Re: Please test RE24

[Quanah Gibson-Mount <quanah@zimbra.com>]

--On Tuesday, March 10, 2009 8:54 PM +0100 Pierangelo Masarati 
<ando@sys-net.it> wrote:

> Michael StrÃder wrote:
> > Quanah Gibson-Mount wrote:
> > > --On Tuesday, March 10, 2009 9:48 AM +0100 Jens Vagelpohl
> > > <jens@dataflake.org> wrote:
> > >
> > > > On Mar 10, 2009, at 00:21 , Quanah Gibson-Mount wrote:
> > > >
> > > > > Please test RE24 as it is being considered for 2.4.16.  Thanks!
> > > >
> > > > On OS X 10.5.6/Intel test045 fails:
> > > This should be fixed now, please test further. :)
> >
> > This test still fails:
>
> Same here.  If I diff the output I get:

test050 is also having interesting failures for me.  I've had one segfault 
(corrupt CSN data), one lockup (waiting on unlocked mutexes to clear) and 
twice I've had server3 be empty:

-rw------- 1 quanah quanah  19849 Mar 10 14:31 server4.out
-rw------- 1 quanah quanah  19849 Mar 10 14:31 server4.flt
-rw------- 1 quanah quanah    377 Mar 10 14:31 server3.out
-rw------- 1 quanah quanah    377 Mar 10 14:31 server3.flt
-rw------- 1 quanah quanah  19849 Mar 10 14:31 server2.out
-rw------- 1 quanah quanah  19849 Mar 10 14:31 server2.flt
-rw------- 1 quanah quanah  19849 Mar 10 14:31 server1.out
-rw------- 1 quanah quanah  19849 Mar 10 14:31 server1.flt


so the comparison fails.  This is after adjusting SLEEP1 to be 20 seconds 
and SLEEP2 to be 30 seconds, which should be more than sufficient.

--Quanah

--

Quanah Gibson-Mount
Principal Software Engineer
Zimbra, Inc
--------------------
Zimbra ::  the leader in open source messaging and collaboration