter the script to work (better) with the distribution. In fact, I currently prefer to utilize these (optimized) scripts rather than shipping my own (generic) script like Quamachi. For that reason, I would just like to see these scripts to standardize on start/stop/restart output and not throw any errors when executed manually.

If your not willing to alter your sysvinit script, then no harm is done. I will adapt Haguichi to rely less (or not at all) on the scripts' output.

Xavion commented on 2013-01-09 21:47

I thought it would have been pretty obvious that the Bash script you referred to removes the need to care about whether a system uses 'sysvinit' or 'systemd'.

It's foolish to assume that all Linux distributions will use the daemon controller supplied with Hamachi. Some of them will always want to do it their own way.

ztefn commented on 2013-01-08 00:07

Quamachi doesn't seem to use the systemctl command either. Instead it uses it's own shell script that --just like the sysvinit script-- starts or kills /opt/logmein-hamachi/bin/hamachid. Seems duplicate effort to me. Why is reinventing the wheel any better than using what's already provided by the logmein-hamachi package?

I'm actually the Haguichi developer... *waves hands*

Xavion commented on 2013-01-06 22:03

If you're using 'systemd', you should only attempt to control the daemon via "systemctl [action] logmein-hamachi.service".

Haguichi shouldn't rely on the 'sysvinit' script and its description string. Tell the developer to fix his code or just use Quamachi.

ztefn commented on 2013-01-05 22:57

Using this package on Cinnarch the sysvinit script produces the following errors when using start/stop/restart:

/etc/rc.d/logmein-hamachi: line 30: /etc/rc.conf: No such file or directory
/etc/rc.d/logmein-hamachi: line 31: /etc/rc.d/functions: No such file or directory
/etc/rc.d/logmein-hamachi: line 88: stat_done: command not found
/etc/rc.d/logmein-hamachi: line 88: rm_daemon: command not found
/etc/rc.d/logmein-hamachi: line 78: stat_done: command not found
/etc/rc.d/logmein-hamachi: line 78: add_daemon: command not found
/etc/rc.d/logmein-hamachi: line 103: stat_done: command not found

Also, because the sysvinit script has a different DESC than the original hamachi-init script Haguichi doesn't recognize the output.

Xavion commented on 2012-12-21 03:47

I have removed all custom variables.

Xavion commented on 2012-12-21 03:46

I have removed most custom variables.

Xavion commented on 2012-12-18 21:24

The syntax highlighting of my text editor led me to believe this was okay. As I can now see that you're correct, I have modified the PKGBUILD accordingly.

Xavion commented on 2012-12-18 08:51

You called the use of such variables 'cute' here:

Now you're calling them 'needless' and the use of them 'silly' here. So, which is it? Are they 'cute', or 'needless' and 'silly'?

Either way, do you actually want me to remove them despite the fact that they work perfectly as is?

Regarding the quoting of variables, here's how the PKGBUILD looks in KWrite:

It certainly appears that the way I've done it does quote the variables. Are you saying that KWrite's syntax highlighting needs fixing?

keenerd commented on 2012-12-17 17:31

Please don't use needless variables.
is just silly.

Please fix all of your quoting.
does not actually quote $pkgdir.

haohaolee commented on 2012-12-12 07:39 is released

Xavion commented on 2012-11-19 04:58

The 'sysvinit' and 'systemd' packages are optional dependencies. There's no need to list either as your system has already booted.

Aelius commented on 2012-11-19 00:49

latest pkgbuild doesn't download sysvinit, but the tarball works.

panthar commented on 2012-11-15 21:24 released upstream, breaking source download

8B3CE273 commented on 2012-11-12 20:17

version released

kite commented on 2012-11-12 17:42

sources url returns 404

ztefn commented on 2012-10-21 11:06

As I wrote on july 30th, this old client cannot connect. So it would be totally useless to provide a mirror. Please remove this package instead.

ReNoM commented on 2012-10-21 10:01

have any mirror?

Anonymous comment on 2012-10-21 08:40

An error occured curl: (22) The requested URL returned error: 400 Bad Request

ztefn commented on 2012-07-30 20:38

As of today this legacy version will be unable to connect. Use the logmein-hamachi package instead.

== quote from LogMeIn website ( ==

As a result of upcoming enhancements to the Hamachi service, those using old clients must update to continue connecting to Hamachi servers. Those who do not update their client to the supported versions by July 30th, 2012 will be unable to connect. Please visit the LogMeIn Download Page to update if you are using versions prior to:


== end quote ==

Aelius commented on 2012-04-13 04:54

I think this version was just axed by logmein though; both of my FreeBSD machines can no longer ping any version client.

rafaelff commented on 2012-04-10 18:12

@link68759: can you point out to FreeBSD's hamachi package and sources ? Maybe we can figure it out how to make it work from their package.

Aelius commented on 2012-04-10 17:30

Unlikely, the same version of hamachi works under FreeBSD.

However this version of hamachi is incompatible with hamachi2 clients.

This is quite frustrating: there is no hamachi2 for freebsd, and no hamachi 0.9 for archlinux, thus these systems cannot communicate under a hamachi network :/

Anonymous comment on 2012-04-09 18:42

Got the same error "bad response 00007f00"

Apparently hamachi uses ifconfig which recently has been removed/moved in favor of the new ip commands.
Try installing "net-tools", that should fix it, could also be added the dependencies.

But currently the login fails for me too, I hope it's only a temporary issue with the hamachi host.

Aelius commented on 2012-04-04 16:59

Fails to login upon install

$ hamachi login
logging in......failed

After reboot, this happens

$hamachi start
04 12:58:05.468 [ 0] [ 959] tap: bad response 00007f00
04 12:58:05.468 [ 0] [ 959] Failed to configure tap device to use

dundee commented on 2011-11-25 12:16

Thanks, josephgbr! Package updated.

blackdeagle commented on 2011-11-24 11:36

Thanks, josephgbr! The patch and your PKGBUILD works perfectly!

Xavion commented on 2011-11-20 22:59

Thanks for letting me know about this; I've modified the package accordingly.

sl1pkn07 commented on 2011-11-20 10:13

is problem /etc/rc.d/logmein-hamachi

change "#! /bin/sh" to "#! /bin/bash"


Xavion commented on 2011-11-20 02:23

I'm receiving the same error after upgrading 'initscripts' to v2011.11.3-1 from 'core'. Since issuing "sh /etc/rc.d/functions" invokes the same error, I'm confident that it has nothing to do with this "logmein-hamachi" package.

sl1pkn07 commented on 2011-11-19 16:41

(1/1) instalando logmein-hamachi [-----------------------------------------------------------------------------] 100%
/etc/rc.d/functions: line 497: syntax error near unexpected token `<'
/etc/rc.d/functions: line 497: ` done < <(findmnt -runRo TARGET,FSTYPE,OPTIONS / | tac)'
Starting LogMeIn Hamachi VPN tunneling engine logmein-hamachi [DONE]

rafaelff commented on 2011-10-01 03:48

Well, since the package is still here (not deleted), please fix z0id's problem with this patch

Also, may I suggest a reorganized PKGBUILD?

Xavion commented on 2011-08-25 05:38

You should install the "logmein-hamachi" (v2) package instead. This (v1) one was deprecated by LogMeIn many months ago.

z0id commented on 2011-08-24 14:45

Compilation fails with

In function ‘open’,
inlined from ‘main’ at tuncfg.c:102:5:
/usr/include/bits/fcntl2.h:51:24: error: call to ‘__open_missing_mode’ declared with attribute error: open with O_CREAT in second argument needs 3 arguments
make: *** [tuncfg] Error 1

exterm commented on 2011-05-31 06:42

version is on the hamachi site.

jcci commented on 2011-02-11 03:18

Just a warning in case nobody noticed so far:
In case you change corresponding clients all to logmein-hamachi 2xx you will en up with an unusable slow network, as long as you don't have a logmein-pro contract. I just had this issue and 2xx clients just connect "via relay" or " via server" instead of "direct UDP". It seems this is not possible as long as one side uses the old Hamachi software. Once I changes back, everything was OK.
This issue I found confirmed in some forums. So this version is just good if you plan to purchase a commercial logmein-pro contract.

Xavion commented on 2010-08-10 05:44

I plan to do that when other packages - that still depend on Hamachi v1 - become compatible with LogMeIn Hamachi v2.

f0x commented on 2010-08-09 11:23

add provides=("${product}")

Xavion commented on 2010-07-16 07:34

You're wrong that 'pacman' doesn't parse custom variables these days. My intention wasn't to show my personal skills to others, as I wasn't expecting anyone to bother reading this PKGBUILD.

Anyone who can't figure out what I've done in the PKGBUILD shouldn't become the maintainer of this package. You're making a big deal out of nothing and I urge you to focus your attention elsewhere.

baghera commented on 2010-07-15 11:39

Why do you say I have no idea of what I'm talking about? You use company in pkgname , in url and in sources (not the part of the filename), it can be replaced by logmein without problems, mainly because
1. the informations stored in that variable is not subject to change (I guess url will always be for example)
2. that variable is not parsed by pacman -Qii or pacman -Sii or whatever.
I can say the same things about product, for example there's no problem doing install -m 755 hamachid instead of install -m 755 ${product}d (which is harder to read BTW).

Besides, please consider you're not the only one with a degree around here, advices are intended to make aur and the community better (making easier to maintain packages when the original submitter/maintainer steps down for example, and that is the case) and not to show our personal skills to others.

Xavion commented on 2010-07-15 09:40

You think you know what you're talking about, but your comment about the 'company' and 'product' variables being useless shows that you really have no idea.

I've already had this discussion many times and I'm going to avoid delving into it again. I have a degree in computer engineering and I won't be following orders given by the likes of you.

Xavion commented on 2010-07-15 09:31

I've already had this discussion many times and I'm going to avoid delving into it again. I have a degree in computer engineering and I won't be following orders given by the likes of you.

baghera commented on 2010-07-14 10:16

I'm sure the variables you introduced are not in conflict with others used by makepkg, but this is a matter of _standards_ , people on the aur should adhere to arch packaging standards to avoid future problems.
BTW, company and product are really useless, you can simply have pkgname=logmein-hamachi , filearch has reason to be there, but as _filearch .

Xavion commented on 2010-07-14 10:10

As usual, I've ensured that none of the variables I've introduced conflict with anything. If this changes later on, I'll be the first to make any necessary changes.

baghera commented on 2010-07-14 10:00

As stated here (the third point) custom variables (such as company , product and filearch in your PKGBUILD) should really have an underscore as prefix. Please fix this.

Xavion commented on 2010-07-12 06:02

For the reasons I've stated below, I think v2 should remain "logmein-hamachi" and v0.9 should become "(logmein-)hamachi-legacy".

Ape commented on 2010-07-11 06:15

Don't you think that when v2 goes stable it should become hamachi? And this would become hamachi1 or hamachi-old.

Xavion commented on 2010-07-10 23:17

The 'logmein-' prefix is now relevant, since the v2 series is linked with other LogMeIn services. I relayed the beta notice from the LogMeIn website in my first comment for that package. I've handled this the same way that LogMeIn did, which is another reason why it's unnecessary to add the '-beta' suffix.

Xavion commented on 2010-07-10 11:26

I've already explained to you several times why your suggestions probably don't fit these uncommon circumstances. If you still don't understand, just keep rereading my comments until they sink in. For the benefit of others, they're located here:

Ape commented on 2010-07-10 06:35

This should be renamed to hamachi-beta. hamachi2 might be also a good name.

Ape commented on 2010-07-10 06:35

I really think that you should remove logmein-hamachi and upload it as hamachi-beta. The logmein prefix has no real purpose.

Ape commented on 2010-07-10 06:34

I really thing that you should remove logmein-hamachi and upload it as hamachi-beta.

Xavion commented on 2010-07-10 01:31

This replaces the 'hamachi' (v0.9.9.9) package, which is no longer advertised on its website. LogMeIn warns that the v2 series is currently beta software and should not be used in a production environment.

Xavion commented on 2010-07-10 01:16

To save you the trouble of upgrading this package, I've just uploaded a "logmein-hamachi" package to AUR. You should leave this package at v0.9.9.9, even after the v2 series is officially classified as stable.

Ape commented on 2010-07-09 09:53

Well this discussion is pretty much pointless until someone creates a PKGBUILD for the new version.

Xavion commented on 2010-07-09 09:27

Yeah, I knew that already. My point is that LogMeIn probably considers v2.0.0.11 to be more stable than v0.9.9.9, given that the latter no longer works on some Linux distributions. I'm guessing that's why they've removed all links to it from their website.

They might just be stating that v2 currently isn't stable to avoid any potential legal issues if it fails. In any event, some might argue that since Arch Linux is a bleeding-edge distribution, this package should be upgraded to the v2.0.0.11 beta edition now :-).

Ape commented on 2010-07-09 07:27

LogMeIn clearly states that v2 for Linux is in beta and that it's not stable [1]. Moreover, it also support v0.9.9.9 still on it's website even if it has cut down the advertisements for it [2].


Xavion commented on 2010-07-09 07:21

It's good to know that we all agree on creating a separate package, one way or another :-). The next release of Quamachi will feature basic support for Hamachi v2.

If LogMeIn still considers Hamachi v0.9.9.9 to be stable, why have all references to it vanished from their website? It's now failing on other Linux distributions and LogMeIn probably realises that it's not so stable anymore!

Ape commented on 2010-07-09 07:04 is still the stable version. Please create a new package for the new version.

Garret commented on 2010-07-09 06:26

I think a separate pkgbuild is the better solution. Then at the moment a gui for the new hamachi doesn't exist.

Xavion commented on 2010-07-09 06:22

A link to the tarball of the v0.9.9.9 edition is no longer advertised on the Hamachi website. I think this is an indication that LogMeIn wants Linux users to upgrade immediately. Based on this, it's still my opinion that you should do as I suggested below relatively soon.

Ape commented on 2010-07-08 06:09

"this is Beta software and is in no way meant to be used in a production environment."

As v2.0.0.11 is still a development version it should not replace this stable package yet. Instead you should create a new package called hamachi-beta, hamachi-dev, or hamachi2 etc. When it goes stable it will replace this package.

Xavion commented on 2010-07-08 02:06

The v0.9.9.9 edition has been replaced by v2.0.0.11, which is available here: Before upgrading this package, it should be re-uploaded to AUR as 'hamachi-legacy'.