Re: (ITS#4476) connection deadlock

[ando@sys-net.it]

On Wed, 2006-04-05 at 19:17 -0700, Howard Chu wrote:

> Right. It looks like all the threads are busy, and the connection_write 
> handler that would have signaled the condition is queued in the thread pool.
> 
> I think we need to bring connection_write handling back inline instead 
> of running it in a separate thread.

After your fix the problem didn't surface any more.  Both the proxy the
remote servers remain responsive even when overloaded.  I must say I had
to apply an innatural load to the proxy to cause the issue.  I didn't
try running a similar load to slapd itself, so I haven't any idea if the
proxy played an essential role or not.

Note that when limiting the number of clients to a level that didn't
cause the issue, I seem to get slightly better search rates (just a
feeling...).

p.




Ing. Pierangelo Masarati
Responsabile Open Solution
OpenLDAP Core Team

SysNet s.n.c.
Via Dossi, 8 - 27100 Pavia - ITALIA
http://www.sys-net.it
------------------------------------------
Office:   +39.02.23998309          
Mobile:   +39.333.4963172
Email:    pierangelo.masarati@sys-net.it
------------------------------------------