Josh Durgin [Tue, 26 Jun 2012 20:49:46 +0000 (13:49 -0700)]
cls_rbd: add methods for dealing with rbd_directory objects
These hide the structure of the directory from clients, and provide
mappings from id to name and name to id. These mappings aren't needed
for old format images, but with layering, they're useful to show
the user e.g. the name of an image's parent instead of its id.
This also makes the directory more efficient for new-style images, since
it uses omap instead of tmap.
Josh Durgin [Mon, 25 Jun 2012 19:47:31 +0000 (12:47 -0700)]
librbd: add indirection between name and header object
This adds a new rbd_id.$name object that contains the id of an image.
Rename the header in new-format images to rbd_header.$id, so that
it doesn't need to change when an image is renamed - the rbd_id.$name
object is copied to a new object instead.
With layering, this becomes more important because many child images
could be depending on the header object of a parent image.
This is another change to the new format, but since no one uses it yet,
we don't need to worry about backwards compatibility.
Yehuda Sadeh [Wed, 27 Jun 2012 00:16:11 +0000 (17:16 -0700)]
rest-bench: mark request as complete later
We marked a request as complete in the callback, however
it might be that we're still inside S3_runall_request_context()
which means that request is not really complete yet.
Possibly fixes bug #2652.
Sage Weil [Wed, 20 Jun 2012 18:07:29 +0000 (11:07 -0700)]
objecter: do not feed session to op_submit()
The linger_send() method was doing this, but it is problematic because the
new Op doesn't get its pgid or acting vector set correctly. The result is
that the request goes to the right OSD, but has the wrong pgid, and makes
the OSD complain about misdirected requests and drop it on the floor. It
didn't affect the test results because we weren't testing whether the
watch was working in that case.
Instead, we'll just recalculate and get the same value the parent linger
op did. Which is fine, and goes through all the usual code paths so
nothing is missed.
Also, increment num_homeless_ops before we recalc_op_target(), so that we
don't (harmlessly, but confusingly) underflow.
Fixes: #2022 Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Wed, 20 Jun 2012 18:07:29 +0000 (11:07 -0700)]
objecter: do not feed session to op_submit()
The linger_send() method was doing this, but it is problematic because the
new Op doesn't get its pgid or acting vector set correctly. The result is
that the request goes to the right OSD, but has the wrong pgid, and makes
the OSD complain about misdirected requests and drop it on the floor. It
didn't affect the test results because we weren't testing whether the
watch was working in that case.
Instead, we'll just recalculate and get the same value the parent linger
op did. Which is fine, and goes through all the usual code paths so
nothing is missed.
Also, increment num_homeless_ops before we recalc_op_target(), so that we
don't (harmlessly, but confusingly) underflow.
Fixes: #2022 Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Thu, 21 Jun 2012 14:31:47 +0000 (07:31 -0700)]
mon: encoding new monmap using quorum feature set
It is probably unlikely that someone will expand the mon cluster with a
mixed feature set, but we know the quorum features here, so we should use
them.
Sage Weil [Thu, 21 Jun 2012 03:41:17 +0000 (20:41 -0700)]
mon: conditionally encode auth incremental with quorum feature bits
If the quorum does not yet all have the MONENC feature, stick to the old
encoding.
It might be more polite to require a super-quorum before switching over,
and take note so that thereafter we can stick to the new encoding, but
that has more moving parts and I'm not sure it's worth the complexity.
Sage Weil [Thu, 21 Jun 2012 03:33:41 +0000 (20:33 -0700)]
mon: track intersection of quorum member features
When we form a quorum, also note the intersection of the quorum members'
feature bits. This will inform decisions about what encodings we use.
This is an imperfect strategy because the quorum may change, and we may
have a mon with old code join in and not understand what is going on.
However, it does ensure that a majority of the members run new code, so in
the absence of other failures we can make progress.
This was an ill-conceived approach to getting atomic transactions out of
btrfs. It doesn't offer rollback, which means that any error means we need
to wedge the file system and reboot in order to avoid corrupting the
data set. And that's silly!
Snapshots are more robust and only marginally slower (because we have to
quiesce our writes while waiting for the snap to start, and btrfs resume
work in-kernel slightly faster...maybe).
Fixes: #2623 Signed-off-by: Sage Weil <sage@inktank.com>
Sage Weil [Fri, 15 Jun 2012 17:42:49 +0000 (10:42 -0700)]
cls_rbd: check for LAYERING feature for parent methods
You need to set the feature on the image before you can use these methods.
We *could* allow get_parent, but that just means they'll get ENOENT instead
of ENOEXEC, and the latter is more informative.
Samuel Just [Tue, 19 Jun 2012 16:11:57 +0000 (09:11 -0700)]
PG: improve find_best_info
07f853db3982e68b952a337cf91cbf7ec0709de9 is actually too conservative,
it suffices to find any info with a last_update of at least the least
last_update from the last period to go active. An info from a previous
interval is acceptable if the last interval never reported a commited
operation and thus still has the same last_update.
Signed-off-by: Samuel Just <sam.just@inktank.com> Reviewed-by: Sage Weil <sage@inktank.com>
Sage Weil [Mon, 18 Jun 2012 21:00:06 +0000 (14:00 -0700)]
mon: gracefully handle slow 'ceph -w' clients
If we are sending log updates to a client (ceph -w), and they are far
enough behind to drop behind first_committed, include a friendly message
in their stream but continue.
Drop useless return value from _create_sub_incremental(). Assert that we
can read the state file.
Samuel Just [Sat, 16 Jun 2012 00:09:42 +0000 (17:09 -0700)]
PG: best_info must have a last_epoch_started as high as any other info
We disregard incomplete infos during find_best_info, but we can't an
info with a last_epoch_started less that of the incomplete info.
This should avoid cases like #2462. In that case, it appears that
a peer with empty info/log was chosen as authoritative even though
there was a non-empty incomplete peer.
Samuel Just [Sat, 16 Jun 2012 00:09:42 +0000 (17:09 -0700)]
PG: best_info must have a last_epoch_started as high as any other info
We disregard incomplete infos during find_best_info, but we can't an
info with a last_epoch_started less that of the incomplete info.
This should avoid cases like #2462. In that case, it appears that
a peer with empty info/log was chosen as authoritative even though
there was a non-empty incomplete peer.
projects / ceph.git / log