]> git.apps.os.sepia.ceph.com Git - ceph.git/log
ceph.git
13 years agoradosgw-admin: add --purge-keys option
Yehuda Sadeh [Tue, 20 Dec 2011 19:00:39 +0000 (11:00 -0800)]
radosgw-admin: add --purge-keys option

13 years agorgw: add timeout to init path
Yehuda Sadeh [Tue, 20 Dec 2011 00:00:27 +0000 (16:00 -0800)]
rgw: add timeout to init path

13 years agoMOSDRepScrub,ReplicatedPG: Add scrub_to to MOSDRepScrub
Samuel Just [Mon, 19 Dec 2011 22:50:17 +0000 (14:50 -0800)]
MOSDRepScrub,ReplicatedPG: Add scrub_to to MOSDRepScrub

When scrub_from is set, also set scrub_to to the primary's
last_update_applied (which will also be the official last_update before
finalizing scrub began).  The replica instead of waiting for
last_update_applied to catch up to last_update will wait for
last_update_applied to catch up to active_rep_scrub->scrub_to.  This
avoids a race where the replica scrub is requeued before all of the
currently queued sub-ops have been processed.

Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
13 years agomon: update man page to document --mkfs stuff
Sage Weil [Mon, 19 Dec 2011 21:47:43 +0000 (13:47 -0800)]
mon: update man page to document --mkfs stuff

Signed-off-by: Sage Weil <sage@newdream.net>
13 years agomds: misc assertions about truncation
Sage Weil [Wed, 30 Nov 2011 01:21:42 +0000 (17:21 -0800)]
mds: misc assertions about truncation

Signed-off-by: Sage Weil <sage@newdream.net>
13 years agoosd: EINVAL on truncate to huge object size
Sage Weil [Wed, 30 Nov 2011 01:17:25 +0000 (17:17 -0800)]
osd: EINVAL on truncate to huge object size

Signed-off-by: Sage Weil <sage@newdream.net>
13 years agoMerge pull request #6 from kylemarsh/wip-obsync-swift
Sage Weil [Mon, 19 Dec 2011 22:00:59 +0000 (14:00 -0800)]
Merge pull request #6 from kylemarsh/wip-obsync-swift

Wip obsync swift

13 years agorgw: remove temp context in prepare_get_obj
Yehuda Sadeh [Mon, 19 Dec 2011 21:54:04 +0000 (13:54 -0800)]
rgw: remove temp context in prepare_get_obj

13 years agorgw: fix xml parser internal structure leak
Yehuda Sadeh [Mon, 19 Dec 2011 21:38:58 +0000 (13:38 -0800)]
rgw: fix xml parser internal structure leak

13 years agorgw: fix a leak of acl structure (in req_state)
Yehuda Sadeh [Mon, 19 Dec 2011 21:38:40 +0000 (13:38 -0800)]
rgw: fix a leak of acl structure (in req_state)

13 years agoMerge branch 'wip-osd-maybe-created'
Greg Farnum [Mon, 19 Dec 2011 21:25:27 +0000 (13:25 -0800)]
Merge branch 'wip-osd-maybe-created'

13 years agoMerge branch 'wip-osd-fsid'
Sage Weil [Mon, 19 Dec 2011 21:24:46 +0000 (13:24 -0800)]
Merge branch 'wip-osd-fsid'

13 years agodoc: fix typo
Sage Weil [Mon, 19 Dec 2011 21:24:31 +0000 (13:24 -0800)]
doc: fix typo

Signed-off-by: Sage Weil <sage@newdream.net>
13 years agoosd: --get-journal-fsid
Sage Weil [Wed, 30 Nov 2011 22:13:47 +0000 (14:13 -0800)]
osd: --get-journal-fsid

Signed-off-by: Sage Weil <sage@newdream.net>
13 years agofilejournal: uuid for fsid
Sage Weil [Wed, 30 Nov 2011 22:13:14 +0000 (14:13 -0800)]
filejournal: uuid for fsid

Decode old header struct, but encode new class using more normal encoding
style.  Embed in a bufferlist for later extensibility.  Use the first
64 bits of the uuid for the per-entry magic, as before.

When we load an old fsid, we mirror the 64-bits.

Signed-off-by: Sage Weil <sage@newdream.net>
13 years agofilestore: make fsid uuid_d instead of uint64_t
Sage Weil [Wed, 30 Nov 2011 05:49:49 +0000 (21:49 -0800)]
filestore: make fsid uuid_d instead of uint64_t

Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
13 years agoosd: --get-osd-fsid and --get-cluster-fsid
Sage Weil [Wed, 30 Nov 2011 05:06:29 +0000 (21:06 -0800)]
osd: --get-osd-fsid and --get-cluster-fsid

Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
13 years agoosd: store osd_fsid as text in osd_data dir
Sage Weil [Wed, 30 Nov 2011 05:06:10 +0000 (21:06 -0800)]
osd: store osd_fsid as text in osd_data dir

along with ceph_fsid (the cluster fsid) and a few other things.

Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
13 years agoosd: include osd_fsid in OSDSuperblock
Sage Weil [Wed, 30 Nov 2011 05:05:20 +0000 (21:05 -0800)]
osd: include osd_fsid in OSDSuperblock

Generated during mkfs.

Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
13 years agoosd: rename OSDSuperblock::fsid -> cluster_fsid
Sage Weil [Wed, 30 Nov 2011 04:46:42 +0000 (20:46 -0800)]
osd: rename OSDSuperblock::fsid -> cluster_fsid

Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
13 years agodoc: fix mon cluster expansion docs
Sage Weil [Mon, 19 Dec 2011 16:04:53 +0000 (08:04 -0800)]
doc: fix mon cluster expansion docs

Signed-off-by: Sage Weil <sage@newdream.net>
13 years agomon: pull addr from ceph.conf, mon_host as needed when joining mon cluster
Sage Weil [Mon, 19 Dec 2011 16:03:59 +0000 (08:03 -0800)]
mon: pull addr from ceph.conf, mon_host as needed when joining mon cluster

Signed-off-by: Sage Weil <sage@newdream.net>
13 years agomon: fix setting of mon addr when joining a cluster
Sage Weil [Mon, 19 Dec 2011 15:57:45 +0000 (07:57 -0800)]
mon: fix setting of mon addr when joining a cluster

Signed-off-by: Sage Weil <sage@newdream.net>
13 years agoman: Update the configuration example for radosgw
Wido den Hollander [Fri, 16 Dec 2011 21:01:05 +0000 (22:01 +0100)]
man: Update the configuration example for radosgw

Signed-off-by: Wido den Hollander <wido@widodh.nl>
13 years agoman: It is capital -C instead of -c when for creating a new keyring
Wido den Hollander [Fri, 16 Dec 2011 21:00:46 +0000 (22:00 +0100)]
man: It is capital -C instead of -c when for creating a new keyring

Signed-off-by: Wido den Hollander <wido@widodh.nl>
13 years agorgw: fix updating of object metadata
Yehuda Sadeh [Sat, 17 Dec 2011 06:04:43 +0000 (22:04 -0800)]
rgw: fix updating of object metadata

being used in swift POST. We were updating wrong object
size and etag

13 years agorgw: bucket cannot be recreated if already exists
Yehuda Sadeh [Sat, 17 Dec 2011 01:13:53 +0000 (17:13 -0800)]
rgw: bucket cannot be recreated if already exists

13 years agoobsync: add authurl to CLI 6/head
Kyle Marsh [Sat, 17 Dec 2011 00:05:46 +0000 (16:05 -0800)]
obsync: add authurl to CLI

s3 connections require the hostname and swift connections require the
authurl.  obsync treats these as equivalent internally, but breaks them
apart on the command line interface for clarity for the users.

13 years agoobject.h: initialize max in hobject_t(sobject_t) constructor
Samuel Just [Fri, 16 Dec 2011 22:42:36 +0000 (14:42 -0800)]
object.h: initialize max in hobject_t(sobject_t) constructor

Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
13 years agoReplicatedPG: fix handle_watch_timeout ctx->at_version
Samuel Just [Fri, 16 Dec 2011 19:30:46 +0000 (11:30 -0800)]
ReplicatedPG: fix handle_watch_timeout ctx->at_version

ctx->at_version should match the head of the new log entries
during issue_repop.  This could cause the scrub hang bug as
last_update would be less than last_update_applied.

Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
13 years agoReplicatedPG: add asserts to catch scrub error
Samuel Just [Fri, 16 Dec 2011 19:22:58 +0000 (11:22 -0800)]
ReplicatedPG: add asserts to catch scrub error

If last_update_applied skipped over last_update, we would see
scrub hang.

Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
13 years agodoc: fix filename in mon addition process
Sage Weil [Fri, 16 Dec 2011 18:39:33 +0000 (10:39 -0800)]
doc: fix filename in mon addition process

Signed-off-by: Sage Weil <sage@newdream.net>
13 years agofilejournal: preallocate journal bytes on create
Sage Weil [Fri, 16 Dec 2011 17:19:59 +0000 (09:19 -0800)]
filejournal: preallocate journal bytes on create

This should reduce fragmentation for large journals that are written
slowly the first time around.

Signed-off-by: Sage Weil <sage@newdream.net>
13 years agoMerge pull request #5 from homac/master
Sage Weil [Fri, 16 Dec 2011 17:08:17 +0000 (09:08 -0800)]
Merge pull request #5 from homac/master

Minor fix for init files and cleaned up spec file. Please pull

13 years agoman: clarify mount.ceph auth options
Josh Durgin [Thu, 15 Dec 2011 22:03:00 +0000 (14:03 -0800)]
man: clarify mount.ceph auth options

Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
13 years agoman: update rule definition for ceph-rbdnamer
Josh Durgin [Thu, 15 Dec 2011 21:49:10 +0000 (13:49 -0800)]
man: update rule definition for ceph-rbdnamer

This is the rule we install since 891025e539a92b5d75011e2e75c475fc0c272042.

Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
13 years agoauthx -> cephx everywhere it's used
Josh Durgin [Thu, 15 Dec 2011 21:41:56 +0000 (13:41 -0800)]
authx -> cephx everywhere it's used

The term authx was in the mount.ceph man page, and got accidentally
copied into rbd help.

Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
13 years agoceph.spec: Clean up and fix spec file and build for a couple of distributions 5/head
Holger Macht [Thu, 15 Dec 2011 16:51:04 +0000 (17:51 +0100)]
ceph.spec: Clean up and fix spec file and build for a couple of distributions

Clean up and fix the spec file. This includes cleaning up of build and
installed system dependencies, LSB compliance fixes, splitting up into
several sub-packages (lib*) and so on. It now builds fine for the
following distributions in the Open Build Service and should be
considered as a starting point for further fixes:

 - CentOS 6
 - Fedora 15
 - RedHat Enterprise Linux 6
 - openSUSE 11.4
 - openSUSE 12.1
 - openSUSE Factory
 - SUSE Linux Enterprise 11 (SP1 and SP2)

Signed-off-by: Holger Macht <hmacht@suse.de>
13 years agoinit-ceph/init-radosgw: Don't use unspecified runlevel 4
Holger Macht [Thu, 15 Dec 2011 16:49:15 +0000 (17:49 +0100)]
init-ceph/init-radosgw: Don't use unspecified runlevel 4

Don't use runlevel 4 in init scripts. AFAIK, no distribution is using it
and at least the Open Build Service complains about it.

Signed-off-by: Holger Macht <hmacht@suse.de>
13 years agoclient: fix logger deregistration
Sage Weil [Thu, 15 Dec 2011 01:43:18 +0000 (17:43 -0800)]
client: fix logger deregistration

Only unregister logger if it is non-NULL (and thus registered) to avoid
running afoul of the cct assertions.

Signed-off-by: Sage Weil <sage@newdream.net>
13 years agoobsync: add vvprint back in
Kyle Marsh [Wed, 14 Dec 2011 23:37:47 +0000 (15:37 -0800)]
obsync: add vvprint back in

Commit ebe5fc60d20f92a0037c53c1e7bd7ae512be3da4 removed the definition of
vvprint without removint all the places that used it.

13 years agodoc: Make overview.rst valid reStructuredText, so I can stop seeing warnings.
Tommi Virtanen [Wed, 14 Dec 2011 19:38:08 +0000 (11:38 -0800)]
doc: Make overview.rst valid reStructuredText, so I can stop seeing warnings.

It's still wrong, but now it won't clutter the output.

Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com>
13 years agopybind: Add a description to docstring.
Tommi Virtanen [Wed, 14 Dec 2011 19:32:34 +0000 (11:32 -0800)]
pybind: Add a description to docstring.

This avoids a Sphinx warning like this:

.../src/pybind/rbd.py:docstring of rbd.RBD.version:2: WARNING: Field list ends without a blank line; unexpected unindent.

Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com>
13 years agodoc: reStructuredText syntax fix.
Tommi Virtanen [Wed, 14 Dec 2011 19:25:40 +0000 (11:25 -0800)]
doc: reStructuredText syntax fix.

Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com>
13 years agoMerge branch 'wip-messenger'
Greg Farnum [Wed, 14 Dec 2011 19:14:02 +0000 (11:14 -0800)]
Merge branch 'wip-messenger'

13 years agodoc: "ceph -w" clarification.
Tommi Virtanen [Wed, 14 Dec 2011 19:18:21 +0000 (11:18 -0800)]
doc: "ceph -w" clarification.

Stop saying "watch cluster state" so many times.

Don't say stdout, that's the assumption.

Don't call showing things on terminal "printing".

Don't show -C option here.

Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com>
13 years agodoc: Fix sentence that ended too abruptly.
Tommi Virtanen [Wed, 14 Dec 2011 19:14:41 +0000 (11:14 -0800)]
doc: Fix sentence that ended too abruptly.

Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com>
13 years agodoc: Split intro to talk about the DFS separately. Mention petabytes.
Tommi Virtanen [Wed, 14 Dec 2011 19:13:57 +0000 (11:13 -0800)]
doc: Split intro to talk about the DFS separately. Mention petabytes.

Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com>
13 years agodoc: Add more missing commands to control.
Tommi Virtanen [Wed, 14 Dec 2011 18:48:25 +0000 (10:48 -0800)]
doc: Add more missing commands to control.

This is too unstructured, that will have to be fixed later.

Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com>
13 years agodoc: Add misc explanations of Ceph internals from email.
Tommi Virtanen [Wed, 14 Dec 2011 18:47:38 +0000 (10:47 -0800)]
doc: Add misc explanations of Ceph internals from email.

Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com>
13 years agodoc: Use todo directive, rescue list of missing commands from wiki.
Tommi Virtanen [Wed, 14 Dec 2011 17:38:59 +0000 (09:38 -0800)]
doc: Use todo directive, rescue list of missing commands from wiki.

Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com>
13 years agodoc: Link to manpage when command is mentioned.
Tommi Virtanen [Wed, 14 Dec 2011 17:36:01 +0000 (09:36 -0800)]
doc: Link to manpage when command is mentioned.

Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com>
13 years agodoc: More consistency on formatting placeholder names.
Tommi Virtanen [Wed, 14 Dec 2011 17:33:50 +0000 (09:33 -0800)]
doc: More consistency on formatting placeholder names.

Signed-off-by: Tommi Virtanen <tommi.virtanen@dreamhost.com>
13 years agomsgr: unset did_bind in stop().
Greg Farnum [Wed, 14 Dec 2011 19:11:37 +0000 (11:11 -0800)]
msgr: unset did_bind in stop().

We use did_bind as a flag on whether or not to stop the Accepter thread
and we should clear it when we do the stopping.
This fixes #1829.

Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
13 years agoobjecter: fix use-after-free
Sage Weil [Wed, 14 Dec 2011 18:59:51 +0000 (10:59 -0800)]
objecter: fix use-after-free

messenger consumes the m reference.  Yay valgrind.

Signed-off-by: Sage Weil <sage@newdream.net>
13 years agoclient: move PerfCounter into Client
Sage Weil [Wed, 14 Dec 2011 18:49:10 +0000 (10:49 -0800)]
client: move PerfCounter into Client

globals are evil.

Fixes: #1826
Signed-off-by: Sage Weil <sage@newdream.net>
13 years agoswift: auth response returns X-Auth-Token instead of X-Storage-Token
Yehuda Sadeh [Wed, 14 Dec 2011 18:49:42 +0000 (10:49 -0800)]
swift: auth response returns X-Auth-Token instead of X-Storage-Token

13 years agoosd: fix build_incremental_map_msg
Sage Weil [Wed, 14 Dec 2011 17:31:17 +0000 (09:31 -0800)]
osd: fix build_incremental_map_msg

We keep both the inc and the full for our oldest osdmap.

Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
13 years agoosd: clean up _delete_head
Sage Weil [Tue, 13 Dec 2011 23:04:05 +0000 (15:04 -0800)]
osd: clean up _delete_head

Might be fixing a subtle logic bug, but old flow was confusing, so not
sure.  :)

Signed-off-by: Sage Weil <sage@newdream.net>
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
13 years agoosd: simplify creation logic in do_osd_ops
Sage Weil [Tue, 13 Dec 2011 23:03:36 +0000 (15:03 -0800)]
osd: simplify creation logic in do_osd_ops

Drop the maybe_created variable, and track exists over the course of the
transaction.

Fixes: #1825
Signed-off-by: Sage Weil <sage@newdream.net>
Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
13 years agoqa: trivial_libceph test
Sage Weil [Wed, 14 Dec 2011 00:31:02 +0000 (16:31 -0800)]
qa: trivial_libceph test

This currently fails... see #1827

Signed-off-by: Sage Weil <sage@newdream.net>
13 years agolibceph: add missing #includes
Sage Weil [Wed, 14 Dec 2011 00:28:50 +0000 (16:28 -0800)]
libceph: add missing #includes

Signed-off-by: Sage Weil <sage@newdream.net>
13 years agolibceph: catch errors from Client::init()
Sage Weil [Wed, 14 Dec 2011 00:28:38 +0000 (16:28 -0800)]
libceph: catch errors from Client::init()

And clean up error paths.

Signed-off-by: Sage Weil <sage@newdream.net>
13 years agoclient: return errors from init
Sage Weil [Wed, 14 Dec 2011 00:28:18 +0000 (16:28 -0800)]
client: return errors from init

Signed-off-by: Sage Weil <sage@newdream.net>
13 years agofilestore: dump transaction to log if we hit an error
Sage Weil [Tue, 13 Dec 2011 17:07:36 +0000 (09:07 -0800)]
filestore: dump transaction to log if we hit an error

This will let us see which operation in the transaction failed.

Signed-off-by: Sage Weil <sage@newdream.net>
13 years agoobjectstore: implement Transaction::dump()
Sage Weil [Tue, 13 Dec 2011 17:05:37 +0000 (09:05 -0800)]
objectstore: implement Transaction::dump()

Signed-off-by: Sage Weil <sage@newdream.net>
13 years agoobjectstore: create Transaction::iterator class
Sage Weil [Tue, 13 Dec 2011 17:05:10 +0000 (09:05 -0800)]
objectstore: create Transaction::iterator class

Remove iterator state from Transaction itself.

Signed-off-by: Sage Weil <sage@newdream.net>
13 years agoqa: rados load-gen workunits
Sage Weil [Tue, 13 Dec 2011 16:19:37 +0000 (08:19 -0800)]
qa: rados load-gen workunits

Signed-off-by: Sage Weil <sage@newdream.net>
13 years agopybind/rados: add asynchronous write,append,read,write_full operations
Samuel Just [Sat, 10 Dec 2011 01:55:57 +0000 (17:55 -0800)]
pybind/rados: add asynchronous write,append,read,write_full operations

Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
13 years agodoc: Clarify documentation of reweight command.
Andre Noll [Fri, 9 Dec 2011 17:57:45 +0000 (18:57 +0100)]
doc: Clarify documentation of reweight command.

This caused some discussions on the mailing list, so let's try to be clear
about the meaning of an OSD weight.

Signed-Off-By: Andre Noll <maan@systemlinux.org>
13 years agodoc: Add missing documentation for osd pool get.
Andre Noll [Fri, 9 Dec 2011 17:57:44 +0000 (18:57 +0100)]
doc: Add missing documentation for osd pool get.

"osd pool set" was already documented, but the corresponding "get"
command was not. This patch adds the list of valid fields for this
command, together with short descriptions.

Thanks to Gregory Farnum for pointing out a flaw in the description
of lpl_num and lpgp_num in an earlier version of the patch.

Signed-Off-By: Andre Noll <maan@systemlinux.org>
13 years agodoc: Update the list of fields for the pool set command.
Andre Noll [Fri, 9 Dec 2011 17:57:43 +0000 (18:57 +0100)]
doc: Update the list of fields for the pool set command.

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>
13 years agodoc: Document pause and unpause osd commands.
Andre Noll [Fri, 9 Dec 2011 17:57:42 +0000 (18:57 +0100)]
doc: Document pause and unpause osd commands.

These two commands were undocumented so far. This patch adds a short
description.

Signed-Off-By: Andre Noll <maan@systemlinux.org>
13 years agodoc: Add documentation of missing osd commands.
Andre Noll [Fri, 9 Dec 2011 17:57:41 +0000 (18:57 +0100)]
doc: Add documentation of missing osd commands.

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>
13 years agodoc: Import the list of ceph subcommands from wiki.
Andre Noll [Fri, 9 Dec 2011 17:57:40 +0000 (18:57 +0100)]
doc: Import the list of ceph subcommands from wiki.

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>
13 years agomessenger: fix up fault()'s "onconnect" parameter.
Greg Farnum [Thu, 8 Dec 2011 00:44:28 +0000 (16:44 -0800)]
messenger: fix up fault()'s "onconnect" parameter.

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>
13 years agomds: mark_disposable when closing a Client connection.
Greg Farnum [Wed, 7 Dec 2011 22:19:23 +0000 (14:19 -0800)]
mds: mark_disposable when closing a Client connection.

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>
13 years agoSimpleMessenger: remove void send_keepalive.
Greg Farnum [Wed, 7 Dec 2011 18:56:01 +0000 (10:56 -0800)]
SimpleMessenger: remove void send_keepalive.

Nobody uses this; they all call the version that returns an int.

Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
13 years agoMerge branch 'wip-mon-stats'
Sage Weil [Mon, 12 Dec 2011 17:29:36 +0000 (09:29 -0800)]
Merge branch 'wip-mon-stats'

13 years agomon: add mds, mon info to cluster_logger
Sage Weil [Thu, 8 Dec 2011 21:13:29 +0000 (13:13 -0800)]
mon: add mds, mon info to cluster_logger

Signed-off-by: Sage Weil <sage@newdream.net>
13 years agomdsmap: rename get_num_*_mds() methods
Sage Weil [Thu, 8 Dec 2011 21:02:12 +0000 (13:02 -0800)]
mdsmap: rename get_num_*_mds() methods

Signed-off-by: Sage Weil <sage@newdream.net>
13 years agomon: report basic cluster stats via perfcounters
Sage Weil [Thu, 8 Dec 2011 20:53:14 +0000 (12:53 -0800)]
mon: report basic cluster stats via perfcounters

These are basic point-in-time cluster stats.

Signed-off-by: Sage Weil <sage@newdream.net>
13 years agocrush: drop unused label
Sage Weil [Mon, 12 Dec 2011 17:22:49 +0000 (09:22 -0800)]
crush: drop unused label

Signed-off-by: Sage Weil <sage@newdream.net>
13 years agoMerge remote branch 'gh/stable'
Sage Weil [Mon, 12 Dec 2011 17:20:26 +0000 (09:20 -0800)]
Merge remote branch 'gh/stable'

13 years agocrush: fix force to behave with non-root TAKE
Sage Weil [Wed, 7 Dec 2011 17:07:38 +0000 (09:07 -0800)]
crush: fix force to behave with non-root TAKE

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>
13 years agocrush: simplify force argument check
Sage Weil [Wed, 7 Dec 2011 17:09:32 +0000 (09:09 -0800)]
crush: simplify force argument check

force isn't used past this point, only force_pos.  Collapse the if
conditions.

Signed-off-by: Sage Weil <sage@newdream.net>
13 years agoworkunits: print tests pjd runs
Josh Durgin [Sat, 10 Dec 2011 00:31:07 +0000 (16:31 -0800)]
workunits: print tests pjd runs

This will tell us which ones actually failed within a test suite.

Signed-off-by: Josh Durgin <josh.durgin@dreamhost.com>
13 years agoMerge branch 'wip_pgls'
Samuel Just [Fri, 9 Dec 2011 23:23:32 +0000 (15:23 -0800)]
Merge branch 'wip_pgls'

13 years agopybind: add object locator support to pybind pool listing
Samuel Just [Fri, 9 Dec 2011 22:42:21 +0000 (14:42 -0800)]
pybind: add object locator support to pybind pool listing

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>
13 years agorados.cc: add --object-locator and object locator output to ls
Samuel Just [Thu, 8 Dec 2011 01:06:01 +0000 (17:06 -0800)]
rados.cc: add --object-locator and object locator output to ls

--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>
13 years agoReplicatedPG: collection_list_handle_t is now an hobject_t
Samuel Just [Sat, 3 Dec 2011 01:06:59 +0000 (17:06 -0800)]
ReplicatedPG: collection_list_handle_t is now an hobject_t

Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
13 years agoosd: delay pg list on a snapid until missing is empty
Samuel Just [Fri, 2 Dec 2011 23:42:01 +0000 (15:42 -0800)]
osd: delay pg list on a snapid until missing is empty

We cannot determine from the missing set whether an object existed
at a given snap.

Signed-off-by: Samuel Just <samuel.just@dreamhost.com>
13 years agoobsync: add swift support to obsync
Kyle Marsh [Thu, 8 Dec 2011 22:30:02 +0000 (14:30 -0800)]
obsync: add swift support to obsync

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.

13 years agomsgr: turn up socket debug printouts
Greg Farnum [Fri, 9 Dec 2011 01:45:47 +0000 (17:45 -0800)]
msgr: turn up socket debug printouts

These shouldn't be too common and will help in debugging
socket leaks.

Signed-off-by: Greg Farnum <gregory.farnum@dreamhost.com>
13 years agoudev: drop device number from name
Josh Durgin [Fri, 9 Dec 2011 00:36:45 +0000 (16:36 -0800)]
udev: drop device number from name

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>
13 years agopybind: trivial fix of missing argument
Henry C Chang [Thu, 8 Dec 2011 14:35:07 +0000 (22:35 +0800)]
pybind: trivial fix of missing argument

Signed-off-by: Henry C Chang <henry.cy.chang@gmail.com>
13 years agocrush: whitespace
Sage Weil [Thu, 8 Dec 2011 16:42:59 +0000 (08:42 -0800)]
crush: whitespace

Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
13 years agoosdmap: initialize cluster_snapshot_epoch
Sage Weil [Thu, 8 Dec 2011 16:41:54 +0000 (08:41 -0800)]
osdmap: initialize cluster_snapshot_epoch

Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
13 years agocrush: set max_devices=0 for map with empty buckets
Sage Weil [Thu, 8 Dec 2011 16:41:39 +0000 (08:41 -0800)]
crush: set max_devices=0 for map with empty buckets

Signed-off-by: Sage Weil <sage.weil@dreamhost.com>
13 years agocrush: fix stepping on unallocated memory
Sage Weil [Thu, 8 Dec 2011 16:01:12 +0000 (08:01 -0800)]
crush: fix stepping on unallocated memory

If size is 0 we can't write here.

Reported-by: pankaj singh <psingh.ait@gmail.com>
Signed-off-by: Sage Weil <sage.weil@dreamhost.com>