[Date Prev][Date Next] [Chronological] [Thread] [Top]

Re: core dump! help!

At 02:01 PM 2001-10-20, oberwetter, josh wrote:
>We've installed OpenLDAP on our new production machines, and one just core
>dumped. I wasn't doing much at all to the server -- a small sample data set
>(fewer than 30 records), infrequent lookups, & even less frequent writes.
>Hopefully somebody can shed some light on why this happened & how to prevent
>it in the future.
>We're running the following:
>Solaris 7
>OpenLDAP 2.0.15
>BerkeleyDB 3.2 (Sleepy Cat)
>OpenSSL 0.9.6b
>We haven't done much configuration such as tuning memory management yet.
>Unfortunately I had logging turned off (darn), and now I'm unable to
>reproduce the problem (the data files don't seem to be corrupted), so I
>can't say exactly what went wrong. I've turned logging on again, and we're
>doing stress testing over the next few days, so I might have more info
>later. All I can offer is the beginning of the core file obtained via 
>strings core | more
>If anybody can read this & finds something telling, would you let me & the
>rest of the group know?

Likely no.   A stack back trace would likely be more useful.

% gdb slapd core
> bt