This list was lacking a few fields: crash_replay_interval, pg_num,
pgp_num and crush_ruleset. Include these fields and add add short
descriptions.
Signed-Off-By: Andre Noll <maan@systemlinux.org>
The set of OSD commands which added by the previous commit is
incomplete. This patch adds documentation for the following
OSD commands which were previously missing: dump, tree, crush,
cluster_snap, lost, create, rm.
Signed-Off-By: Andre Noll <maan@systemlinux.org>
This adds the content of the wiki page at
http://ceph.newdream.net/wiki/Monitor_commands
to doc/control.rst in order to make it available at the new official
location for the ceph documentation. This first patch is just the
result of a cut-and-paste operation. There are no changes in content,
but the text was converted to rst format.
Signed-Off-By: Andre Noll <maan@systemlinux.org>
We should be setting this true when calling fault() from connect().
And rename it in the header -- it does produce less output when set,
but it's actually useful because it changes the shutdown logic
slightly.
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
This is causing issues since the Client's ack of the MClientSession
is somehow not getting back to the MDS. We should make the
Client/SimpleMessenger be better-behaved about shutdown, but for
now this solution matches how we're working elsewhere in our codebase.
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
If the (first) TAKE in the crush rule is not the root, see if they picked
a point somewhere beneath the appropriate part of the hierachy. If not,
continue with a normal mapping.
Previously we would simply error out in this case and map nothing.
Signed-off-by: Sage Weil <sage@newdream.net>
list_objects returns Object(). Object therefore now has an optional
locator_key parameter which will set up the object locator on Object()
methods to allow objects returned from list_objects with locator keys to
be used normally.
Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
--object-locator locator causes io to use the specified locator. For
objects with non-empty locators, rados pool ls will output the locator
as well.
Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
A single "url" doesn't make sense for a swift object store the way it does
for an S3 store or local file, so this commit explodes the interface from a
single url argument to three arguments: host, bucket, and object_prefix.
It also adds a SwiftStore and related classes. Syncing ACLs with swift is
not yet implemented.
The device number depends on how many rbd images have been
mapped. Removing it makes the name determined solely by the name,
image, and snapshot that are mapped, for ease of scripting or persistence
across reboots.
Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
No need for the old collection_list_partial instance: it's cleaner to
just use an hobject_t as the collection list handle.
Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
Previously, there was no way to recover the locator key used to create
and object. Now, rados_objects_list_next and ObjectIterator will return
the key as well as the object name.
Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
To match the HashIndex ordering, we need to sort hobject_t by the nibble
reversed hash. We store objects in the filestore in a directory tree
with the least significant nibble at the top and the most at the bottom
to facilitate pg splitting in the future.
Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
If data_subset is empty (i.e., the data we pulled is no longer useful),
we should mark complete false and continue rather than fail the
assert in range_end().
Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
This function is passing a reference from PG::active_rep_scrub to
the req_scrub_wq, not eliminating the reference (and the WorkQueue
doesn't grab a new reference itself, either).
The other alternative is to convert the WorkQueue to grab a
reference, but since they can cycle through the WorkQueue more than
once, and need to be ->put() outside the WorkQueue, I don't like
that option.
This should fix#1758.
Also add an assert to PG::_request_scrub_map to check on the other
possible cause of this bug (and fix the indentation).
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
The conditional before running pip install was unnecessary,
"pip install" on already installed packages is fast (as long
as it's not --upgrade), and --quiet makes it not spam the
console.
Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com>
Now you can create diagrams easily with the ".. ditaa::"
directive in the Sphinx documents.
admin/build-doc now checks for debs required for building
the documentation, or just lists commands missing for hosts
not using dpkg.
For more on Ditaa, see http://ditaa.sourceforge.net/
Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com>
