Subject: Re: Slurpd crashes during exit
To: matthew sporleder <msporleder@gmail.com>
From: Jean-Luc Wasmer <jl+netbsd@lists.wasmer.ca>
List: current-users
Date: 08/31/2005 00:02:43
When running in the foreground, I would simply Ctrl-C it (SIGINT).
With the rc script, it looks like it's SIGTERM.

I first noticed the slurpd.core file (in the directory where I put my 
replica log file... which happens to be the directory I use for the db) 
which is why I tried running in the foreground (with -d 4).

I don't think it matters, but I never stopped slapd (only slurpd).

The current version in pkgsrc is 2.2.27. I'll upgrade by the end of the 
week.

Jean-Luc

matthew sporleder wrote:

>What signal are you using to kill it?
>It looks like this might be exposing a bug in LWP?  Since when you
>kill slurpd and slapd it basically sends an exit to all of its
>children (the slave connections) and then drops out.  By the way, that
>openldap is a little old,, is it the default from pkgsrc?
>
>
>On 8/30/05, Jean-Luc Wasmer <jl+netbsd@lists.wasmer.ca> wrote:
>  
>
>>Hi,
>>
>>I'm not sure who's to blame (pthread or slurpd code) and it's not a big
>>issue for me, but someone may be interested in that problem.
>>Whether I kill slurpd from the foreground or with the rc.d script, I get
>>a "core dumped".
>>
>>end replication thread for my.server.com:636
>>fm: exiting
>>assertion "unreachable" failed: file
>>"/usr/src/lib/libpthread/pthread.c", line 620, function "pthread_exit"
>>Abort (core dumped)
>>
>>
>>Configuration:
>>- NetBSD 3.99.7
>>- openldap-2.2.20nb2
>>
>>Jean-Luc
>>