The MDS Autoscaler provides automation to maintain a required count of
MDSs in the system.
The MDS count in the system can be afected by:
* max_mds config option
* standby_count_wanted config option
* death of an active MDS Rank
Fixes: https://tracker.ceph.com/issues/46884
Signed-off-by: Milind Changire <mchangir@redhat.com>
* refs/pull/37529/head:
qa: set rados op timeouts for mds/ceph-fuse
qa: print debug info on mount cleanup
qa: remove redundant rmr
qa: use null mode to prevent undesired changes to mountpoint
qa: unmount all clients before deleting the file system
osdc: add timeout configs for mons/osds
common: accept timespan for SaferCond.wait_for
Reviewed-by: Neha Ojha <nojha@redhat.com>
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
This way, we can do a bulk scan of the store without building up
an unbounded amount of state in Transaction::read_set. Note that
such transactions will not be snapshot isolated.
Signed-off-by: Samuel Just <sjust@redhat.com>
Currently, we need to rely on the Transaction::read_set to ensure cache
residence of a leaf node until TransactionManager adds the logical
extent to the cache. The next patch, however, will introduce a lazy
flag for Transaction's to enable doing snapshot inconsistent scans
without populating the read_set, so we'll want this to work without
it.
Signed-off-by: Samuel Just <sjust@redhat.com>
Add request timeout beast
The beast frontend will use the same parameter as the civetweb one, request_timeout_ms and will be configured to 65 seconds by default
Fixes: https://tracker.ceph.com/issues/45431
Signed-off-by: Or Friedmann <ofriedma@redhat.com>
The Dashboard can get cluster information from the Orchestrator.
For settings that are set by previous revisions, the Dashboard will
check them and ask user to remove them.
Signed-off-by: Kiefer Chang <kiefer.chang@suse.com>
This change make the Dashboard support two types of Ganesha clusters:
- Orchestrator clusters (Since Octopus)
- Deployed by the Orchestrator.
- The Dashboard gets the pool/namespace that stores Ganesha
configuration objects from the Orchestrator.
- The Dashboard gets the daemons in a cluster from the Orchestrator.
- User-defined clusters (Since Nautilus)
- Clusters defined by using `ceph dashboard
set-ganesha-clusters-rados-pool-namespace` command is treated as
user-defined clusters.
- Each daemon has its own RADOS configuration objects. The
Dashboard uses these objects to deduce daemons.
Fixes: https://tracker.ceph.com/issues/46492
Signed-off-by: Kiefer Chang <kiefer.chang@suse.com>
After deciding to always enable tracking log in early phase, there's no
need to keep "log_early" option here and remove it directly.
Suggested-by: Kefu Chai <kefu@redhat.com>
Signed-off-by: Changcheng Liu <changcheng.liu@aliyun.com>
When notifying the journal recorder of an overflow or if the object
close request has completed due to no more in-flight IO, it was
possible for a race between a flush request and the processing of
an append completion to attempt to kick off duplicate notifications.
Since the overflowed and closed callbacks are properly protected from
duplicates, use a counter instead of a boolean to track possible
in-flight handler callbacks.
Fixes: https://tracker.ceph.com/issues/47880
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
otherwise we could call do_until() recursively if we have other tasks
which need to prempt the reactor and current future's state is actually
always available.
Signed-off-by: Kefu Chai <kchai@redhat.com>
before this change, in our specialization of seastar::do_until(),
we access `f` after calling `f.get()`, this is not correct. as `f.get()`
actually moves `f._state` away and detaches the associated promise if any.
so we cannot call `f._then()` anymore after calling `f.get()`. as
`f._then()` schedules `f` by detaching the future from promise and
attaching the scheduled task to the promise. but `future_base::detach_promise()`
does not check `_promise` before accessing it, hence the segfault.
after this change, the order of the checks is rearranged so that
`f.get()` is called at the end. and also use `f.get0()` to be more
explicit, as we are accessing the only element of the returned
value.
Signed-off-by: Kefu Chai <kchai@redhat.com>
The teuthology testing for rgw-orphan-list needs to install
`s3cmd`. Switch from using dnf to yum to work on a wider variety of
platforms.
Signed-off-by: J. Eric Ivancich <ivancich@redhat.com>
Since when starting up the datalog we check whether Omap and FIFO
backed versions already exist, don't fill the error log with
complaints if they happen not to.
Signed-off-by: Adam C. Emerson <aemerson@redhat.com>
