Re: How do I get a good gdb trave of slapd?

[Aaron Richton <richton@nbcs.rutgers.edu>]

> and not through pam. Even slaptest segfaults after saying
Does it say NULL, or did you miss an error message here?

> Therefore: How should I use gdb to get a good trace? My problem is that
Try "info threads", "thread <#>", and "thread apply all" (as in "thread
apply all bt").

> If this is a known issue with this version (that I know is old) then
> please say so, but if not I'd rather get it working than have to build
> my own rpms from time to time.
See

http://www.openldap.org/devel/cvsweb.cgi/~checkout~/Attic/CHANGES?rev=1.5.6.237&hideattic=1&sortbydate=0

for the 2.2 changelog. Four fixes for "SASL" since 2.2.15. Nine fixes for
"crash" since 2.2.15. If you're going to take the time to ponder the
output of gdb, wouldn't you feel better knowing you weren't duplicating
effort by running against 2.2.26/2.2.27? While you're at it, make sure
your version of cyrus-sasl is recent, as well.

> >From slaptest I got this stacktrace:

Personally, I'd be interested in starting with -d -1 output from a recent
version sooner than a backtrace.