From: Haomai Wang <haomaiwang@gmail.com>
Date: Wed, 14 Jan 2015 14:51:58 +0000 (+0800)
Subject: AsyncConnection: Don't increment connect_seq if connect failed
X-Git-Tag: v0.93~247^2~3
X-Git-Url: http://git-server-git.apps.pok.os.sepia.ceph.com/?a=commitdiff_plain;h=f7f25b4cbb148ba157fc84d9c37d45baefd0156a;p=ceph.git

AsyncConnection: Don't increment connect_seq if connect failed

If connection sent many messages without acked, then it was marked down.
Next we get a new connection, it will issue a connect_msg with connect_seq=0,
server side need to detect "connect_seq==0 && existing->connect_seq >0",
so it will reset out_q and detect remote reset. But if client side failed
before sending connect_msg, now it will issue a connect_msg with non-zero
connect_seq which will cause server-side can't detect exist remote reset.
Server-side will reply a non-zero in_seq and cause client crash.

Signed-off-by: Haomai Wang <haomaiwang@gmail.com>
---

diff --git a/src/msg/async/AsyncConnection.cc b/src/msg/async/AsyncConnection.cc
index 556704ec8e4d..16830e360605 100644
--- a/src/msg/async/AsyncConnection.cc
+++ b/src/msg/async/AsyncConnection.cc
@@ -1998,7 +1998,7 @@ void AsyncConnection::fault()
     return;
   }
 
-  if (state != STATE_CONNECTING) {
+  if (!(state >= STATE_CONNECTING && state < STATE_CONNECTING_READY)) {
     // policy maybe empty when state is in accept
     if (policy.server || (state >= STATE_ACCEPTING && state < STATE_ACCEPTING_WAIT_SEQ)) {
       ldout(async_msgr->cct, 0) << __func__ << " server, going to standby" << dendl;
@@ -2017,6 +2017,7 @@ void AsyncConnection::fault()
       if (backoff > async_msgr->cct->_conf->ms_max_backoff)
         backoff.set_from_double(async_msgr->cct->_conf->ms_max_backoff);
     }
+    state = STATE_CONNECTING;
     ldout(async_msgr->cct, 10) << __func__ << " waiting " << backoff << dendl;
   }