linux-stable.git/net/ceph, branch v4.0.7

crush: ensuring at most num-rep osds are selected

2015-06-23T00:03:20+00:00

commit 45002267e8d2699bf9b022315bee3dd13b044843 upstream.

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: Greg Kroah-Hartman

Revert "libceph: clear r_req_lru_item in __unregister_linger_request()"

2015-06-06T15:20:51+00:00

commit 521a04d06a729e5971cdee7f84080387ed320527 upstream.

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.

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

libceph: request a new osdmap if lingering request maps to no osd

2015-06-06T15:20:51+00:00

commit b0494532214bdfbf241e94fabab5dd46f7b82631 upstream.

This commit does two things.  First, if there are any homeless
lingering requests, we now request a new osdmap even if the osdmap that
is being processed brought no changes, i.e. if a given lingering
request turned homeless in one of the previous epochs and remained
homeless in the current epoch.  Not doing so leaves us with a stale
osdmap and as a result we may miss our window for reestablishing the
watch and lose notifies.

MON=1 OSD=1:

    # cat linger-needmap.sh
    #!/bin/bash
    rbd create --size 1 test
    DEV=$(rbd map test)
    ceph osd out 0
    rbd map dne/dne # obtain a new osdmap as a side effect (!)
    sleep 1
    ceph osd in 0
    rbd resize --size 2 test
    # rbd info test | grep size -> 2M
    # blockdev --getsize $DEV -> 1M

N.B.: Not obtaining a new osdmap in between "osd out" and "osd in"
above is enough to make it miss that resize notify, but that is a
bug^Wlimitation of ceph watch/notify v1.

Second, homeless lingering requests are now kicked just like those
lingering requests whose mapping has changed.  This is mainly to
recognize that a homeless lingering request makes no sense and to
preserve the invariant that a registered lingering request is not
sitting on any of r_req_lru_item lists.  This spares us a WARN_ON,
which commit ba9d114ec557 ("libceph: clear r_req_lru_item in
__unregister_linger_request()") tried to fix the _wrong_ way.

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

Revert "libceph: use memalloc flags for net IO"

2015-04-07T16:08:35+00:00

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

Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client

2015-02-19T22:14:42+00:00

Pull Ceph changes from Sage Weil:
 "On the RBD side, there is a conversion to blk-mq from Christoph,
  several long-standing bug fixes from Ilya, and some cleanup from
  Rickard Strandqvist.

  On the CephFS side there is a long list of fixes from Zheng, including
  improved session handling, a few IO path fixes, some dcache management
  correctness fixes, and several blocking while !TASK_RUNNING fixes.

  The core code gets a few cleanups and Chaitanya has added support for
  TCP_NODELAY (which has been used on the server side for ages but we
  somehow missed on the kernel client).

  There is also an update to MAINTAINERS to fix up some email addresses
  and reflect that Ilya and Zheng are doing most of the maintenance for
  RBD and CephFS these days.  Do not be surprised to see a pull request
  come from one of them in the future if I am unavailable for some
  reason"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client: (27 commits)
  MAINTAINERS: update Ceph and RBD maintainers
  libceph: kfree() in put_osd() shouldn't depend on authorizer
  libceph: fix double __remove_osd() problem
  rbd: convert to blk-mq
  ceph: return error for traceless reply race
  ceph: fix dentry leaks
  ceph: re-send requests when MDS enters reconnecting stage
  ceph: show nocephx_require_signatures and notcp_nodelay options
  libceph: tcp_nodelay support
  rbd: do not treat standalone as flatten
  ceph: fix atomic_open snapdir
  ceph: properly mark empty directory as complete
  client: include kernel version in client metadata
  ceph: provide seperate {inode,file}_operations for snapdir
  ceph: fix request time stamp encoding
  ceph: fix reading inline data when i_size > PAGE_SIZE
  ceph: avoid block operation when !TASK_RUNNING (ceph_mdsc_close_sessions)
  ceph: avoid block operation when !TASK_RUNNING (ceph_get_caps)
  ceph: avoid block operation when !TASK_RUNNING (ceph_mdsc_sync)
  rbd: fix error paths in rbd_dev_refresh()
  ...

libceph: kfree() in put_osd() shouldn't depend on authorizer

2015-02-19T11:27:51+00:00

a255651d4cad ("ceph: ensure auth ops are defined before use") made
kfree() in put_osd() conditional on the authorizer.  A mechanical
mistake most likely - fix it.

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

libceph: fix double __remove_osd() problem

2015-02-19T11:27:50+00:00

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 
Cc: stable@vger.kernel.org # 3.9+: 7c6e6fc53e73: libceph: assert both regular and lingering lists in __remove_osd()
Cc: stable@vger.kernel.org # 3.9+: cc9f1f518cec: libceph: change from BUG to WARN for __remove_osd() asserts
Cc: stable@vger.kernel.org # 3.9+
Signed-off-by: Ilya Dryomov 
Reviewed-by: Sage Weil 
Reviewed-by: Alex Elder

libceph: tcp_nodelay support

2015-02-19T10:31:40+00:00

TCP_NODELAY socket option set on connection sockets,
disables Nagle’s algorithm and improves latency characteristics.
tcp_nodelay(default)/notcp_nodelay option flags provided to
enable/disable setting the socket option.

Signed-off-by: Chaitanya Huilgol 
[idryomov@redhat.com: NO_TCP_NODELAY -> TCP_NODELAY, minor adjustments]
Signed-off-by: Ilya Dryomov

libceph: use mon_client.c/put_generic_request() more

2015-02-19T10:31:37+00:00

Signed-off-by: Ilya Dryomov

libceph: nuke pool op infrastructure

2015-02-19T10:31:37+00:00

On Mon, Dec 22, 2014 at 5:35 PM, Sage Weil  wrote:
> On Mon, 22 Dec 2014, Ilya Dryomov wrote:
>> Actually, pool op stuff has been unused for over two years - looks like
>> it was added for rbd create_snap and that got ripped out in 2012.  It's
>> unlikely we'd ever need to manage pools or snaps from the kernel client
>> so I think it makes sense to nuke it.  Sage?
>
> Yep!

Signed-off-by: Ilya Dryomov