Hi, I've released syslog-ng 1.3.8 and libol 0.2.9 with a couple of bugfixes. -- Bazsi PGP info: KeyID 9AF8D0A9 Fingerprint CD27 CFB0 802C 0944 9CFD 804E C82C 8EB1 url: http://www.balabit.hu/pgpkey.txt
syslog is now stable on my machines, but it seems that it has been restarting itself. i control everything in a cluster fashion so when i rotate syslog into sql i run a stop/start script in perl which logs into all the machines and issues a stop/start through init scripts. sometimes it works but sometimes it wont due to the fact that the pid in /var/run is differant than the current pid, telling me that syslog-ng has somehow restarted itself and is now running under a new pid. has this issue been brought up before? if so might it be fixed in 1.3.8? thanks -jacob * * * * * * * * * * * * * * * * * * Jacob Zehnder * * * * * * * * * Grill Master * * * * * * * * "The dude abides." * * * * 'The Big Lebowski' * * * * * * * * * * * * * *
On Wed, Dec 15, 1999 at 03:47:19PM -0800, Jacob Zehnder wrote:
syslog is now stable on my machines, but it seems that it has been restarting itself. i control everything in a cluster fashion so when i rotate syslog into sql i run a stop/start script in perl which logs into all the machines and issues a stop/start through init scripts. sometimes it works but sometimes it wont due to the fact that the pid in /var/run is differant than the current pid, telling me that syslog-ng has somehow restarted itself and is now running under a new pid. has this issue been brought up before? if so might it be fixed in 1.3.8?
It's fixed in 1.3.8. Syslog-ng didn't truncate it's pidfile before writing the new pid, so if the new pid is shorter, than the pidfile was, the contents became invalid. -- Bazsi PGP info: KeyID 9AF8D0A9 Fingerprint CD27 CFB0 802C 0944 9CFD 804E C82C 8EB1 url: http://www.balabit.hu/pgpkey.txt
participants (2)
-
Balazs Scheidler
-
Jacob Zehnder