Ummm... Importnat word mising. I'd not use the single-threading version
NOW. The reasoning was in the mail. I may ask some time later for it,
though.

Sorry...

Rainer

> -----Original Message-----
> From: [EMAIL PROTECTED] [mailto:rsyslog-
> [EMAIL PROTECTED] On Behalf Of Rainer Gerhards
> Sent: Monday, September 24, 2007 11:17 AM
> To: rsyslog-users
> Subject: Re: [rsyslog] rsyslog 1.19.6 and segfaults
> 
> Thanks for the report. I'd not yet use the single-threading version.
> The
> performance hit should not be severe, but I think I'll be able to
> release another version today. Not sure, of course, if it actually
> fixes
> the bug, but I am working on it. I got a report that it is indeed
> related to a threading issue and I am now checking all runtime
> functions
> that get called. At least one I have found to be not thread-safe. I
> intend to release a version with that fix and a couple of minor
things,
> even if I have not yet completed the full runtime-checks.
> 
> Rainer
> 
> > -----Original Message-----
> > From: [EMAIL PROTECTED] [mailto:rsyslog-
> > [EMAIL PROTECTED] On Behalf Of Jan-Frode Myklebust
> > Sent: Sunday, September 23, 2007 11:28 AM
> > To: [email protected]
> > Subject: Re: [rsyslog] rsyslog 1.19.6 and segfaults
> >
> > On 2007-09-11, Rainer Gerhards <[EMAIL PROTECTED]> wrote:
> > >
> > > But one important thing: I would appreciate if those of you that
> > > experienced the segfaults could try out the new version. I
> currently
> > > think it will *NOT* fix the problem, but I would like to verify
> that
> > > (and I have to admit I am not angry if I am wrong...).
> >
> > Sorry, you were right:
> >
> > *** glibc detected *** rsyslogd: corrupted double-linked list:
> > 0xb7214c98 ***
> > ======= Backtrace: =========
> > /lib/libc.so.6[0x4152ce3e]
> > /lib/libc.so.6(cfree+0x90)[0x415305d0]
> > rsyslogd(MsgDestruct+0x46)[0x80580b6]
> > rsyslogd[0x804df9a]
> > rsyslogd(llExecFunc+0x3f)[0x805f11f]
> > rsyslogd[0x804d9ea]
> > rsyslogd[0x804db3f]
> > /lib/libpthread.so.0[0x416112db]
> > /lib/libc.so.6(clone+0x5e)[0x4159414e]
> >
> > With 1.19.6.
> >
> > > If it fails, it would be even greater if you could try it in
> > > single-threaded mode:
> > >
> > > ./configure --disable-pthreads
> >
> > Is this safe to use on a busy system? I wouldn't want to start
> dropping
> > lots of packets..
> >
> >
> >   -jf
> >
> > _______________________________________________
> > rsyslog mailing list
> > http://lists.adiscon.net/mailman/listinfo/rsyslog
> _______________________________________________
> rsyslog mailing list
> http://lists.adiscon.net/mailman/listinfo/rsyslog
_______________________________________________
rsyslog mailing list
http://lists.adiscon.net/mailman/listinfo/rsyslog

Reply via email to