The API call is a task and the response status is determined by whether
the call is completed within a pre-defined duration (2 seconds) or not.
We should also allow the status when the call takes longer.
Fixes: https://tracker.ceph.com/issues/46812
Signed-off-by: Kiefer Chang <kiefer.chang@suse.com>
when applying a placement of '1' for mon, mgr, rbd-mirror, crash, alertmanager, grafana, node-exporter, prometheus, or mds a dry-run was activated
Fixes: https://tracker.ceph.com/issues/46819
Signed-off-by: Daniel-Pivonka <dpivonka@redhat.com>
Since e5b585d15d ceph-volume depends on
python-ceph-common. This commit introduces this dependency for the
ceph-osd rpm (which includes ceph-volume) and installs the dependency
for tox runs.
Fixes: https://tracker.ceph.com/issues/46772
Fixes: e5b585d15d
Signed-off-by: Jan Fajerski <jfajerski@suse.com>
cephadm: Fix error bootstraping with '--registry-json'
Reviewed-by: Adam King <adking@redhat.com>
Reviewed-by: Sebastian Wagner <sebastian.wagner@suse.com>
The problem only happens when a host is reported both from Ceph and the
Orchestrator. There is an error on merging Ceph hosts with Orchestrator
hosts. The unit test is refined to test this case.
Fixes: https://tracker.ceph.com/issues/46761
Signed-off-by: Kiefer Chang <kiefer.chang@suse.com>
ignore SIGHUP, as crimson does not write log messages using the
facilities offered by ceph::logging. it uses seastar::logger instead.
and seastar::logger writes to stderr, stdout and/or syslog. there is no
need to reopen the log file when the settings change.
and we do send SIGHUP to OSD processes when performing thrash tests. to
avoid killing crimson-osd processes by accident. SIGHUP is ignored.
Signed-off-by: Kefu Chai <kchai@redhat.com>
since we now depends on GCC-7 and up for C++17 support, we can now use
std::atomic<time_point> for better readability.
Signed-off-by: Kefu Chai <kchai@redhat.com>
This commit extends the object dispatch read function to support multiple object extents (using a single librados ObjectOperation).
In addition, we add an option to return the version number for that read.
This new function will be used by the new crypto object dispatch layer to support read-modify-write operation required for encrypting and writing unaligned data.
Signed-off-by: Or Ozeri <oro@il.ibm.com>
* track dependency for running tox tests with tox.ini. the fix
test is not performed by "make check", so `install-deps.sh`
is not responsible for preparing the autopep8 wheels. hence
we can just put it in tox.ini
* remove "python_version >= '3'" as we do not support python2
this change addresses a regression introduced by
7d8a3b6032, which fails the rpmbuild like:
RPM build errors:
Installed (but unpackaged) file(s) found:
/usr/share/ceph/mgr/requirements-fix.txt
Signed-off-by: Kefu Chai <kchai@redhat.com>
mgr/cephadm: Introduce tox and autopep8
Reviewed-by: Juan Miguel Olmo Martínez <jolmomar@redhat.com>
Reviewed-by: Sebastian Wagner <sebastian.wagner@suse.com>
* set the osd_stat in pg_stat message sent to mgr, otherwise the seq id
in the message would be always 0.
* return the seq id composed of up_epoch and osd_stat_seq when serving
"flush_pg_stats" tell command.
Signed-off-by: Kefu Chai <kchai@redhat.com>
extract check_old_style_section_names() out from
md_config_t::parse_config_files() into ConfFile class.
for two reasons:
* so it can be reused
* for smaller md_config_t::parse_config_files(), hence better
readability
Signed-off-by: Kefu Chai <kchai@redhat.com>
for two reasons
* to consolidate the rules for figuring out the cluster name
* to reuse it in another implementation of parse_config_files(),
crimson will need to restructure the existing parse_config_files(),
and implement an async version of it.
Signed-off-by: Kefu Chai <kchai@redhat.com>
instead of passing the output parameter just let it return the section
names. C++17 enfoces the copy elision in this case, so no need to pass
the output parameter for saving the overhead of creating an temporary
object and copying it.
Signed-off-by: Kefu Chai <kchai@redhat.com>
md_config_t::parse_config_files() performs blocked i/o, and it might be
detected by the stall detector of seastar, and it spills out warning
messages when it sees the hang.
so i plan to implement an asynchronized parse_config_files() for
crimson, and reuse the synchronized part from the blocking
implementation of parse_config_files(). this is the first step.
Signed-off-by: Kefu Chai <kchai@redhat.com>
