[Bug 854894] New: booth-arbitrator init script broken
https://bugzilla.novell.com/show_bug.cgi?id=854894 https://bugzilla.novell.com/show_bug.cgi?id=854894#c0 Summary: booth-arbitrator init script broken Classification: openSUSE Product: openSUSE Factory Version: 13.2 Milestone 0 Platform: Other OS/Version: Other Status: NEW Severity: Major Priority: P5 - None Component: High Availability AssignedTo: lmb@suse.com ReportedBy: tserong@suse.com QAContact: qa-bugs@suse.de Found By: --- Blocker: --- On openSUSE 13.1 and Factory, if I try to start booth-arbitrator: # /etc/init.d/booth-arbitrator start redirecting to systemctl start booth-arbitrator ..(very long delay here)... Job for booth-arbitrator.service failed. See 'systemctl status booth-arbitrator.service' and 'journalctl -xn' for details. factory-1:~ # systemctl status booth-arbitrator.service -l booth-arbitrator.service - LSB: start and stop BOOTH arbitrator daemon Loaded: loaded (/etc/init.d/booth-arbitrator) Active: failed (Result: timeout) since Wed 2013-12-11 20:39:34 EST; 18s ago Process: 1907 ExecStart=/etc/init.d/booth-arbitrator start (code=exited, status=0/SUCCESS) Dec 11 20:34:47 factory-1 booth-arbitrator[1916]: [1916]: debug: catchup result: name: ticketA, owner: -1, ballot: 0, expires: 0 Dec 11 20:34:47 factory-1 booth-arbitrator[1916]: [1916]: debug: attempting catchup from 192.168.4.250 Dec 11 20:34:50 factory-1 booth-arbitrator[1916]: [1916]: ERROR: connection to 192.168.4.250 error No route to host Dec 11 20:34:50 factory-1 booth-arbitrator[1916]: [1916]: debug: attempting catchup from 192.168.4.251 Dec 11 20:34:53 factory-1 booth-arbitrator[1916]: [1916]: ERROR: connection to 192.168.4.251 error No route to host Dec 11 20:34:53 factory-1 booth-arbitrator[1916]: [1916]: debug: catchup result: name: ticketB, owner: -1, ballot: 0, expires: 0 Dec 11 20:34:53 factory-1 booth-arbitrator[1916]: [1916]: info: BOOTH arbitrator daemon started Dec 11 20:39:34 factory-1 systemd[1]: booth-arbitrator.service operation timed out. Terminating. Dec 11 20:39:34 factory-1 systemd[1]: Failed to start LSB: start and stop BOOTH arbitrator daemon. Dec 11 20:39:34 factory-1 systemd[1]: Unit booth-arbitrator.service entered failed state. `journalctl -xn` gives: Dec 11 20:39:34 factory-1 systemd[1]: booth-arbitrator.service operation timed out. Terminating. Dec 11 20:39:34 factory-1 systemd[1]: Failed to start LSB: start and stop BOOTH arbitrator daemon. -- Subject: Unit booth-arbitrator.service has failed I suspect that when systemd takes over the init script, somehow booth is not daemonizing correctly so remains attached to the terminal. Note that even after the failure reported above, booth is actually running: # ps ax|grep booth 1916 ? SLs 0:00 /usr/sbin/boothd arbitrator But the init script doesn't think it is: # /etc/init.d/booth-arbitrator status BOOTH arbitrator daemon is cat: /var/run/booth.pid: No such file or directory cat: /var/run/booth.pid: No such file or directory Stopped So the pidfile is missing for some reason. Rather than fixing the init script or the daemon, or whatever is causing this behaviour, I'd suggest throwing away the init script and replacing it with a systemd unit file. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
https://bugzilla.novell.com/show_bug.cgi?id=854894 https://bugzilla.novell.com/show_bug.cgi?id=854894#c Lars Marowsky-Bree <lmb@suse.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Priority|P5 - None |P3 - Medium CC| |lmb@suse.com AssignedTo|lmb@suse.com |philipp.marek@linbit.com -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
https://bugzilla.novell.com/show_bug.cgi?id=854894 https://bugzilla.novell.com/show_bug.cgi?id=854894#c1 Philipp Marek <philipp.marek@linbit.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |UPSTREAM --- Comment #1 from Philipp Marek <philipp.marek@linbit.com> 2013-12-12 07:41:02 UTC --- I've already done some changes to the initscript (to prepare for multiple booth instances, ie. multi-tenancy), so that might be fixed. Please note that what you're seeing here is the old behaviour - a single CATCHUP try for each remote site to get the current ticket status. In git HEAD this is already replaced with some retry logic that keeps asking until sufficient information has been collected. So the pid file and the TCP socket (for status queries) should be available much sooner - perhaps that's enough to fix that problem. The systemd integration is still on the TODO list, btw - I'll have yet to read more about that, because we might need multiple instances for multi-tenancy. The init script just started a daemon for each existing config file - I don't know yet what to do for systemd. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
participants (1)
-
bugzilla_noreply@novell.com