Sage Weil [Wed, 28 Aug 2013 16:50:11 +0000 (09:50 -0700)]
mon: discover mon addrs, names during election state too
Currently we only detect new mon addrs and names during the probing phase.
For non-trivial clusters, this means we can get into a sticky spot when
we discover enough peers to form an quorum, but not all of them, and the
undiscovered ones are enough to break the mon ranks and prevent an
election.
One way to work around this is to continue addr and name discovery during
the election. We should also consider making the ranks less sensitive to
the undefined addrs; that is a separate change.
Fixes: #4924
Backport: dumpling Signed-off-by: Sage Weil <sage@inktank.com> Tested-by: Bernhard Glomm <bernhard.glomm@ecologic.eu>
Loic Dachary [Tue, 27 Aug 2013 11:58:33 +0000 (13:58 +0200)]
common: move SharedPtrRegistry test after t.join
The thread created to test SharedPtrRegistry race conditions updates a
value ( ptr ) that is tested by the main gtest thread but is not
protected by a lock. Instead of adding a lock, the main thread tests
the value after pthread_join() on the child thread.
mon: DataHealthService: monitor backing store's size and report it
If the store's size grows beyond what we believe to be reasonable, we must
let the user know that something fishy may be going on. This intends to
act as an early warning system for monitors suffering from leveldb
compaction issues. However, if the monitor's store is just growing a lot
due to normal cluster behaviour, we made sure that the warning threshold
is adjustable by tuning 'mon_leveldb_size_warn' (defaulting to 40GB).
Fixes: #5909 Signed-off-by: Joao Eduardo Luis <jecluis@gmail.com>
Dan Mick [Fri, 23 Aug 2013 00:30:24 +0000 (17:30 -0700)]
ceph_rest_api.py: create own default for log_file
common/config thinks the default log_file for non-daemons should be "".
Override that so that the default is
/var/log/ceph/{cluster}-{name}.{pid}.log
since ceph-rest-api is more of a daemon than a client.
Fixes: #6099
Backport: dumpling Signed-off-by: Dan Mick <dan.mick@inktank.com>
David Disseldorp [Mon, 29 Jul 2013 15:05:44 +0000 (17:05 +0200)]
mds: remove waiting lock before merging with neighbours
CephFS currently deadlocks under CTDB's ping_pong POSIX locking test
when run concurrently on multiple nodes.
The deadlock is caused by failed removal of a waiting_locks entry when
the waiting lock is merged with an existing lock, e.g:
Note that the waiting 4116@1:1 lock entry is merged with the existing
4116@0:1 held lock to become a 4116@0:2 held lock. However, the now
handled 4116@1:1 waiting_locks entry remains.
When handling a lock request, the MDS calls adjust_locks() to merge
the new lock with available neighbours. If the new lock is merged,
then the waiting_locks entry is not located in the subsequent
remove_waiting() call because adjust_locks changed the new lock to
include the old locks.
This fix ensures that the waiting_locks entry is removed prior to
modification during merge.
Signed-off-by: David Disseldorp <ddiss@suse.de> Reviewed-by: Greg Farnum <greg@inktank.com>
os: KeyValueDB: expose interface to obtain estimated store size
On LevelDBStore, instead of using leveldb's GetApproximateSizes() function,
we will instead assess what's the store's raw size from the contents of
the store dir (this means .sst's, .log's, etc). The reason behind this
approach is that GetApproximateSizes() would expect us to provide a range
of keys for which to obtain an approximate size; on the other hand, what we
really want is to obtain the size of the store -- not the size of the
data (besides, with the compaction issues we've been seeing, we wonder
how reliable such approximation would be).
Signed-off-by: Joao Eduardo Luis <jecluis@gmail.com>
Sage Weil [Thu, 22 Aug 2013 22:54:48 +0000 (15:54 -0700)]
mon/Paxos: fix another uncommitted value corner case
It is possible that we begin the paxos recovery with an uncommitted
value for, say, commit 100. During last/collect we discover 100 has been
committed already. But also, another node provides an uncommitted value
for 101 with the same pn. Currently, we refuse to learn it, because the
pn is not strictly > than our current uncommitted pn... even though it is
the next last_committed+1 value that we need.
There are two possible fixes here:
- make this a >= as we can accept newer values from the same pn.
- discard our uncommitted value metadata when we commit the value.
Let's do both!
Fixes: #6090 Signed-off-by: Sage Weil <sage@inktank.com>
Yehuda Sadeh [Mon, 19 Aug 2013 23:56:27 +0000 (16:56 -0700)]
rgw: bucket meta remove don't overwrite entry point first
Fixes: #6056
When removing a bucket metadata entry we first unlink the bucket
and then we remove the bucket entrypoint object. Originally
when unlinking the bucket we first overwrote the bucket entrypoint
entry marking it as 'unlinked'. However, this is not really needed
as we're just about to remove it. The original version triggered
a bug, as we needed to propagate the new header version first (which
we didn't do, so the subsequent bucket removal failed).
Sandon Van Ness [Fri, 23 Aug 2013 02:44:40 +0000 (19:44 -0700)]
QA: Compile fsstress if missing on machine.
Some distro's have a lack of ltp-kernel packages and all we need is
fstress. This just modified the shell script to download/compile
fstress from source and copy it to the right location if it doesn't
currently exist where it is expected. It is a very small/quick
compile and currently only SLES and debian do not have it already.
Reviewed-by: Sage Weil <sage@inktank.com> Signed-off-by: Sandon Van Ness <sandon@inktank.com>
Sandon Van Ness [Fri, 23 Aug 2013 02:44:40 +0000 (19:44 -0700)]
QA: Compile fsstress if missing on machine.
Some distro's have a lack of ltp-kernel packages and all we need is
fstress. This just modified the shell script to download/compile
fstress from source and copy it to the right location if it doesn't
currently exist where it is expected. It is a very small/quick
compile and currently only SLES and debian do not have it already.
Reviewed-by: Sage Weil <sage@inktank.com> Signed-off-by: Sandon Van Ness <sandon@inktank.com>
Gary Lowell [Thu, 22 Aug 2013 18:07:16 +0000 (11:07 -0700)]
ceph.spec.in: Don't invoke debug_package macro on centos.
If the redhat-rpm-config package is installed, the debuginfo rpms will
be built by default. The build will fail when the package installed
and the specfile also invokes the macro.
Signed-off-by: Gary Lowell <gary.lowell@inktank.com>
Alexandre Oliva [Thu, 22 Aug 2013 06:40:22 +0000 (03:40 -0300)]
enable mds rejoin with active inodes' old parent xattrs
When the parent xattrs of active inodes that the mds attempts to open
during rejoin lack pool info (struct_v < 5), this field will be filled
in with -1, causing the mds to retry fetching a backtrace with a pool
number that matches the expected value, which fails and causes the
err==-ENOENT branch to be taken and retry pool 1, which succeeds, but
with pool -1, and so keeps on bouncing between the two retry cases
forever.
This patch arranges for the mds to go along with pool -1 instead of
insisting that it be refetched, enabling it to complete recovery
instead of eating cpu, network bandwidth and metadata osd's resources
like there's no tomorrow, in what AFAICT is an infinite and very busy
loop.
This is not a new problem: I've had it even before upgrading from
Cuttlefish to Dumpling, I'd just never managed to track it down, and
force-unmounting the filesystem and then restarting the mds was an
easier (if inconvenient) work-around, particularly because it always
hit when the filesystem was under active, heavy-ish use (or there
wouldn't be much reason for caps recovery ;-)
There are two issues not addressed in this patch, however. One is
that nothing seems to proactively update the parent xattr when it is
found to be outdated, so it remains out of date forever. Not even
renaming top-level directories causes the xattrs to be recursively
rewritten. AFAICT that's a bug.
The other is that inodes that don't have a parent xattr (created by
even older versions of ceph) are reported as non-existing in the mds
rejoin message, because the absence of the parent xattr is signaled as
a missing inode (?failed to reconnect caps for missing inodes?). I
suppose this may cause more serious recovery problems.
I suppose a global pass over the filesystem tree updating parent
xattrs that are out-of-date would be desirable, if we find any parent
xattrs still lacking current information; it might make sense to
activate it as a background thread from the backtrace decoding
function, when it finds a parent xattr that's too out-of-date, or as a
separate client (ceph-fsck?).
Loic Dachary [Tue, 13 Aug 2013 15:28:31 +0000 (17:28 +0200)]
ReplicatedPG: create ObjectContext with SharedPtrRegistry
All new ObjectContext are replaced with calls to
SharedPtrRegistry::lookup_or_create to ensure that they are all
registered. Because the constructor is invoked with no argument, care
is taken to always initialize the destructor_callback data member
immediately afterwards.
ReplicatedPG::get_object_context contains a redundant call to
get_snapset_context that is removed.
Loic Dachary [Mon, 12 Aug 2013 14:47:42 +0000 (16:47 +0200)]
ReplicatedPG: ObjectContext is made compatible with SharedPtrRegistry
When creating a new object SharedPtrRegistry::lookup_or_create uses
the default ObjectContext constructor with no argument. The existing
ObjectContext constructor is modified to have no argument and the
initialization that was previously done within the constructor is done
by the caller (that only happens three times).
The ObjectContext::get method is removed: its only purpose is to
increment the ref.
The ObjectContext::registered data member is removed as well as all
the associated assert()
The ObjectContext::destructor_callback data member Context is added
and called by the destructor. It will allow the caller to perform
additional cleanup, if necessary.
All ObjectContext * data members are replaced with shared_ptr.
Loic Dachary [Thu, 15 Aug 2013 18:15:03 +0000 (20:15 +0200)]
ReplicatedPG: add Mutex to protect snapset_contexts
snapset_contexts_locks is added and locked in each function where
snapset_contexts or the SnapSetContext::ref data member needs to be
accessed or modified.
Loic Dachary [Mon, 12 Aug 2013 12:05:38 +0000 (14:05 +0200)]
sharedptr_registry: add a variant of get_next() and the empty() method
The SharedPtrRegistry::get_next() method with a value of type VPtr
instead of V is added because it is sometime more convenient to not
copy the value when walking the registry. The
SharedPtrRegistry::empty() predicate method is added.
Josh Durgin [Wed, 21 Aug 2013 21:28:49 +0000 (14:28 -0700)]
objecter: resend unfinished lingers when osdmap is no longer paused
Plain Ops that haven't finished yet need to be resent if the osdmap
transitions from full or paused to unpaused. If these Ops are
triggered by LingerOps, they will be cancelled instead (since
should_resend = false), but the LingerOps that triggered them will not
be resent.
Fix this by checking the registered flag for all linger ops, and
resending any of them that aren't paused anymore.
Fixes: #6070 Signed-off-by: Josh Durgin <josh.durgin@inktank.com> Reviewed-by: Sage Weil <sage.weil@inktank.com>
Yehuda Sadeh [Mon, 19 Aug 2013 15:40:16 +0000 (08:40 -0700)]
rgw: change cache / watch-notify init sequence
Fixes: #6046
We were initializing the watch-notify (through the cache
init) before reading the zone info which was much too
early, as we didn't have the control pool name yet. Now
simplifying init/cleanup a bit, cache doesn't call watch/notify
init and cleanup directly, but rather states its need
through a virtual callback.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> Reviewed-by: Sage Weil <sage@inktank.com>
projects / ceph.git / log