I?ve been having the same issue with Ubuntu, thankfully we have access to Upstart: ``` start on (local-filesystems and net-device-up IFACE!=lo) stop on stopping network-services author "Mark Lopez" description "Tinc Upstart Job" version "0.1" env network=master respawn exec /usr/sbin/tincd -n "$network" -D --debug=3 --logfile ``` I removed the default init script and switched over to Upstart. Restarts will occur as soon as the process dies. Save the job as ?/etc/init/tinc.conf?. From: tinc [mailto:tinc-bounces at tinc-vpn.org] On Behalf Of Daniel J. Grinkevich Sent: Friday, January 29, 2016 8:25 AM To: tinc at tinc-vpn.org Subject: Re: respawning dead tinc process This is what we use on our routers, running once a minute via crontab. if pgrep "tincd" >/dev/null; then echo "tincd is running" else echo "tincd isn't running, restarting" tincd -n nycmesh fi On Fri, Jan 29, 2016 at 9:07 AM, pjv <pjv at pjv.me<mailto:pjv at pjv.me>> wrote: I have tinc 1.1pre11 running on various routers and linux cloud servers. On one of the cloud servers, under Ubuntu 12.04, tinc is mysteriously dying once in a while, leaving a dangling PID. I have been unable to track down why it is dying, but it happens infrequently enough that I care less about why it is dying than how to robustly respawn it when it dies. Before I re-invent the wheel, has anyone come up with a solution for automatically detecting a dead tincd and respawning the process on linux? _______________________________________________ tinc mailing list tinc at tinc-vpn.org<mailto:tinc at tinc-vpn.org> http://www.tinc-vpn.org/cgi-bin/mailman/listinfo/tinc -- --- GPG Key: 0x160B24D1C08FB4E4<https://pgp.mit.edu/pks/lookup?op=get&search=0x160B24D1C08FB4E4> -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://www.tinc-vpn.org/pipermail/tinc/attachments/20160129/b9d1a224/attachment.html>
Hi Daniel, Yes, thanks and I should have mentioned that I am using pretty much the same thing you suggested on my routers, but for whatever reason couldn?t get that approach working on the linux box in question. Mark, Thanks for the upstart script. I had tried to do the same myself but couldn?t get my upstart script to work right. I was missing the ?-D? flag in my script. Added that and now it?s working great.> On Jan 29, 2016, at 8:42 AM, Mark Lopez <m at silvenga.com> wrote: > > I?ve been having the same issue with Ubuntu, thankfully we have access to Upstart: > > ``` > start on (local-filesystems and net-device-up IFACE!=lo) > stop on stopping network-services > > author "Mark Lopez" > description "Tinc Upstart Job" > version "0.1" > > env network=master > > respawn > > exec /usr/sbin/tincd -n "$network" -D --debug=3 --logfile > ``` > > I removed the default init script and switched over to Upstart. Restarts will occur as soon as the process dies. Save the job as ?/etc/init/tinc.conf?. > > From: tinc [mailto:tinc-bounces at tinc-vpn.org <mailto:tinc-bounces at tinc-vpn.org>] On Behalf Of Daniel J. Grinkevich > Sent: Friday, January 29, 2016 8:25 AM > To: tinc at tinc-vpn.org <mailto:tinc at tinc-vpn.org> > Subject: Re: respawning dead tinc process > > This is what we use on our routers, running once a minute via crontab. > > if pgrep "tincd" >/dev/null; then > echo "tincd is running" > else > echo "tincd isn't running, restarting" > tincd -n nycmesh > fi > > On Fri, Jan 29, 2016 at 9:07 AM, pjv <pjv at pjv.me <mailto:pjv at pjv.me>> wrote: > I have tinc 1.1pre11 running on various routers and linux cloud servers. On one of the cloud servers, under Ubuntu 12.04, tinc is mysteriously dying once in a while, leaving a dangling PID. I have been unable to track down why it is dying, but it happens infrequently enough that I care less about why it is dying than how to robustly respawn it when it dies. > > Before I re-invent the wheel, has anyone come up with a solution for automatically detecting a dead tincd and respawning the process on linux? > > _______________________________________________ > tinc mailing list > tinc at tinc-vpn.org <mailto:tinc at tinc-vpn.org> > http://www.tinc-vpn.org/cgi-bin/mailman/listinfo/tinc <http://www.tinc-vpn.org/cgi-bin/mailman/listinfo/tinc>-------------- next part -------------- An HTML attachment was scrubbed... URL: <http://www.tinc-vpn.org/pipermail/tinc/attachments/20160129/f0d25e16/attachment-0001.html> -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 801 bytes Desc: Message signed with OpenPGP using GPGMail URL: <http://www.tinc-vpn.org/pipermail/tinc/attachments/20160129/f0d25e16/attachment-0001.sig>
There arre systemd services here which also do restart on a failure: http://tinc-vpn.org/git/browse?p=tinc;a=tree;f=systemd;hb=refs/heads/1.1 Cheers> Am 29.01.2016 um 16:15 schrieb pjv <pjv at pjv.me>: > > Hi Daniel, > > Yes, thanks and I should have mentioned that I am using pretty much the same thing you suggested on my routers, but for whatever reason couldn?t get that approach working on the linux box in question. > > Mark, > > Thanks for the upstart script. I had tried to do the same myself but couldn?t get my upstart script to work right. I was missing the ?-D? flag in my script. Added that and now it?s working great. > >> On Jan 29, 2016, at 8:42 AM, Mark Lopez <m at silvenga.com> wrote: >> >> I?ve been having the same issue with Ubuntu, thankfully we have access to Upstart: >> >> ``` >> start on (local-filesystems and net-device-up IFACE!=lo) >> stop on stopping network-services >> >> author "Mark Lopez" >> description "Tinc Upstart Job" >> version "0.1" >> >> env network=master >> >> respawn >> >> exec /usr/sbin/tincd -n "$network" -D --debug=3 --logfile >> ``` >> >> I removed the default init script and switched over to Upstart. Restarts will occur as soon as the process dies. Save the job as ?/etc/init/tinc.conf?. >> >> From: tinc [mailto:tinc-bounces at tinc-vpn.org] On Behalf Of Daniel J. Grinkevich >> Sent: Friday, January 29, 2016 8:25 AM >> To: tinc at tinc-vpn.org >> Subject: Re: respawning dead tinc process >> >> This is what we use on our routers, running once a minute via crontab. >> >> if pgrep "tincd" >/dev/null; then >> echo "tincd is running" >> else >> echo "tincd isn't running, restarting" >> tincd -n nycmesh >> fi >> >> On Fri, Jan 29, 2016 at 9:07 AM, pjv <pjv at pjv.me> wrote: >> I have tinc 1.1pre11 running on various routers and linux cloud servers. On one of the cloud servers, under Ubuntu 12.04, tinc is mysteriously dying once in a while, leaving a dangling PID. I have been unable to track down why it is dying, but it happens infrequently enough that I care less about why it is dying than how to robustly respawn it when it dies. >> >> Before I re-invent the wheel, has anyone come up with a solution for automatically detecting a dead tincd and respawning the process on linux? >> >> _______________________________________________ >> tinc mailing list >> tinc at tinc-vpn.org >> http://www.tinc-vpn.org/cgi-bin/mailman/listinfo/tinc > > > > _______________________________________________ > tinc mailing list > tinc at tinc-vpn.org > http://www.tinc-vpn.org/cgi-bin/mailman/listinfo/tinc-------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 801 bytes Desc: Message signed with OpenPGP using GPGMail URL: <http://www.tinc-vpn.org/pipermail/tinc/attachments/20160211/c329860b/attachment.sig>