Re: slapd problems under stress

[Mark Valence <kurash@sassafras.com>]

Christian-

> I have integrated two backends into slapd and we are now
> observing problems with slapd if it got under stress:
>
> Our test program (written by using libldap) starts 80 async
> request and waits for their results. This is done every
> 10 seconds. In case of a time-out (10 seconds) all pending
> operations will be abandoned. Sometimes the slapd crashes
> with a SEGV, sometimes it hangs, sometimes it aborts (from
> assert() during a memory allocation)!

Can you send the source for the test program?  Does the problem 
happen after replying to a bunch of search requests, or does it 
happen as soon as you start the test program?  Do the three cases 
(SEGV, hang, abort) happen with equal likelihood, or is one case more 
likely than the others?

> After that, we concentrated on a single backend which I
> stripped down so that the search does nothing but a
> 'send_ldap_result(conn,op,LDAP_NO_SUCH_OBJECT,NULL,NULL,NULL,0);'.
> Same problems!
>
> The next step was to use the 'standard' ldbm backend.
> And again: we have the same problems!
>
> We are running linux 2.2.14, OpenLDAP 2.0-devel (last
> update from the CVS: 28 June 2000), debug level for all
> tests was 255.

Can you try a test?  In libraries/libldap_r/tpool.c, add the 
following line after line 161:

  return ldap_pvt_thread_create( &thr, 1, start_routine, arg );

Then rebuild and try running your tests and see if the problems still happen.

> I think that the problems are thread related.
> Are there any known problems? Is it possible to set a limit
> on the number of concurrent threads?

You can change line 97 in servers/slapd/init.c to something like:

  ldap_pvt_thread_pool_init(&connection_pool, 5, 0);

where the second parameter is the maximum number of concurrent threads.

Mark.