| Summary: | dev-util/buildbot-slave-0.8.3 init-script issues | ||
|---|---|---|---|
| Product: | Gentoo Linux | Reporter: | Johan Bergström <bugs> |
| Component: | New packages | Assignee: | Python Gentoo Team <python> |
| Status: | RESOLVED DUPLICATE | ||
| Severity: | normal | ||
| Priority: | Normal | ||
| Version: | unspecified | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Package list: | Runtime testing required: | --- | |
| Attachments: | init-debug.log | ||
|
Description
Johan Bergström
2011-03-08 15:05:41 UTC
Guess I should also note that the script itself has no issues actually starting, it's just start-stop-daemon that has a hard time. Can you confirm this is still an issue for you? It appears to have been taken care of in the tree. Sorry to say that it still fails: s.nu /etc # eix -cs buildbot-slave [I] dev-util/buildbot-slave (0.8.3@05/04/2011): BuildBot Slave Daemon s.nu /etc # /etc/init.d/buildslave start * Starting buildslave in /var/lib/build/node ... [ !! ] # ps aux | grep buildslave buildbot 17958 0.0 1.1 25536 9096 ? Sl 23:49 0:00 /usr/bin/python2.7 /usr/bin/twistd --no_save --logfile=/var/lib/build/node/twistd.log --pidfile=/var/lib/build/node/buildslave.pid --python=/var/lib/build/node/buildbot.tac Guess I should be a good gentoo user and upgrade to baselayout-2. I'll report back after Ah I see what you mean now. You are correct, it appears to be the way we call python. Try this and let me know if it works...
--env HOME=${BASEDIR}" \
--pidfile="${BASEDIR}/buildslave.pid" \
--exec /usr/bin/twistd -- \
--no_save \
Keep the other lines the same. Basically change the command that is being executed to twistd and add the --pidfile to start-stop-daemon
Created attachment 272155 [details]
init-debug.log
Unfortunately, the patched init-script didn't work. Attaching the debug output of patched --start.
Thanks
(In reply to comment #6) > Created attachment 272155 [details] > init-debug.log > > Unfortunately, the patched init-script didn't work. Attaching the debug output > of patched --start. > > Thanks Took a brief look at it, and by re-adding --pidfile to twisted as well, we have a working script. Here's the "whole" ordeal: start-stop-daemon --start -c "${USERNAME}" \ --env HOME="${BASEDIR}" \ --pidfile="${BASEDIR}/buildslave.pid" \ --exec /usr/bin/twistd -- \ --no_save \ --pidfile="${BASEDIR}/buildslave.pid" \ --logfile="${BASEDIR}/twistd.log" \ --python="${BASEDIR}/buildbot.tac" eend $? This most likely applies to buildbot as well (master) (In reply to comment #7) > (In reply to comment #6) > > Created attachment 272155 [details] > > init-debug.log > > > > Unfortunately, the patched init-script didn't work. Attaching the debug output > > of patched --start. > > > > Thanks > > Took a brief look at it, and by re-adding --pidfile to twisted as well, we have > a working script. Here's the "whole" ordeal: > > start-stop-daemon --start -c "${USERNAME}" \ > --env HOME="${BASEDIR}" \ > --pidfile="${BASEDIR}/buildslave.pid" \ > --exec /usr/bin/twistd -- \ > --no_save \ > --pidfile="${BASEDIR}/buildslave.pid" \ > --logfile="${BASEDIR}/twistd.log" \ > --python="${BASEDIR}/buildbot.tac" > eend $? > > This most likely applies to buildbot as well (master) Yep. This is what I had intended you set it to. You need to provide the pidfile to both twisted and start-stop-daemon So do we need it on both s-s-d and twistd, or will just the latter suffice? (In reply to comment #9) > So do we need it on both s-s-d and twistd, or will just the latter suffice? According to comment 8, both. I haven't seen this happen in a while now. Can't say if there's been an upstream change or not, but our init scripts seems similar to what they were before. Close? Add it anyway? I took the fix from bug 397759, which I think was basically a duplicate? Closing this for now, feel free to reopen if there are still issues. *** This bug has been marked as a duplicate of bug 397759 *** |