linux-stable/net/ceph
Ilya Dryomov 51f6afddb1 libceph: fix PG split vs OSD (re)connect race
commit a561372405 upstream.

We can't rely on ->peer_features in calc_target() because it may be
called both when the OSD session is established and open and when it's
not.  ->peer_features is not valid unless the OSD session is open.  If
this happens on a PG split (pg_num increase), that could mean we don't
resend a request that should have been resent, hanging the client
indefinitely.

In userspace this was fixed by looking at require_osd_release and
get_xinfo[osd].features fields of the osdmap.  However these fields
belong to the OSD section of the osdmap, which the kernel doesn't
decode (only the client section is decoded).

Instead, let's drop this feature check.  It effectively checks for
luminous, so only pre-luminous OSDs would be affected in that on a PG
split the kernel might resend a request that should not have been
resent.  Duplicates can occur in other scenarios, so both sides should
already be prepared for them: see dup/replay logic on the OSD side and
retry_attempt check on the client side.

Cc: stable@vger.kernel.org
Fixes: 7de030d6b1 ("libceph: resend on PG splits if OSD has RESEND_ON_SPLIT")
Link: https://tracker.ceph.com/issues/41162
Reported-by: Jerry Lee <leisurelysw24@gmail.com>
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Tested-by: Jerry Lee <leisurelysw24@gmail.com>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2019-08-29 08:28:50 +02:00
..
crush crush: fix using plain integer as NULL warning 2018-08-13 17:55:44 +02:00
armor.c
auth.c libceph: add authorizer challenge 2018-08-02 21:33:24 +02:00
auth_none.c ceph: fix whitespace 2018-08-02 21:33:21 +02:00
auth_none.h ceph: fix whitespace 2018-08-02 21:33:21 +02:00
auth_x.c libceph: weaken sizeof check in ceph_x_verify_authorizer_reply() 2018-08-02 21:33:26 +02:00
auth_x.h ceph: fix whitespace 2018-08-02 21:33:21 +02:00
auth_x_protocol.h libceph: add authorizer challenge 2018-08-02 21:33:24 +02:00
buffer.c
ceph_common.c libceph: wait for latest osdmap in ceph_monc_blacklist_add() 2019-03-27 14:14:39 +09:00
ceph_fs.c
ceph_hash.c
ceph_strings.c
cls_lock_client.c libceph: use timespec64 in for keepalive2 and ticket validity 2018-08-02 21:26:12 +02:00
crypto.c libceph, ceph: add __init attribution to init funcitons 2018-04-02 10:12:48 +02:00
crypto.h
debugfs.c libceph, ceph: add __init attribution to init funcitons 2018-04-02 10:12:48 +02:00
Kconfig ceph: fix whitespace 2018-08-02 21:33:21 +02:00
Makefile ceph: fix whitespace 2018-08-02 21:33:21 +02:00
messenger.c libceph: handle an empty authorize reply 2019-02-27 10:08:50 +01:00
mon_client.c libceph: wait for latest osdmap in ceph_monc_blacklist_add() 2019-03-27 14:14:39 +09:00
msgpool.c
osd_client.c libceph: fix PG split vs OSD (re)connect race 2019-08-29 08:28:50 +02:00
osdmap.c The main piece is a set of libceph changes that revamps how OSD 2018-06-15 07:24:58 +09:00
pagelist.c
pagevec.c ceph: fix whitespace 2018-08-02 21:33:21 +02:00
snapshot.c
string_table.c
striper.c