Bug#340983: Hopefully fixed now

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

Bug#340983: Hopefully fixed now

Kari Pahula-2
I stopped using s-s-d generated pid file in 0.2-2.  My feeling about
failing logrotates was that logrotate sent the HUP signal to a wrong
process.  There isn't code in place to prevent multiple preload
daemons from starting so this could well have happened.

Hopefully this change will fix this bug, too.  Does anybody see
logrotate still failing?


--
To UNSUBSCRIBE, email to [hidden email]
with a subject of "unsubscribe". Trouble? Contact [hidden email]

Reply | Threaded
Open this post in threaded view
|

Bug#340983: Hopefully fixed now

Mario 'BitKoenig' Holbe
On Thu, Dec 08, 2005 at 07:35:00PM +0200, Kari Pahula wrote:
> I stopped using s-s-d generated pid file in 0.2-2.  My feeling about
> failing logrotates was that logrotate sent the HUP signal to a wrong
...
> Hopefully this change will fix this bug, too.  Does anybody see
> logrotate still failing?

Well, I didn't see it failing before 0.2-2 but now I do :)

/etc/cron.daily/logrotate:
start-stop-daemon: stat preload: No such file or directory (No such file or directory)
error: error running postrotate script for /var/log/preload.log
run-parts: /etc/cron.daily/logrotate exited with return code 1

ii  preload        0.2-2          an adaptive readahead daemon

You should probably give the full path to the preload binary to
start-stop-daemon.


regards
   Mario
--
Whenever you design a better fool-proof software,
the genetic pool will always design a better fool.

signature.asc (492 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Bug#340983: Hopefully fixed now

Mario 'BitKoenig' Holbe
On Fri, Dec 09, 2005 at 10:10:44AM +0100, Mario 'BitKoenig' Holbe wrote:
> You should probably give the full path to the preload binary to
> start-stop-daemon.

Oh, and btw.... please consider using /etc/init.d/preload reload or
restart in the logrotate script instead of invoking start-stop-daemon
directly. This is probably less error-prone in the future when you
change the init-script and forget to change the logrotate script :)


Mario
--
The question of whether a computer can think is no more interesting than
the question of whether a submarine can swim.          -- E. W. Dijkstra

signature.asc (492 bytes) Download Attachment