KDC tests failing since this week
Andreas Schneider
asn at samba.org
Thu Jun 16 11:51:27 UTC 2016
On Thursday, 16 June 2016 10:44:39 CEST Andrew Bartlett wrote:
> On Wed, 2016-06-15 at 15:15 +0200, Andreas Schneider wrote:
> > Hi Andrew,
> >
> > I'm working on refactoring the KDC code. Last week
> >
> > make -j test TESTS="samba4.krb5.kdc"
> >
> > worked just fine for me, this week if I run it the fl2008r2dc tests
> > runs into
> > a timeout. The changes which went into master are mostly your code.
> >
> > auth_check_password_recv: sam_ignoredomain authentication for user
> > [SAMBA2008R2\Administrator] succeeded
> > NTLMSSP Sign/Seal - Initialising with flags:
> > Got NTLMSSP neg_flags=0x62088235
> > NTLMSSP_NEGOTIATE_UNICODE
> > NTLMSSP_REQUEST_TARGET
> > NTLMSSP_NEGOTIATE_SIGN
> > NTLMSSP_NEGOTIATE_SEAL
> > NTLMSSP_NEGOTIATE_NTLM
> > NTLMSSP_NEGOTIATE_ALWAYS_SIGN
> > NTLMSSP_NEGOTIATE_EXTENDED_SESSIONSECURITY
> > NTLMSSP_NEGOTIATE_VERSION
> > NTLMSSP_NEGOTIATE_128
> > NTLMSSP_NEGOTIATE_KEY_EXCH
> > Terminating connection - 'ldapsrv_call_loop:
> > tstream_read_pdu_blob_recv() -
> > NT_STATUS_INVALID_BUFFER_SIZE'
> > single_terminate: reason[ldapsrv_call_loop:
> > tstream_read_pdu_blob_recv() -
> > NT_STATUS_INVALID_BUFFER_SIZE]
> > Timed out (60 sec) waiting for working LDAP and a RID Set to be
> > allocated by
> > DC7 PID 32142 at
> > /home/asn/workspace/projects/samba/selftest/target/Samba4.pm
> > line 222.
> > Samba 32142 failed to start up at
> > /home/asn/workspace/projects/samba/selftest/
> > target/Samba4.pm line 165.
> > failed to start up environment 'fl2008r2dc' at
> > /home/asn/workspace/projects/
> > samba/selftest/target/Samba.pm line 49.
> > samba can't start up known environment 'fl2008r2dc' at
> > /home/asn/workspace/
> > projects/samba/selftest/selftest.pl line 898
> >
> >
> > This is what I get. I'm not sure what is really going wrong. Help
> > would be
> > much appreciated. I will continue digging ...
>
> G'Day,
Hi Andrew,
I played around a bit.
in source4/selftest/tests.py line 696
for env in ["rodc", "promoted_dc", "ad_dc", "fl2000dc", "fl2008r2dc"]:
which sets the targets for the test. If I remove the fl2000dc target, the
fl2008r2dc works just fine.
If I change the line to:
for env in ["rodc", "promoted_dc", "fl2000dc", "ad_dc"]:
sefltest fails to provision the ad_dc target.
So the fl2000dc taints the environment!
I need to investigate further but the questions which comes to my mind is: Why
do we still support fl2000?
Cheers,
-- andreas
--
Andreas Schneider GPG-ID: CC014E3D
Samba Team asn at samba.org
www.samba.org
More information about the samba-technical
mailing list