From: Markus Armbruster <armbru@pond.sub.org>
Date: Thu, 31 Dec 2009 11:12:39 +0000 (+0100)
Subject: Fix new_seqno()'s recovery from seqno mismatch
X-Git-Tag: v4.3.24~45
X-Git-Url: http://git.pond.sub.org/?p=empserver;a=commitdiff_plain;h=18148f03fa96cad1dbca2e9a087670153a6b19dd

Fix new_seqno()'s recovery from seqno mismatch

A sequence number mismatch occurs when an invalid copy (with a stale
sequence number) gets written.  To continue after the error, we need
to use the current sequence number.  But new_seqno() used the larger
one of the two, which is correct only as long as sequence numbers do
not overflow.  Overflow is rather unlikely, though.
---

diff --git a/src/lib/common/file.c b/src/lib/common/file.c
index af02cf184..f32bafa4b 100644
--- a/src/lib/common/file.c
+++ b/src/lib/common/file.c
@@ -603,8 +603,7 @@ new_seqno(struct empfile *ep, void *buf)
     if (!(ep->flags & EFF_TYPED))
 	return;
     old_seqno = get_seqno(ep, elt->uid);
-    if (CANT_HAPPEN(old_seqno != elt->seqno))
-	old_seqno = MAX(old_seqno, elt->seqno);
+    CANT_HAPPEN(old_seqno != elt->seqno);
     elt->seqno = old_seqno + 1;
 }