* refs/pull/32806/head:
common/bl: fix accessibility of bptr's _off and _len fields.
common/bl: drop get_raw() from the public buffer::ptr interface.
common: drop sharing of buffer::raw outside bufferlist.
Reviewed-by: Kefu Chai <kchai@redhat.com>
mgr/dashboard: Add Modified filter to Configuration table
Reviewed-by: Stephan Müller <smueller@suse.com>
Reviewed-by: Tatjana Dehler <tdehler@suse.com>
Avoid leaking connections that had partially-consumed
client data on unexpected disconnect.
Resolves CVE-2020-1700 (moderate impact flaw).
Fixes: https://tracker.ceph.com/issues/42531
Signed-off-by: Or Friedmann <ofriedma@redhat.com>
Signed-off-by: Matt Benjamin <mbenjamin@redhat.com>
This also makes `_cancel_jobs()` thread safe, which was not the
case earlier (with `_cancel_purge_job()`) -- this also makes the
code simpler by sharing the lock betweent two condition variables.
Signed-off-by: Venky Shankar <vshankar@redhat.com>
This will be required when creating a clone as the clone would
inherit source subvolumes creation mode and uid/gid.
Signed-off-by: Venky Shankar <vshankar@redhat.com>
we are using spdk/dpdk for async msgr's dpdk backend since 01a9f178. so
no need to keep dpdk submodule anymore.
Fixes: http://tracker.ceph.com/issues/24032
Signed-off-by: Kefu Chai <kchai@redhat.com>
* refs/pull/32981/head:
mon: elector: return after triggering a new election
Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: Neha Ojha <nojha@redhat.com>
This reverts commit cb8087dfac.
The seastar submodule's .gitmodules links to `../dpdk` which is no longer present after removing dpdk from ceph.git's .gitmodules.
```
<dwfreed> the ceph/seastar repo uses awful URLs for the submodules
<dwfreed> and those awful URLs are the real reason it's failing
<dwfreed> dgalloway: ^^^
<dwfreed> seastar's .gitmodules references repos in the parent directory, so that when it's checked out as a submodule of ceph, you don't download the repos twice (and git will probably also use references instead of duplicating the local .git); however, ceph doesn't have a submodule for dpdk anymore
<dwfreed> so seastar's referencing a dpdk repo that doesn't exist
<dgalloway> i think i follow. so you're suggesting revert cb8087dfac ?
<dwfreed> yep
<dwfreed> that'd be one way to fix it
...
<joshd> dgalloway: I'd suggest revert for now, and let the crimson folks figure out the longer term fix when they're back
```
Signed-off-by: David Galloway <dgallowa@redhat.com>
The self._health_slot value needs to be updated too if the current key
is removed.
Fixes: https://tracker.ceph.com/issues/43886
Signed-off-by: Sage Weil <sage@redhat.com>
This is harmless if logging is low, but adds useful info when it is turned
up.
Hunting bug https://tracker.ceph.com/issues/43914
Signed-off-by: Sage Weil <sage@redhat.com>
If we haven't scrubbed everything, we occasinoally re-request scrub in case
the request was missed by the OSD (this can happen). But we were
re-requesting scrub on ALL pgs, and if they are done in a
semi-deterministic order and are slow, then we may never get to the final
ones.
Signed-off-by: Sage Weil <sage@redhat.com>
