codebrowse is writing it's PID file into the source tree

Bug #689961 reported by Steve McInerney
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Launchpad itself
Triaged
High
Unassigned

Bug Description

From an init.d stop:
open("/srv/codebrowse.launchpad.net/production/launchpad-rev-12021/scripts/loggerhead.pid", O_RDONLY) = -1 ENOENT (No such file or directory)

The startup/stop commands are:
ROOTDIR=/srv/codebrowse.launchpad.net
name=loggerhead
rundir=${ROOTDIR}/production/launchpad
export LPCONFIG=codebrowse1

make -C "$rundir" LPCONFIG=${LPCONFIG} start_codebrowse

make -C "$rundir" LPCONFIG=${LPCONFIG} stop_codebrowse

As we have 2 codebrowses running on guava, ideally the PID would be something like:
/var/tmp/${LPCONFIG}.pid or thereabouts, so as to keep the two separately identified.

Steve McInerney (spm)
tags: added: canonical-losa-lp
Revision history for this message
Tom Haddon (mthaddon) wrote :

Actually, can we have the pidfile be written to /srv/codebrowse.launchpad.net/var/ (for production), which I think is where it was before. We also need to make sure there's some way of preventing pidfile locations from changing in production (unless we really want to, and have warning ahead of time) because initscripts depend on them, and if they change, it breaks rollouts in a very painful way.

Changed in launchpad-code:
importance: Undecided → High
Curtis Hovey (sinzui)
Changed in launchpad:
status: New → Triaged
Revision history for this message
Robert Collins (lifeless) wrote :

I think this is a dup; dunno the #.

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.