TestFileStoreState: distinguish between 'get_coll()' and 'get_coll_at()'
get_coll_at(int pos) should return the collection at the map's position
'pos', but 'pos' was being used as a map key. Therefore, we add a new
function 'get_coll(int key)' to mimic this behavior, and we make
'get_coll_at()' follow its intended behavior.
This patch may affect the test_filestore_idempotent_sequence tester, since
it uses the 'get_coll_at()' function a lot, and we changed this function's
behavior.
Signed-off-by: Joao Eduardo Luis <jecluis@gmail.com>
run_seed_to.sh: rework the script, make it more flexible and broaden the tests.
Allow for '-h' and other options such as disabling the journal sync tests,
defining it is to be run on a btrfs FS, enabling exit on error (default is
now 'off'), and allow certain env variables to specify additional options
to each store.
Signed-off-by: Joao Eduardo Luis <jecluis@gmail.com>
Sage Weil [Sat, 21 Apr 2012 00:13:08 +0000 (17:13 -0700)]
librbd: fix ictx_check pointer weirdness by using std::string
I was seeing failures of LibRBD.TestIOToSnapshot where we would fail to
refresh after rollback, even though the snap existed. I assume it is
because the std::string whose c_str() we were pointing to was reallocated.
Sage Weil [Fri, 20 Apr 2012 23:56:57 +0000 (16:56 -0700)]
filestore: fix collection_add journal replay problem
In collection_add we have a two-phase guard set on the linked object via
the old name. During replay, we might see that the dest name is missing
and replay the operation, and in the process overwrite a newer guard with
an older one.
Avoid this by checking the source name too, and skipping the operation
entirely if a new guard exists.
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
This code should be on a stand-alone class, instead of being embedded on
a single test, in case someone or something find it useful somewhere down
the line.
Signed-off-by: Joao Eduardo Luis <jecluis@gmail.com>
Sage Weil [Fri, 20 Apr 2012 23:36:54 +0000 (16:36 -0700)]
log: prefix dump with line numbers
This makes it easier to interpret the dump, and makes it obvious what is
dump (and potentially a dup of something that was already logged) and what
is not.
Sage Weil [Wed, 18 Apr 2012 21:39:18 +0000 (14:39 -0700)]
osd: dump old ops singly rather than all at once.
Fixes #2269. Convert the OpTracker::check_ops_in_flight interface
to take a vector<string> and create a separate warning for each old
Op, and dump those singly to the clog in the OSD.
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com> Reviewed-by: Sage Weil <sage@newdream.net>
pgmap: allow Incrementals to specify [near]full_ratios of 0
This commit isn't entirely safe: old monitors used 0 to mean "no change".
We can revert this (and the PGMonitor.cc portion of 841f2885318d1bcf37aab3f2947b1f40fee772a9) if we don't want to allow
0 as a valid ratio setting, and to maintain perfect backwards
compatibility.
pgmon: remove the PGMonitor update_full_ratios stuff
Making it a config watcher is just a huge mess in terms of consistently
updating it appropriately.
The next commit will add a monitor command for changing it.
Sage Weil [Tue, 17 Apr 2012 17:32:38 +0000 (10:32 -0700)]
mon: only fill in full/nearfull sets if the ratio > 0
This avoids putting all OSDs in both sets when the ratios are 0, as they
are with a fresh cluster and pgmap. This also makes setting the ratio to
0 effectively disable the full/nearfull feature.
mon: unconditionally encode PGMap full ratios in the Incremental
This properly spreads the real value to peon monitors -- they weren't
seeing the right values at all before.
Initialize all related values to zero so that it's obvious if they
somehow avoided becoming set properly.
This doesn't require any kind of protocol revision, luckily -- mixing
monitors from before and after this change might result in extra work
recalculating full sets, but it won't spread bad values or anything.
Alex Elder [Tue, 17 Apr 2012 13:33:42 +0000 (08:33 -0500)]
qa: comment out xfstest 232
Test 232 in the xfstests suite produces an XFS error in the log
when run over an RBD device. This is most likely an XFS problem
that will be tracked separately (in tracker 2302).
My original plan with getting this checked in was to have it run a
baseline set of the tests--all known to pass on rbd devices--with
the intention of doing ongoing work to add back missing tests (at
least from the "auto" group) as we understand and fix whatever
makes them produce failures.
So just comment out test 232 so the xfstests script is able to
run to completion without error.
objecter: recalc pgid in case it wasn't a precalculated pgid
We had an issue where we didn't call it since we relied on the
existance of the object name and we were passed an empty one.
However, we still need to handle the case of list_objects, where
we don't want to recalc pgid.
Alex Elder [Sat, 14 Apr 2012 16:43:15 +0000 (11:43 -0500)]
run_xfstests.sh: ensure cleanup on errors
Because we exit on any error (due to 'set -e'), the cleanup call was
never getting made in the event of an error. The net effect of that
was that a filesystem could be left mounted, and rbd cleanup then
couldn't complete because the module was in use.
Fix the trap call so it calls cleanup on exit as well as error.
Switch to using the capitalized signal names in the call.
Alex Elder [Sat, 14 Apr 2012 16:26:21 +0000 (11:26 -0500)]
run_xfstests.sh: pass test result via exit status
It turns out that xfstests *does* exit with non-zero status
when a test fails. Its exit status is the number of tests
that failed (which, now that we have over 255 tests could be
an issue...)
Save the exit status and make it be the result of the run.
Alex Elder [Sat, 14 Apr 2012 02:26:22 +0000 (21:26 -0500)]
qa: add run_xfsests.sh script
Add a script that runs xfstests over a pair of devices that are
specified using command line arguments. The tests are run using
a specified filesystem type (xfs, ext4, or btrfs).
A default set of tests is run if none is specified on the command
line. Normally there's an "auto" group used for this purpose, but
for now I've laid out a (large) subset of them that I know pass on
rbd devices. These can be updated as we find they work reliably.
projects / ceph.git / log