... to avoid leaving log events that reference log
segments by offsets which no longer exist.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 386f2d7c82)
Reviewed-by: Greg Farnum <greg@inktank.com>
Using a stringstream that is only displayed on error when calling the
erasure code factory, instead of cerr. The user expects the output to be
clean when there is no error.
http://tracker.ceph.com/issues/9429Fixes: #9429
Signed-off-by: Loic Dachary <loic-201408@dachary.org>
If we have nolaggy waiters but are laggy we want to sleep. Otherwise,
we will just spin and spam the log ...
Signed-off-by: Sage Weil <sage@redhat.com>
Since the erasure code plugin version check has been introduced,
whenever a library/binary that can load plugin needs to be recompiled,
the erasure code plugins must also be considered. If the reason for
recompiling the library/binary is a new commit, the plugins will fail to
load.
The dependency is not based on source compilation and a shared library
dependency on liberasure-code.la is added instead. This library is
uniformly used whenever a plugin is to be loaded and therefore covers
all library/binaries that need it.
http://tracker.ceph.com/issues/9413Fixes: #9413
Signed-off-by: Loic Dachary <loic-201408@dachary.org>
When access key contains ':', e.g. `some_info:for_user',
authorization header looks like:
"AWS some_info:for_user:request_signature"
so `auth_str.find(':')` result in auth_id = "some_info",
auth_sign = "for_user:request_signature".
auth_str.rfind(':') solves this issue.
Signed-off-by: Roman Haritonov <reclosedev@gmail.com>
When replaying EImportFinish/EFragment event, the replay thread may call
MDS::queue_waiters. MDS::queue_waiters() requires its caller to hold the
mds_lock. Otherwise assert(waiter_mutex == __null || waiter_mutex->is_locked())
in Cond::Signal() will be tiggered.
Signed-off-by: Yan, Zheng <zyan@redhat.com>
When upgrading from a build without the promotion on 2nd read feature,
should set min_read_recency_for_promote to the default value 1, instead
of 0.
Signed-off-by: Zhiqiang Wang <wonzhq@hotmail.com>
Currently in CrushWrapper, the member "struct crush_map *crush" is a public member,
so people can break the encapsulation and manipulate directly to the crush structure.
This is not a good practice for encapsulation and will lead to inconsistent if code
mix use the CrushWrapper API and crush C API.A simple example could be:
1.some code use crush_add_rule(C-API) to add a rule, which will not set the have_rmap flag to false in CrushWrapper
2.another code using CrushWrapper trying to look up the newly added rule by name will get a -ENOENT.
This patch move CrushWrapper::crush to private, together with three reverse map(type_rmap, name_rmap, rule_name_rmap)
and also change codes accessing the CrushWrapper::crush to make it compile.
Signed-off-by: Xiaoxi Chen <xiaoxi.chen@intel.com>
This is a bad assert. Specifically, handle_osd_op_reply may still be
holding the session ref while it is calling the completion for a previous
request. This is safe: it is only holding the session ref after it dropped
the global map rwlock because of the per-session completion locks. The
request in question was already marked completed by the time our thread
took the session lock.
Fixes: #9241
Signed-off-by: Sage Weil <sage@redhat.com>
If we cancel a read, revoke the rx buffers to avoid a use-after-free and/or
other undefined badness by using user buffers that may no longer be
present.
Fixes: #9362
Backport: firefly, dumpling
Reported-by: Matthias Kiefer <matthias.kiefer@1und1.de>
Signed-off-by: Sage Weil <sage@redhat.com>
Verify we don't receive data after a timeout.
Based on reproducer for #9362 written by
Matthias Kiefer <matthias.kiefer@1und1.de>.
Signed-off-by: Sage Weil <sage@redhat.com>
