linux-stable.git/net/ceph, branch v3.18.76

libceph: don't set weight to IN when OSD is destroyed

2017-04-18T05:55:54+00:00

commit b581a5854eee4b7851dedb0f8c2ceb54fb902c06 upstream.

Since ceph.git commit 4e28f9e63644 ("osd/OSDMap: clear osd_info,
osd_xinfo on osd deletion"), weight is set to IN when OSD is deleted.
This changes the result of applying an incremental for clients, not
just OSDs.  Because CRUSH computations are obviously affected,
pre-4e28f9e63644 servers disagree with post-4e28f9e63644 clients on
object placement, resulting in misdirected requests.

Mirrors ceph.git commit a6009d1039a55e2c77f431662b3d6cc5a8e8e63f.

Fixes: 930c53286977 ("libceph: apply new_state before new_up_client on incrementals")
Link: http://tracker.ceph.com/issues/19122
Signed-off-by: Ilya Dryomov 
Reviewed-by: Sage Weil 
Signed-off-by: Greg Kroah-Hartman

libceph: verify authorize reply on connect

2017-01-15T14:49:52+00:00

[ Upstream commit 5c056fdc5b474329037f2aa18401bd73033e0ce0 ]

After sending an authorizer (ceph_x_authorize_a + ceph_x_authorize_b),
the client gets back a ceph_x_authorize_reply, which it is supposed to
verify to ensure the authenticity and protect against replay attacks.
The code for doing this is there (ceph_x_verify_authorizer_reply(),
ceph_auth_verify_authorizer_reply() + plumbing), but it is never
invoked by the the messenger.

AFAICT this goes back to 2009, when ceph authentication protocols
support was added to the kernel client in 4e7a5dcd1bba ("ceph:
negotiate authentication protocol; implement AUTH_NONE protocol").

The second param of ceph_connection_operations::verify_authorizer_reply
is unused all the way down.  Pass 0 to facilitate backporting, and kill
it in the next commit.

Cc: stable@vger.kernel.org
Signed-off-by: Ilya Dryomov 
Reviewed-by: Sage Weil 
Signed-off-by: Sasha Levin

libceph: apply new_state before new_up_client on incrementals

2016-08-08T01:59:11+00:00

[ Upstream commit 930c532869774ebf8af9efe9484c597f896a7d46 ]

Currently, osd_weight and osd_state fields are updated in the encoding
order.  This is wrong, because an incremental map may look like e.g.

    new_up_client: { osd=6, addr=... } # set osd_state and addr
    new_state: { osd=6, xorstate=EXISTS } # clear osd_state

Suppose osd6's current osd_state is EXISTS (i.e. osd6 is down).  After
applying new_up_client, osd_state is changed to EXISTS | UP.  Carrying
on with the new_state update, we flip EXISTS and leave osd6 in a weird
"!EXISTS but UP" state.  A non-existent OSD is considered down by the
mapping code

2087    for (i = 0; i < pg->pg_temp.len; i++) {
2088            if (ceph_osd_is_down(osdmap, pg->pg_temp.osds[i])) {
2089                    if (ceph_can_shift_osds(pi))
2090                            continue;
2091
2092                    temp->osds[temp->size++] = CRUSH_ITEM_NONE;

and so requests get directed to the second OSD in the set instead of
the first, resulting in OSD-side errors like:

[WRN] : client.4239 192.168.122.21:0/2444980242 misdirected client.4239.1:2827 pg 2.5df899f2 to osd.4 not [1,4,6] in e680/680

and hung rbds on the client:

[  493.566367] rbd: rbd0: write 400000 at 11cc00000 (0)
[  493.566805] rbd: rbd0:   result -6 xferred 400000
[  493.567011] blk_update_request: I/O error, dev rbd0, sector 9330688

The fix is to decouple application from the decoding and:
- apply new_weight first
- apply new_state before new_up_client
- twiddle osd_state flags if marking in
- clear out some of the state if osd is destroyed

Fixes: http://tracker.ceph.com/issues/14901

Cc: stable@vger.kernel.org # 3.15+: 6dd74e44dc1d: libceph: set 'exists' flag for newly up osd
Cc: stable@vger.kernel.org # 3.15+
Signed-off-by: Ilya Dryomov 
Reviewed-by: Josh Durgin 
Signed-off-by: Sasha Levin

libceph: set 'exists' flag for newly up osd

2016-08-08T01:59:06+00:00

[ Upstream commit 6dd74e44dc1df85f125982a8d6591bc4a76c9f5d ]

Signed-off-by: Yan, Zheng 
Reviewed-by: Sage Weil 
Signed-off-by: Ilya Dryomov 
Signed-off-by: Sasha Levin

libceph: don't bail early from try_read() when skipping a message

2016-03-10T14:43:52+00:00

[ Upstream commit e7a88e82fe380459b864e05b372638aeacb0f52d ]

The contract between try_read() and try_write() is that when called
each processes as much data as possible.  When instructed by osd_client
to skip a message, try_read() is violating this contract by returning
after receiving and discarding a single message instead of checking for
more.  try_write() then gets a chance to write out more requests,
generating more replies/skips for try_read() to handle, forcing the
messenger into a starvation loop.

Cc: stable@vger.kernel.org # 3.10+
Reported-by: Varada Kari 
Signed-off-by: Ilya Dryomov 
Tested-by: Varada Kari 
Reviewed-by: Alex Elder 
Signed-off-by: Sasha Levin

crush: ensuring at most num-rep osds are selected

2015-06-15T18:29:41+00:00

[ Upstream commit 45002267e8d2699bf9b022315bee3dd13b044843 ]

Crush temporary buffers are allocated as per replica size configured
by the user.  When there are more final osds (to be selected as per
rule) than the replicas, buffer overlaps and it causes crash.  Now, it
ensures that at most num-rep osds are selected even if more number of
osds are allowed by the rule.

Reflects ceph.git commits 6b4d1aa99718e3b367496326c1e64551330fabc0,
                          234b066ba04976783d15ff2abc3e81b6cc06fb10.

Signed-off-by: Ilya Dryomov 
Signed-off-by: Sasha Levin

Revert "libceph: clear r_req_lru_item in __unregister_linger_request()"

2015-06-09T17:43:54+00:00

[ Upstream commit 521a04d06a729e5971cdee7f84080387ed320527 ]

This reverts commit ba9d114ec5578e6e99a4dfa37ff8ae688040fd64.

.. which introduced a regression that prevented all lingering requests
requeued in kick_requests() from ever being sent to the OSDs, resulting
in a lot of missed notifies.  In retrospect it's pretty obvious that
r_req_lru_item item in the case of lingering requests can be used not
only for notarget, but also for unsent linkage due to how tightly
actual map and enqueue operations are coupled in __map_request().

The assertion that was being silenced is taken care of in the previous
("libceph: request a new osdmap if lingering request maps to no osd")
commit: by always kicking homeless lingering requests we ensure that
none of them ends up on the notarget list outside of the critical
section guarded by request_mutex.

Cc: stable@vger.kernel.org # 3.18+, needs b0494532214b "libceph: request a new osdmap if lingering request maps to no osd"
Signed-off-by: Ilya Dryomov 
Reviewed-by: Sage Weil 
Signed-off-by: Sasha Levin

Revert "libceph: clear r_req_lru_item in __unregister_linger_request()"

2015-06-09T17:43:53+00:00

[ Upstream commit 521a04d06a729e5971cdee7f84080387ed320527 ]

This reverts commit ba9d114ec5578e6e99a4dfa37ff8ae688040fd64.

.. which introduced a regression that prevented all lingering requests
requeued in kick_requests() from ever being sent to the OSDs, resulting
in a lot of missed notifies.  In retrospect it's pretty obvious that
r_req_lru_item item in the case of lingering requests can be used not
only for notarget, but also for unsent linkage due to how tightly
actual map and enqueue operations are coupled in __map_request().

The assertion that was being silenced is taken care of in the previous
("libceph: request a new osdmap if lingering request maps to no osd")
commit: by always kicking homeless lingering requests we ensure that
none of them ends up on the notarget list outside of the critical
section guarded by request_mutex.

Cc: stable@vger.kernel.org # 3.18+, needs b0494532214b "libceph: request a new osdmap if lingering request maps to no osd"
Signed-off-by: Ilya Dryomov 
Reviewed-by: Sage Weil 
Signed-off-by: Sasha Levin

Revert "libceph: use memalloc flags for net IO"

2015-04-17T00:13:14+00:00

[ Upstream commit 6d7fdb0ab351b33d4c12d53fe44be030b90fc9d4 ]

This reverts commit 89baaa570ab0b476db09408d209578cfed700e9f.

Dirty page throttling should be sufficient for us in the general case
so there is no need to use __GFP_MEMALLOC - it would be needed only in
the swap-over-rbd case, which we currently don't support.  (It would
probably take approximately the commit that is being reverted to add
that support, but we would also need the "swap" option to distinguish
from the general case and make sure swap ceph_client-s aren't shared
with anything else.)  See ceph-devel threads [1] and [2] for the
details of why enabling pfmemalloc reserves for all cases is a bad
thing.

On top of potential system lockups related to drained emergency
reserves, this turned out to cause ceph lockups in case peers are on
the same host and communicating via loopback due to sk_filter()
dropping pfmemalloc skbs on the receiving side because the receiving
loopback socket is not tagged with SOCK_MEMALLOC.

[1] "SOCK_MEMALLOC vs loopback"
    http://www.spinics.net/lists/ceph-devel/msg22998.html
[2] "[PATCH] libceph: don't set memalloc flags in loopback case"
    http://www.spinics.net/lists/ceph-devel/msg23392.html

Conflicts:
	net/ceph/messenger.c [ context: tcp_nodelay option ]

Cc: Mike Christie 
Cc: Mel Gorman 
Cc: Sage Weil 
Cc: stable@vger.kernel.org # 3.18+, needs backporting
Signed-off-by: Ilya Dryomov 
Acked-by: Mike Christie 
Acked-by: Mel Gorman 
[idryomov@gmail.com: backport to 3.18, 3.19: context]
Signed-off-by: Sasha Levin

libceph: fix double __remove_osd() problem

2015-03-06T22:53:05+00:00

commit 7eb71e0351fbb1b242ae70abb7bb17107fe2f792 upstream.

It turns out it's possible to get __remove_osd() called twice on the
same OSD.  That doesn't sit well with rb_erase() - depending on the
shape of the tree we can get a NULL dereference, a soft lockup or
a random crash at some point in the future as we end up touching freed
memory.  One scenario that I was able to reproduce is as follows:

            

con_fault_finish()
  osd_reset()
                              
                              ceph_osdc_handle_map()
                                
                                kick_requests()
                                  
                                  reset_changed_osds()
                                    __reset_osd()
                                      __remove_osd()
                                  
                                
    
    
    __kick_osd_requests()
      __reset_osd()
        __remove_osd() <-- !!!

A case can be made that osd refcounting is imperfect and reworking it
would be a proper resolution, but for now Sage and I decided to fix
this by adding a safe guard around __remove_osd().

Fixes: http://tracker.ceph.com/issues/8087

Cc: Sage Weil 
Signed-off-by: Ilya Dryomov 
Reviewed-by: Sage Weil 
Reviewed-by: Alex Elder 
Signed-off-by: Greg Kroah-Hartman