Samuel Just [Fri, 18 Apr 2014 00:26:17 +0000 (17:26 -0700)]
ReplicatedPG: block scrub on blocked object contexts
Fixes: #8011 Signed-off-by: Samuel Just <sam.just@inktank.com> Reviewed-by: Sage Weil <sage@inktank.com>
(cherry picked from commit e66f2e36c06ca00c1147f922d3513f56b122a5c0)
Samuel Just [Thu, 24 Apr 2014 19:48:44 +0000 (12:48 -0700)]
ECBackend::continue_recovery_op: handle a source shard going down
get_min_avail_to_read_shards might return an error if there are
no longer enough sources to reconstruct the missing shards.
This is possible if osds went down while we were writing the
previous chunk -- we already notice in check_recovery_sources
if a source goes down during a read.
Samuel Just [Fri, 11 Apr 2014 01:15:30 +0000 (18:15 -0700)]
ReplicatedPG: do not preserve op context during flush
Any information stashed in the OpContext may be obsolete by the time we
actually mark the object clean. Instead, let the start_flush caller
clean up its OpContext and in try_flush_mark_clean we'll create a new
one. The primary reason to keep the OpContext would have been locking,
but we can set the obc as blocking without holding an OpContext, and
that would allow trimming to happen in the mean time (which is good
since trim_object does not respect rw locks since it doesn't change user
visible state). In try_flush_mark_clean, we requeue the fop->op along
with (but ahead of) the fop->dup_ops.
Modified qemu-iotests workunit script to check for versions
that use the latest qemu (currently only Trusty). Limit the
tests to those that are applicable to rbd.
client: check cap ID when handling cap export message
handle following sequence of events:
- mds0 exports an inode to mds1. client receives the cap import
message from mds1. caps from mds0 are removed while handling
the cap import message.
- mds1 exports an inode to mds0. client receives the cap export
message from mds1. handle_cap_export() adds placeholder caps
for mds0
- client receives the first cap export message (for exporting
inode from mds0 to mds1)
Sage Weil [Sun, 20 Apr 2014 05:04:33 +0000 (22:04 -0700)]
osd: change in up set primary constitutes a peering interval change
In several places, a change in the up_primary triggers a new peering
interval, but the palces that actually generate the new past intervals,
including check_new_interval(), did not enforce that. This becomes
somewhat obvious when you see that those callers are ignoring the
up_primary output argument for pg_to_up_acting_osds().
Fix this by adding arguments to check_new_interval and fixing the callers
to pass them in properly. Add a unit test case to verify this.
Note that the past interval struct itself does not record who the
up_primary was; possibly it should.
Fixes: #8139 Signed-off-by: Sage Weil <sage@inktank.com>
Samuel Just [Tue, 8 Apr 2014 21:03:59 +0000 (14:03 -0700)]
ReplicatedPG: do not create whiteout clones
First, make_writeable treats whiteout heads like snapdir for
cloning purposes. Second, to ensure that we send the correct
deletes on flush to the backing pool, we instead use oi.snaps
on any clone we are flushing to infer the snaps during which
head did not exist and send a delete as appropriate prior to
the copy_from.
Normally, we'd have a problem if the delete and the copy_from
completed, but an interval change intervened before the dirty
flag was cleared since we'd end up re-deleting the object.
To avoid that, we use the CEPH_OSD_FLAG_ORDERSNAP flag.
Additionally, we will use the correct snap_seq on the delete
or flush as appropriate to ensure that the previous clone
gets created with the same clone id as in the cache pool.
Fixes: #7942 Signed-off-by: Samuel Just <sam.just@inktank.com>
Samuel Just [Thu, 17 Apr 2014 19:27:07 +0000 (12:27 -0700)]
osd/: propogate hit_set history with repop
We don't actually send the whole info on each repop, just the log
entries, updated stats, and a few other bits. For hit_set ops, we need
to also communicate the new hit_set history status atomically with the
log entries and the transaction. Thus, we add a channel for an optional
pg_hit_set_history_t field in PGBackend::submit_transaction interface
and associated messages and implementations to update the hit_set info
field along with the log entries.
This also means that hit_set_(persist|trim) update an
updated_hit_set_history field on the OpContext instead of directly
modifying the info field.
Fixes: #8124 Signed-off-by: Samuel Just <sam.just@inktank.com>
Sage Weil [Fri, 18 Apr 2014 18:12:23 +0000 (11:12 -0700)]
mon: wait for PaxosService readable in handle_get_version
We were waiting for the election to finish, but we need to *also* wait for
paxos to recover. Being a peon or leader is not sufficient and we may
return a map that is still old.
Fixes: #7997 Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Thu, 17 Apr 2014 20:11:54 +0000 (13:11 -0700)]
osd/ReplicatedPG: check clones for degraded
We check whether the head is degraded, and we check whether a clone is
unreadable, but in the case where we have a cache op on a degraded object,
we don't check. That leads to an assert when the repop hits the replica
and the object is in the peer's missing set.
Fix this by adding a check on the clone when write_ordered is true. Note
that checking write_ordered is better than whether it is a cache op because
we want to preserve write ordering even for reads that are flagged by the
client.
Fixes: #8048 Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Thu, 17 Apr 2014 17:48:26 +0000 (10:48 -0700)]
osdc/Objecter: fix osd target for newly-homeless op
If we recalculate the mapping and find that there is no primary, we need
to set the 'osd' field to -1. Otherwise, the caller will try to resend
to a dead session with bad results.
Sage Weil [Thu, 17 Apr 2014 16:33:44 +0000 (09:33 -0700)]
mon: set leader commands prior to first election
If we have just started and receive a command, we currently will reply with
EINVAL because the leader commands are empty. Note that this race is very
difficult to reach because the (old) peon needs to forward a command to
the mon while it still thinks it has quorum, and the message needs to get
sent after the leader mon has restarted and reset its connection but before
it has declared a new election.
To fix this, we should assume at startup time that our commands are
valid. If it is an internal command that does not require quorum, that
is fine. If it does require quorum, we will retry the command after the
election completes and we will revalidate the command then.
Fixes: #8132 Signed-off-by: Sage Weil <sage@inktank.com>
John Spray [Thu, 17 Apr 2014 14:28:22 +0000 (15:28 +0100)]
mon: EBUSY instead of EAGAIN when pgs creating
In 69321bf, EAGAIN changed behaviour to block indefinitely
rather than returning to user. Change the return for
`osd pool set` operations that are blocked by creating PGs
to return EBUSY instead of EAGAIN, so that they are excepted
from this blocking behaviour.
Signed-off-by: John Spray <john.spray@inktank.com>
projects / ceph.git / log