summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorVolker Lendecke <vl@samba.org>2008-03-03 18:12:26 +0100
committerKarolin Seeger <ks@sernet.de>2008-03-04 08:55:25 +0100
commita94313e06a030e1a1afff18c797e700eb6b25248 (patch)
treee3b6bea077fb0bba82ee3ae570ff8b2c9820a13d
parent4170259e5f83c281966a80916e4ece6937e08988 (diff)
downloadsamba-a94313e06a030e1a1afff18c797e700eb6b25248.tar.gz
samba-a94313e06a030e1a1afff18c797e700eb6b25248.tar.xz
samba-a94313e06a030e1a1afff18c797e700eb6b25248.zip
Zero out the out policy handler in lsa_Close
... after a REALLY long session staring at sniffs we can now join XP to v3-2-test again... Apparently not doing this makes XP keep an internal handle to LSA open which confuses the hell out of it. Karolin, this needs to be in v3-2-stable :-) Volker (cherry picked from commit 2c42fc21d8bede226e411623aecd69038477373b)
-rw-r--r--source/rpc_server/srv_lsa_nt.c1
1 files changed, 1 insertions, 0 deletions
diff --git a/source/rpc_server/srv_lsa_nt.c b/source/rpc_server/srv_lsa_nt.c
index 1333d656d48..ce3fd1a44fa 100644
--- a/source/rpc_server/srv_lsa_nt.c
+++ b/source/rpc_server/srv_lsa_nt.c
@@ -1216,6 +1216,7 @@ NTSTATUS _lsa_Close(pipes_struct *p, struct lsa_Close *r)
}
close_policy_hnd(p, r->in.handle);
+ ZERO_STRUCTP(r->out.handle);
return NT_STATUS_OK;
}