The test performs shallow scrubs, intentionally using small chunk
sizes to allow dump commands time to check specific details.
Following commit ffda64119f
(PR#44749), shallow scrubs chunks are controlled by a separate
configuration parameter. This PR fixes the test to use the
correct parameter.
An additional minor change is an adjustment to the test loop sleep time:
it is now reduced to guarantee that a dump followed by a counter
increase will be performed in more-or-less the scrubs frequency.
Fixes: https://tracker.ceph.com/issues/58797
Signed-off-by: Ronen Friedman <rfriedma@redhat.com>
Add support for testing CephFS mounted with FUSE using xfstests-dev.
Fixes: https://tracker.ceph.com/issues/55354
Signed-off-by: Rishabh Dave <ridave@redhat.com>
crimson,mon: add guards to avoid accidental crimson deployment and to avoid usage of unsupported features with crimson
Reviewed-by: Radoslaw Zarzynski <rzarzyns@redhat.com>
Reviewed-by: Greg Farnum <gfarnum@redhat.com>
The string returned by get_system_type() for RHEL 8.4 recently wasn't
one of the strings that xfstests_dev.py expects. The string returned
this time was instead "rhel". Add this string to the list of strings
that xfstests_dev.py expects for Red Hat OS family so that this failure
won't occur again in future. Also log the detected distro and version so
that it is easier to debug when this issue occurs next time.
Fixes: https://tracker.ceph.com/issues/58726
Signed-off-by: Rishabh Dave <ridave@redhat.com>
Otherwise the setting doesn't take effect. While at it, replace
home-grown stringify() with standard to_string().
Fixes: https://tracker.ceph.com/issues/58833
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Temporarily disable teuthology cron jobs before
the Reef dev freeze. This commit is intended to
lessen the load on the teuthology queue as we
test final patches for Reef, but should be reverted
after the dev freeze.
Signed-off-by: Laura Flores <lflores@redhat.com>
The "results" directory in xfstests-dev repo contains logs that are
created by it during execution of its tests. Copying this directory to
Ceph log directory saves it from being erased during tear down conducted
by teuthology test runner. These logs allow deeper inspection in to the
cause of failure of tests.
Signed-off-by: Rishabh Dave <ridave@redhat.com>
Create user "fsgqa2" and, since xfstests-dev's README uses option "-m"
while creating user "fsgqa", do the same in our codebase.
Creating user "fsgqa2" is optional along with other users according to
xfstests-dev's README. Create all these users is a preventive step since
tests needing user/users maybe skipped during execution.
Signed-off-by: Rishabh Dave <ridave@redhat.com>
In xfstests-dev repo, local.config options "SCRATCH_DEV" and
"SCRATCH_MNT" were enabled in commit
2ba6b87105 but TODO comment for the
wasn't deleted. Let's delete that and use format string instead of
format function on regular string.
Signed-off-by: Rishabh Dave <ridave@redhat.com>
The intention of this re-arrangement is to make it easier for developers
to use an already cloned and built xfstests-dev repo with
vstart_runner.py.
The re-arrangements are -
* move tearDown() closer to setUp() (so that devs can
enable/disable parts of them easily),
* carve out separate methods (so that these codeblocks can be disabled
in a single go) for -
** deleting groups and users,
** deleting xfstests-dev and xfsprogs-dev repo and
** building xfstests-dev repo
* carve out method for deleting xfstests-dev and xfsprogs-dev repos
* add self.xfstests_repo_path comment (for devs to set path to
xfstests-dev repo), and,
* add a note showing how to use already cloned and built xfstests-dev
repo
Signed-off-by: Rishabh Dave <ridave@redhat.com>
From the xfstests-dev's README doc set DIFF_LENGTH to 0 will print
the full diff instead of the default 10.
This will be very helpful to debug the test failures, or we will
lose many valuable debug infomation.
Signed-off-by: Xiubo Li <xiubli@redhat.com>
Commit dc69033763 moves cephfs-shell from
"<CEPH-REPO-ROOT>/src/tools/cephfs/" to
"<CEPH-REPO-ROOT>/src/tools/cephfs/shell" but cephfs-shell's location in
src/vstart.sh and qa/tasks/cephfs/test_cephfs_shell.py is left
un-updated. This produces a broken vstart_environment.sh and broken
export command in test_cephfs_shell.py.
Introduced-by: dc69033763
Fixes: https://tracker.ceph.com/issues/58795
Signed-off-by: Rishabh Dave <ridave@redhat.com>
At the moment, the latency results are reported in nanoseconds.
In order to improve readability, we'll convert it to seconds.
While at it, we'll fix the fio duration report, which we're
wrongfully dividing by 1000 twice.
Signed-off-by: Lucian Petrut <lpetrut@cloudbasesolutions.com>
Added data regarding completion latency (ns) to the fio results
table, such as min, max, mean, stddev, etc.
Signed-off-by: Stefan Chivu <schivu@cloudbasesolutions.com>
"rbd feature disable" appears to reliably hang if the corresponding
remote request is proxied to rbd-nbd (because rbd-nbd happens to own
the exclusive lock after a series of blkdiscard calls) [1]. Work
around it here by enabling journaling before the image is mapped
and disabling it after the image is unmapped.
Also, don't assert on the output of "rbd journal inspect --verbose"
having a certain number of entries. This is racy: if the script gets
delayed after the last blkdiscard call for some reason, there may be
fewer entries present in the journal or none at all.
[1] https://tracker.ceph.com/issues/58740
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
* Updated the existing tests
* Added new qa tests for --dump and --dumpfs options
* Updated the cluster configuration to 2 mds, 2 clients
Fixes: https://tracker.ceph.com/issues/57014
Signed-off-by: Jos Collin <jcollin@redhat.com>
The test_fscrypt_encrypt will only run the 'encrypt' related test
cases without 'test_dummy_encryption' option enabled. This will
test the filename and content verification.
After this I will add the whole test cases with 'test_dummy_encryption'
option.
Signed-off-by: Xiubo Li <xiubli@redhat.com>
osd/scrub: use separate chunk size configuration for shallow scrubs
Reviewed-by: Samuel Just <sjust@redhat.com>
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
Reviewed-by: Mark Kogan <mkogan@redhat.com>
teuthology/suite/placeholder.py sets it to default otherwise.
Fixes: https://tracker.ceph.com/issues/57873
Signed-off-by: Samuel Just <sjust@redhat.com>
Using the existing common default chunk size for scrubs that are
not deep scrubs is wasteful: a high ratio of inter-OSD messages
per chunk, while the actual OSD work per chunk is minimal.
Signed-off-by: Ronen Friedman <rfriedma@redhat.com>
When upgrading osd with filestore to reef, restart should not be possible
the osd won't boot and error message will be showed in the osd log
Signed-off-by: Nitzan Mordechai <nmordec@redhat.com>
