Control: tags -1 + pending On Tue, 2016-02-16 at 10:09 +0100, Michael Braun wrote: > After installing some unrelated program using apt-get, it was > discovered that nslcd needs a restart und I was promted to restart it > (among other services).
Were you using needrestart for this? I've had needrestart not correctly restart at least ntp, rpcbind and mailman. I suspect there is a bug in needrestart and or the combination with systemd but I've not been able to track it down. > The logs when the error occured read: > > Feb 15 12:12:01 gate nslcd[26487]: caught signal SIGTERM (15), shutting down > Feb 15 12:12:01 gate nslcd[21177]: Stopping LDAP connection daemon: nslcd. > Feb 15 12:12:01 gate nslcd[21302]: version 0.9.4 starting > Feb 15 12:12:04 gate nslcd[26487]: thread 2 is still running, shutting down > anyway > Feb 15 12:12:04 gate nslcd[26487]: version 0.9.4 bailing out > Feb 15 12:12:06 gate nslcd[21302]: accepting connections > Feb 15 12:12:06 gate nslcd[21231]: Starting LDAP connection daemon: nslcd. The last line I think is systemd writing a log entry using the nslcd. The lines are out of order for some reason and the new daemon in started while the old one is still shutting down. The problem could be that the init script is not called with restart but with separate stop and start actions (this is slightly different). I will update the init script to ensure that stop also only returns when nslcd has actually stopped. Thanks for pointing this out. This could also be the issue with some of the other services that have not been restarted correctly. Thanks, -- -- arthur - adej...@debian.org - http://people.debian.org/~adejong --
signature.asc
Description: This is a digitally signed message part
