svn commit: samba r11868 - in branches/SAMBA_4_0/source/torture/rpc: .

[tridge at samba.org]

Volker,

 > As such, the best way to avoid this is to re-use as much of the code as
 > possible.  In particular, it would be great if we could (optionally) use
 > a normal join from 'net join', or a password file as the credentials
 > system normally uses it (see auth/credentials/credentials_files.c).  

Echoing Andrews comments a bit, I found it a bit tricky to work out
how to use the RPC-XPLOGIN test when trying to reproduce those talloc
problems.

One way to make it easier would be to make the use of the wks.pwd and
user.pwd files optional. So by default it would use the same
credentials as all other tests use (and do an automatic join), but you
could optionally have a way of specifying those files for more
advanced testing.

This might be similar to the existing torture:unclist option which is
used to provide a list of servers to run BENCH-NBENCH
against. BENCH-NBENCH works fine without it, but by adding it you can
hit a bunch of servers in parallel.

Then hopefully we could enable the RPC-XPLOGIN test in the build farm
and have it regularly tested.

my apologies for stating the obvious if you already have plans to do
something like the above, which I suspect you do :-)

Cheers, Tridge