This is a note to let you know that I've just added the patch titled
libceph: fix messenger retry
to the 3.4-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary
The filename of the patch is:
0032-libceph-fix-messenger-retry.patch
and it can be found in the queue-3.4 subdirectory.
If you, or anyone else, feels it should not be added to the stable tree,
please let <[email protected]> know about it.
>From 97e7f3ee21ecef94c8cf6914b6c549d683e83739 Mon Sep 17 00:00:00 2001
From: Sage Weil <[email protected]>
Date: Tue, 10 Jul 2012 11:53:34 -0700
Subject: libceph: fix messenger retry
From: Sage Weil <[email protected]>
(cherry picked from commit 5bdca4e0768d3e0f4efa43d9a2cc8210aeb91ab9)
In ancient times, the messenger could both initiate and accept connections.
An artifact if that was data structures to store/process an incoming
ceph_msg_connect request and send an outgoing ceph_msg_connect_reply.
Sadly, the negotiation code was referencing those structures and ignoring
important information (like the peer's connect_seq) from the correct ones.
Among other things, this fixes tight reconnect loops where the server sends
RETRY_SESSION and we (the client) retries with the same connect_seq as last
time. This bug pretty easily triggered by injecting socket failures on the
MDS and running some fs workload like workunits/direct_io/test_sync_io.
Signed-off-by: Sage Weil <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
---
include/linux/ceph/messenger.h | 12 ++----------
net/ceph/messenger.c | 12 ++++++------
2 files changed, 8 insertions(+), 16 deletions(-)
--- a/include/linux/ceph/messenger.h
+++ b/include/linux/ceph/messenger.h
@@ -163,16 +163,8 @@ struct ceph_connection {
/* connection negotiation temps */
char in_banner[CEPH_BANNER_MAX_LEN];
- union {
- struct { /* outgoing connection */
- struct ceph_msg_connect out_connect;
- struct ceph_msg_connect_reply in_reply;
- };
- struct { /* incoming */
- struct ceph_msg_connect in_connect;
- struct ceph_msg_connect_reply out_reply;
- };
- };
+ struct ceph_msg_connect out_connect;
+ struct ceph_msg_connect_reply in_reply;
struct ceph_entity_addr actual_peer_addr;
/* message out temps */
--- a/net/ceph/messenger.c
+++ b/net/ceph/messenger.c
@@ -1423,7 +1423,7 @@ static int process_connect(struct ceph_c
* dropped messages.
*/
dout("process_connect got RESET peer seq %u\n",
- le32_to_cpu(con->in_connect.connect_seq));
+ le32_to_cpu(con->in_reply.connect_seq));
pr_err("%s%lld %s connection reset\n",
ENTITY_NAME(con->peer_name),
ceph_pr_addr(&con->peer_addr.in_addr));
@@ -1450,10 +1450,10 @@ static int process_connect(struct ceph_c
* If we sent a smaller connect_seq than the peer has, try
* again with a larger value.
*/
- dout("process_connect got RETRY my seq = %u, peer_seq = %u\n",
+ dout("process_connect got RETRY_SESSION my seq %u, peer %u\n",
le32_to_cpu(con->out_connect.connect_seq),
- le32_to_cpu(con->in_connect.connect_seq));
- con->connect_seq = le32_to_cpu(con->in_connect.connect_seq);
+ le32_to_cpu(con->in_reply.connect_seq));
+ con->connect_seq = le32_to_cpu(con->in_reply.connect_seq);
ceph_con_out_kvec_reset(con);
ret = prepare_write_connect(con);
if (ret < 0)
@@ -1468,9 +1468,9 @@ static int process_connect(struct ceph_c
*/
dout("process_connect got RETRY_GLOBAL my %u peer_gseq %u\n",
con->peer_global_seq,
- le32_to_cpu(con->in_connect.global_seq));
+ le32_to_cpu(con->in_reply.global_seq));
get_global_seq(con->msgr,
- le32_to_cpu(con->in_connect.global_seq));
+ le32_to_cpu(con->in_reply.global_seq));
ceph_con_out_kvec_reset(con);
ret = prepare_write_connect(con);
if (ret < 0)
Patches currently in stable-queue which might be from [email protected] are
queue-3.4/0020-ceph-ensure-auth-ops-are-defined-before-use.patch
queue-3.4/0025-ceph-add-auth-buf-in-prepare_write_connect.patch
queue-3.4/0021-ceph-have-get_authorizer-methods-return-pointers.patch
queue-3.4/0026-libceph-avoid-unregistering-osd-request-when-not-reg.patch
queue-3.4/0005-crush-fix-memory-leak-when-destroying-tree-buckets.patch
queue-3.4/0002-crush-adjust-local-retry-threshold.patch
queue-3.4/0023-ceph-return-pointer-from-prepare_connect_authorizer.patch
queue-3.4/0018-ceph-define-ceph_auth_handshake-type.patch
queue-3.4/0011-ceph-messenger-reset-connection-kvec-caller.patch
queue-3.4/0032-libceph-fix-messenger-retry.patch
queue-3.4/0019-ceph-messenger-reduce-args-to-create_authorizer.patch
queue-3.4/0029-libceph-use-con-get-put-ops-from-osd_client.patch
queue-3.4/0015-ceph-messenger-check-prepare_write_connect-result.patch
queue-3.4/0003-crush-be-more-tolerant-of-nonsensical-crush-maps.patch
queue-3.4/0028-libceph-osd_client-don-t-drop-reply-reference-too-ea.patch
queue-3.4/0014-ceph-don-t-set-WRITE_PENDING-too-early.patch
queue-3.4/0016-ceph-messenger-rework-prepare_connect_authorizer.patch
queue-3.4/0031-libceph-flush-msgr-queue-during-mon_client-shutdown.patch
queue-3.4/0013-ceph-drop-msgr-argument-from-prepare_write_connect.patch
queue-3.4/0009-ceph-messenger-change-read_partial-to-take-end-arg.patch
queue-3.4/0017-ceph-messenger-check-return-from-get_authorizer.patch
queue-3.4/0001-crush-clean-up-types-const-ness.patch
queue-3.4/0037-ceph-check-PG_Private-flag-before-accessing-page-pri.patch
queue-3.4/0022-ceph-use-info-returned-by-get_authorizer.patch
queue-3.4/0027-libceph-fix-pg_temp-updates.patch
queue-3.4/0008-ceph-messenger-update-to-in-read_partial-caller.patch
queue-3.4/0007-ceph-messenger-use-read_partial-in-read_partial_mess.patch
queue-3.4/0010-libceph-don-t-reset-kvec-in-prepare_write_banner.patch
queue-3.4/0012-ceph-messenger-send-banner-in-process_connect.patch
queue-3.4/0024-ceph-rename-prepare_connect_authorizer.patch
queue-3.4/0004-crush-fix-tree-node-weight-lookup.patch
--
To unsubscribe from this list: send the line "unsubscribe stable" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
