rgw_log_op() expects more server and request variables from the
request environment. Fix rgw_lib to set more that it can
usefully, but also make rgw_log_op() check for the existence
of all keys it will use.
Fixes: http://tracker.ceph.com/issues/21942
Signed-off-by: Matt Benjamin <mbenjamin@redhat.com>
Required to store correct final accounted_size of objects in
RGWWriteWriteRequest::write_finish.
Fixes: http://tracker.ceph.com/issues/21940
Signed-off-by: Matt Benjamin <mbenjamin@redhat.com>
If the symlink points to the right location, do nothing. If it doesn't,
replace it. If it's not a symlink, bail with EEXIST.
Signed-off-by: Sage Weil <sage@redhat.com>
rbd.xfstests task allows spawning xfstests runs on multiple nodes.
Don't unwind task contexts if one of the runs fails -- let the other
runs finish.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Previously, there was only one time we would end up in this region of code:
when the backfill was rejected by the peer. Previously that was apparently
reliably when we had an outstanding SCAN request, because we would
unconditionally cancle the MAX recovery op and clear waiting_on_backfill.
See 624aaf2a4ea9950153a89ff921e2adce683a6f51 for when this code appeared.
Now we have several similar paths, and we don't always have an outstanding
scan call (I don't think!). Regardless, move most these three cases into
a common helper and make the finish_recovery_op completion conditional
on whether there is an outstanding SCAN. This fixes a leak of a recovery
op when we defer while a scan is outstanding (this bug was recently
introduced by e708410542b0a52fbb29e14b76f49c94adbc0a59 and then
duplicated by 2463c6463d1ed38a2e15a0960ed1530a47851489).
Note that there is still one other time we register MAX ops: when we are
finishing backfill. There, we start one per target. But we will always
get back our reply and process it in the normal way (that old commit
did not change the timing for these).
Signed-off-by: Sage Weil <sage@redhat.com>
* refs/pull/18192/head:
qa/cephfs: test ec data pool
qa/suites/fs/basic_functional/clusters: more osds
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
This commit introduces a regression where the admin socket specified in
ceph.conf overrides command line arguments or environment variables.
This reverts commit 1630f4bbef3a3ff6efb56acbaa9b8786b945b1e6.
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
the RGWPutObj_Compress filter relies on a starting offset of 0 to ensure
that we only compress entire objects
for RGWRados::fetch_remote_obj(), we also fetch object metadata. the
replies come back with a 'Rgwx-Embedded-Metadata-Len' header, which
specifies how many bytes of object metadata are at the front of the
request body. when this is present, the offsets passed from
RGWRadosPutObj down to the RGWPutObjDataProcessor filters are offsets
into the http response body, rather than logical offsets into the object
data itself
this commit adds a transformation to RGWRadosPutObj so that only that
logical offset is visible to the RGWPutObjDataProcessor
Fixes: http://tracker.ceph.com/issues/21895
Signed-off-by: Casey Bodley <cbodley@redhat.com>
Also for the most time write_message return zero. So avoid checking
"r < 0 or r > 0" every time, it should firstly check "r == 0".
Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
Fix for:
[src/test/librbd/test_librbd.cc:2610]: (warning) %lld in format string
(no. 1) requires 'long long' but the argument type is 'unsigned long long'.
[src/test/librbd/test_librbd.cc:2610]: (warning) %lld in format string
(no. 2) requires 'long long' but the argument type is 'unsigned long long'.
[src/test/librbd/test_librbd.cc:2610]: (warning) %lld in format string
(no. 3) requires 'long long' but the argument type is 'unsigned long long'.
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
Fix for:
[src/test/librbd/test_internal.cc:1054]: (warning) %ld in format string
(no. 1) requires 'long' but the argument type is 'unsigned long'.
[src/test/librbd/test_internal.cc:1054]: (warning) %ld in format string
(no. 2) requires 'long' but the argument type is 'unsigned long'.
[src/test/librbd/test_internal.cc:1063]: (warning) %ld in format string
(no. 1) requires 'long' but the argument type is 'unsigned long'.
[src/test/librbd/test_internal.cc:1073]: (warning) %ld in format string
(no. 1) requires 'long' but the argument type is 'unsigned long'.
[src/test/librbd/test_internal.cc:1114]: (warning) %ld in format string
(no. 1) requires 'long' but the argument type is 'unsigned long'.
[src/test/librbd/test_internal.cc:1121]: (warning) %ld in format string
(no. 1) requires 'long' but the argument type is 'unsigned long'.
[src/test/librbd/test_internal.cc:1129]: (warning) %ld in format string
(no. 1) requires 'long' but the argument type is 'unsigned long'.
[src/test/librbd/test_internal.cc:1196]: (warning) %ld in format string
(no. 1) requires 'long' but the argument type is 'unsigned long'.
[src/test/librbd/test_internal.cc:1196]: (warning) %ld in format string
(no. 2) requires 'long' but the argument type is 'unsigned long'.
[src/test/librbd/test_internal.cc:1205]: (warning) %ld in format string
(no. 1) requires 'long' but the argument type is 'unsigned long'.
[src/test/librbd/test_internal.cc:1215]: (warning) %ld in format string
(no. 1) requires 'long' but the argument type is 'unsigned long'.
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
Fix for:
[src/rgw/rgw_rados.cc:3752]: (performance) Inefficient usage of
string::find() in condition; string::compare() would be faster.
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
Fix for:
[src/osd/osd_types.cc:439]: (warning) %d in format string (no. 1) requires
'int *' but the argument type is 'unsigned int *'.
Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
