On Fri, 07 May 2010 18:57:49 +0200 hennr <he...@hennr.name> wrote: > I already removed the old (lenny) dump, this attached one is from the > squeeze installation. > > If I leave the "ERLANG_NODE=ejabberd" commented > in /etc/default/ejabberd the 2 logs attached get created. > > If I uncomment it, the dump gets created.
At this moment I have no access to a box with Erlang R13B to inspect the dump, but the logs clearly state that ejabberd failed to start because of requested sockets being already in use (errors in both logs indicate this -- see at the top of ejabberd.log for instance). This *might* indicate a bogus error: that is, you had another ejabberd instance stuck in memory with that sockets open. So, retry this, but first verify no stuck instance is present, this can be done by grepping the list of processes for the "beam" (or "beam.smp" -- depends on configuration) binary and killing it, if present. Then verify you don't have the target sockets open, that is, $ netstat -lt|grep :5222 must show nothing. Will report on the crash dump later. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
