]>
git.apps.os.sepia.ceph.com Git - ceph.git/log
Sage Weil [Mon, 15 Mar 2010 17:58:57 +0000 (10:58 -0700)]
mds: use CDir::item_new for LogSegment::new_dirfrags elist
Greg Farnum [Mon, 15 Mar 2010 17:37:22 +0000 (10:37 -0700)]
auth: update documentation on caps
Greg Farnum [Mon, 15 Mar 2010 17:03:39 +0000 (10:03 -0700)]
mon: return error codes from prepare_new_pool
Greg Farnum [Fri, 12 Mar 2010 23:45:31 +0000 (15:45 -0800)]
mon: add warning/debugging prints to prepare_new_pool
Sage Weil [Sat, 13 Mar 2010 00:48:32 +0000 (16:48 -0800)]
Merge branch 'mds_mem' into unstable
Sage Weil [Sat, 13 Mar 2010 00:08:36 +0000 (16:08 -0800)]
mds: be consistent about list<>::item naming (item_foo)
Sage Weil [Fri, 12 Mar 2010 23:57:04 +0000 (15:57 -0800)]
dlist: fix pop_front/back
Sage Weil [Fri, 12 Mar 2010 23:28:27 +0000 (15:28 -0800)]
xlist: clean up interface
Sage Weil [Fri, 12 Mar 2010 23:26:15 +0000 (15:26 -0800)]
dlist: clean up interface
Sage Weil [Fri, 12 Mar 2010 23:26:02 +0000 (15:26 -0800)]
mds: use elist instead of dlist
Sage Weil [Fri, 12 Mar 2010 18:13:20 +0000 (10:13 -0800)]
elist: simpler embedded list
This is more or less equivalent to the linux kernel list_head:
each embedded item struct has only a next and prev pointer. As
long as the same member item is always used, at a fixed offset
from the containing class, we can go from an item to a contained
class.
The offset can either be passed to the list (head) constructor,
or to the begin(), front(), back() members explicitly.
Iterator has 3 modes.. current (list_for_each), cache_next
(list_for_each_safe), and magic (uses cached next iff current is
empty). Magic will work most of the time... as long as we don't
re-add ourselves to a different list inside the iterator loop.
(Note that if we do, we will iterator up to the other list's
head, not detect it is a head, an get an invalid pointer and
crash.)
elist: add to makefile
elist: require offset for cosntructor
elist: fix pop_front/back
Greg Farnum [Fri, 12 Mar 2010 22:24:17 +0000 (14:24 -0800)]
rgw: remove unused refcount data member from XMLObj. Also clears up ctor
Greg Farnum [Thu, 11 Mar 2010 22:49:27 +0000 (14:49 -0800)]
rgw: Add documentation
Sage Weil [Fri, 12 Mar 2010 17:09:09 +0000 (09:09 -0800)]
Merge branch 'mds_mem' into unstable
Sage Weil [Fri, 12 Mar 2010 16:40:04 +0000 (08:40 -0800)]
cephx: initialize values
Sage Weil [Thu, 11 Mar 2010 17:40:09 +0000 (09:40 -0800)]
synclient: print mount error
Sage Weil [Fri, 12 Mar 2010 05:43:56 +0000 (21:43 -0800)]
xlist: clean up users
Sage Weil [Fri, 12 Mar 2010 05:43:37 +0000 (21:43 -0800)]
mds: use dlist instead of xlist for main cache objects
Sage Weil [Fri, 12 Mar 2010 05:43:23 +0000 (21:43 -0800)]
dlist: simpler version of xlist without _list pointer
Sage Weil [Fri, 12 Mar 2010 05:11:08 +0000 (21:11 -0800)]
mds: reduce MDSCacheObject footprint a bit
Save 8 bytes. Whoop.
Sage Weil [Fri, 12 Mar 2010 05:07:00 +0000 (21:07 -0800)]
mds: replica_nonce can wrap; fix asserts
Sage Weil [Fri, 12 Mar 2010 04:50:29 +0000 (20:50 -0800)]
mds: put rare state in more_bits_t in ScatterLock
Saves 48 bytes per ScatterLock, 144 bytes per CInode.
Sage Weil [Fri, 12 Mar 2010 04:29:21 +0000 (20:29 -0800)]
mds: show xlist overhead
Sage Weil [Thu, 11 Mar 2010 18:52:32 +0000 (10:52 -0800)]
mds: move unstable SimpleLock state into separate struct
Allocated as needed, freed when lock returns to stable state.
Also make remaining int fields a bit smaller.
Sage Weil [Fri, 12 Mar 2010 05:38:04 +0000 (21:38 -0800)]
debian: specify mirror explicitly when creating pbuilder image
Sage Weil [Thu, 11 Mar 2010 17:57:36 +0000 (09:57 -0800)]
mds: simplify LocalLock wrlock methods
Sage Weil [Thu, 11 Mar 2010 17:41:30 +0000 (09:41 -0800)]
mds: add locallock statemachine
This is just for consistency; it's trivial. We lost the get_sm()
pointer test, though.
Sage Weil [Thu, 11 Mar 2010 17:40:09 +0000 (09:40 -0800)]
synclient: print mount error
Sage Weil [Thu, 11 Mar 2010 17:30:10 +0000 (09:30 -0800)]
mds: create abstract LockType that includes numeric type and state machine
This eliminates 1 field in each SimpleLock.
Sage Weil [Thu, 11 Mar 2010 17:05:37 +0000 (09:05 -0800)]
debian: build binary-arch debs only from builddebs.sh
Otherwise we rebuild the source .dsc, which isn't what we want.
Sage Weil [Wed, 10 Mar 2010 05:38:43 +0000 (21:38 -0800)]
debian: clean out unused debhelper rules steps
Sage Weil [Wed, 10 Mar 2010 05:38:30 +0000 (21:38 -0800)]
debian: put debian/ in .diff.gz, not release tarball
Sage Weil [Tue, 9 Mar 2010 23:16:54 +0000 (15:16 -0800)]
msgr: remove dead kill_after code bits
Sage Weil [Tue, 9 Mar 2010 23:08:08 +0000 (15:08 -0800)]
todo multi-mds issue
Greg Farnum [Wed, 10 Mar 2010 20:56:37 +0000 (12:56 -0800)]
mon: users can now change pool owners.
This requires them to have write permission on both the current
and the new auid.
Greg Farnum [Wed, 10 Mar 2010 18:20:44 +0000 (10:20 -0800)]
osd: Check auid permissions in get_pool_cap
Greg Farnum [Wed, 10 Mar 2010 00:46:45 +0000 (16:46 -0800)]
mon: use MonCaps::check_privileges in prepare_new_pool
Greg Farnum [Wed, 10 Mar 2010 00:42:29 +0000 (16:42 -0800)]
Revert "mon: add check_privileges function to OSDMon."
Deprecated by the existence of MonCaps::check_privileges
This reverts commit
07df67f3948e50a24365e0936bd0de3c7dc1a1f2 .
Greg Farnum [Wed, 10 Mar 2010 00:38:33 +0000 (16:38 -0800)]
mon: add check_privileges function to MonCaps
Greg Farnum [Wed, 10 Mar 2010 00:02:31 +0000 (16:02 -0800)]
vstart: use 'allow *' notation instead of 'allow rwx'
Greg Farnum [Wed, 10 Mar 2010 00:01:38 +0000 (16:01 -0800)]
osd: accept 'allow *' as an admin flag that sets allow_all
in the OSDCap.
Greg Farnum [Tue, 9 Mar 2010 23:50:08 +0000 (15:50 -0800)]
osd: OSDCaps parses and stores auid permissions
Greg Farnum [Tue, 9 Mar 2010 22:14:22 +0000 (14:14 -0800)]
mon: AuthMonitor doesn't overwrite allow_all bit
on incoming messages, only on initial connect.
Greg Farnum [Tue, 9 Mar 2010 22:13:38 +0000 (14:13 -0800)]
mon: Add basic operator<< to MonCaps
Greg Farnum [Tue, 9 Mar 2010 22:12:45 +0000 (14:12 -0800)]
mon: Accept 'allow *' as an admin flag that sets allow_all
in MonCaps.
Greg Farnum [Tue, 9 Mar 2010 22:09:36 +0000 (14:09 -0800)]
mon: MonCaps parses and stores auid permissions
Also, rename MonServiceCap -> MonCap
Sage Weil [Tue, 9 Mar 2010 23:02:30 +0000 (15:02 -0800)]
thread: mask all signals on child threads
Mask all signals on any threads we create. Since we don't use
signals for anything, this leaves the signal behavior to the
original parent thread or process linking in librados or
libceph.
Sage Weil [Tue, 9 Mar 2010 22:51:05 +0000 (14:51 -0800)]
timer: kill unused g_timer
Sage Weil [Tue, 9 Mar 2010 22:05:52 +0000 (14:05 -0800)]
mds: create mds.$id debug log symlink
...in addition to the current mds$num
Sage Weil [Tue, 9 Mar 2010 19:00:03 +0000 (11:00 -0800)]
mds: only check for canceled _client_ requests, not slaves
Slave requests don't get added to the session request list.
Sage Weil [Tue, 9 Mar 2010 18:41:42 +0000 (10:41 -0800)]
mds: fix MDentryUnlink
Need to replicate mds dir + stray dir dentry as well as the stray
inode, dir, dentry. This was overlooked when the /.ceph and
per-mds directories were set up.
Yehuda Sadeh [Mon, 8 Mar 2010 23:34:24 +0000 (15:34 -0800)]
messenger: don't use signal SIGUSRx anymore
Was used internally to wake up blocking messenger threads.
Sage Weil [Mon, 8 Mar 2010 21:35:56 +0000 (13:35 -0800)]
qa: make -j on kernel build
Sage Weil [Mon, 8 Mar 2010 23:10:14 +0000 (15:10 -0800)]
Revert "mon: Hack -- set allow_all in caps if client name is admin"
This reverts commit
f07ce468e502b403faaef6cc5e9bd578366c9395 .
Sage Weil [Mon, 8 Mar 2010 22:49:33 +0000 (14:49 -0800)]
mds: note when we last became non-laggy
Sage Weil [Mon, 8 Mar 2010 22:45:33 +0000 (14:45 -0800)]
mkcephfs: cosd mkfs as root, then chown
Sage Weil [Mon, 8 Mar 2010 22:45:23 +0000 (14:45 -0800)]
todo bugs
Sage Weil [Mon, 8 Mar 2010 19:00:50 +0000 (11:00 -0800)]
filestore: clean up btrfs, clone range detection
Use statfs to detect btrfs, not the trans ioctl (which won't work
if you're not root!). Use separate bool for clone range support.
Sage Weil [Sat, 6 Mar 2010 00:22:33 +0000 (16:22 -0800)]
cauthtool: verify set_uid arg
Sage Weil [Fri, 5 Mar 2010 00:46:03 +0000 (16:46 -0800)]
mds: dump misc struct sizes to log
Sage Weil [Fri, 5 Mar 2010 00:45:42 +0000 (16:45 -0800)]
monclient: time out properly from wait_auth_rotating
Sage Weil [Fri, 5 Mar 2010 00:22:56 +0000 (16:22 -0800)]
decay: separate fixed DecayRate from DecayCounter
Avoid eating a double per instance.
Sage Weil [Thu, 4 Mar 2010 23:48:41 +0000 (15:48 -0800)]
mds: drop useless double from DecayCounter
No need to store k AND halflife.
Greg Farnum [Mon, 8 Mar 2010 15:47:47 +0000 (07:47 -0800)]
mon: Hack -- set allow_all in caps if client name is admin
Greg Farnum [Mon, 8 Mar 2010 15:41:35 +0000 (07:41 -0800)]
objecter: librados: add optional auid param to
create_pool functions
Greg Farnum [Mon, 8 Mar 2010 15:40:54 +0000 (07:40 -0800)]
objecter: add auid to PoolOp, fill it in during pool_op_submit
Greg Farnum [Mon, 8 Mar 2010 14:44:51 +0000 (06:44 -0800)]
mon: let prepare_new_pool assign an auid.
Add new version that takes an MPoolOp and checks permissions
first.
Greg Farnum [Mon, 8 Mar 2010 14:43:24 +0000 (06:43 -0800)]
mon: add check_privileges function to OSDMon.
Greg Farnum [Mon, 8 Mar 2010 14:42:00 +0000 (06:42 -0800)]
msg: Add an auid field to MPoolOp; additional constructor to match
Greg Farnum [Sat, 6 Mar 2010 00:47:26 +0000 (16:47 -0800)]
auth: move CEPH_AUTH_UID_DEFAULT into ceph_fs.h
Greg Farnum [Fri, 5 Mar 2010 21:48:04 +0000 (13:48 -0800)]
cauthtool: set auid in data member format instead of as (ignored) text key
Greg Farnum [Fri, 5 Mar 2010 23:35:28 +0000 (15:35 -0800)]
mon: store auid in MonCaps
Greg Farnum [Fri, 5 Mar 2010 20:31:09 +0000 (12:31 -0800)]
osd: store auid in OSDCaps
Greg Farnum [Fri, 5 Mar 2010 01:16:48 +0000 (17:16 -0800)]
mon: fix bad return test!
cephx_build_service_ticket_blob returns a bool, so it's not ever
going to return <0 -- or if it somehow does that'll be true (good, not false!
Greg Farnum [Fri, 5 Mar 2010 20:36:56 +0000 (12:36 -0800)]
auth: add set_uid() to KeyRing, print out auid in print function
Greg Farnum [Thu, 4 Mar 2010 23:52:31 +0000 (15:52 -0800)]
auth: Add an auid parameter to verify_authorizer and handle_request
It defaults to null, in which case it's not assigned to.
auth: add optional auid param to handle_request
Greg Farnum [Fri, 5 Mar 2010 19:13:59 +0000 (11:13 -0800)]
cephx: Piece-wise initialization of structs is BAD!
If you add, say, a field "auid" to the struct and it's being
piece-wise initialized then you need to change every copy to include it.
Or you assume it's just copied wholesale and then spend many hours trying
to find where it isn't. Like in this one.
Greg Farnum [Fri, 5 Mar 2010 20:48:47 +0000 (12:48 -0800)]
auth: rename auth_uid -> auid
Greg Farnum [Thu, 4 Mar 2010 22:35:53 +0000 (14:35 -0800)]
auth: re-add auid to EntityAuth and AuthTicket.
Greg Farnum [Fri, 5 Mar 2010 23:24:42 +0000 (15:24 -0800)]
mon: add auid field to MonCaps. Also make it a struct --
we want those data members accessible!"
Greg Farnum [Fri, 5 Mar 2010 20:37:56 +0000 (12:37 -0800)]
After discussing with Sage, we do want auid to be a part
of AuthTicket, et al. Let's do it that way.
Revert "authtool: give generated key specific uid if one is provided on the cli."
This reverts commit
3a080fa109d1c9106ac6891c97ee5e816b4ea51d .
Revert "cephx: Add auth_uid (if present) to Info for all service types"
This reverts commit
1cb95b419421e99efbebf77950b0e582907060b4 .
Revert "osd: mon: caps store auth_uid as an int for fast checking"
This reverts commit
08e21ca094859143955f1c2acbef0a1451bcab90 .
Sage Weil [Thu, 4 Mar 2010 21:55:48 +0000 (13:55 -0800)]
mds: shut up about request latency
Sage Weil [Thu, 4 Mar 2010 21:34:54 +0000 (13:34 -0800)]
mds: return lease on null dentries (ENOENT lookups)
This provides the client with a lease on a dentry's nonexistence,
allowing it to avoid repetitive ENOENT lookups.
Sage Weil [Thu, 4 Mar 2010 18:30:08 +0000 (10:30 -0800)]
mds: behave when passed NULL *need_issue
Catch two cases: *need_issue = true at top of func
(if (first ...)), and when eval() call resets need_issue again.
Sage Weil [Thu, 4 Mar 2010 15:41:20 +0000 (07:41 -0800)]
crushtool: fail on parse error
Sage Weil [Thu, 4 Mar 2010 00:49:58 +0000 (16:49 -0800)]
mds: accept stray reconnects
Hmm. Ultimately this is the direction we want to go, so that,
for example, a client that is forcefully disconnected can at
least attempt a reconnect (even if all its caps go ESTALE).
There are undoubtably some issues that need to be dealt with to
keep the mds from choking on weird client data, but those things
need to be fixed regardless.
Sage Weil [Wed, 3 Mar 2010 22:14:42 +0000 (14:14 -0800)]
filestore: fix --mkjournal option
Read fsid off disk before creating journal.
Sage Weil [Thu, 4 Mar 2010 17:46:21 +0000 (09:46 -0800)]
filestore: avoid sync_file_range(2) if not present
This disables the flushing 'optimization'.
Sage Weil [Thu, 4 Mar 2010 17:45:56 +0000 (09:45 -0800)]
filejournal: avoid sync_file_range(2) if not present
Sage Weil [Thu, 4 Mar 2010 17:45:44 +0000 (09:45 -0800)]
autoconf: detect sync_file_range(2)
Sage Weil [Thu, 4 Mar 2010 17:29:38 +0000 (09:29 -0800)]
crushtool: allow multiple -v's to increase verbosity
Sage Weil [Thu, 4 Mar 2010 17:29:23 +0000 (09:29 -0800)]
crushtool: fix parsing with new boost library (compress whitespace)
The latest spirit breaks parsing. It seems to choke on
whitespace (e.g., fail to parse if there is any trailing
whitespace). I couldn't find any obvious problems with how
spirit is beign used, so work around it by stripping out any
extra whitespace from the input. Bleh!
Greg Farnum [Thu, 4 Mar 2010 02:40:22 +0000 (18:40 -0800)]
osd: mon: caps store auth_uid as an int for fast checking
Greg Farnum [Thu, 4 Mar 2010 02:40:18 +0000 (18:40 -0800)]
cephx: Add auth_uid (if present) to Info for all service types
Greg Farnum [Wed, 3 Mar 2010 23:57:27 +0000 (15:57 -0800)]
osd: OSDCaps have a better check for ownership
Greg Farnum [Wed, 3 Mar 2010 23:26:11 +0000 (15:26 -0800)]
authtool: give generated key specific uid if one is provided on the cli.
Of course, you can also now just specify it in the caps file.
Greg Farnum [Wed, 3 Mar 2010 21:02:06 +0000 (13:02 -0800)]
Revert a number of auth_uid commits.
Attempting to add an auth_uid to all the relevant structs has gotten too messy;
it *will* get messed up at some point in the future. So we're moving it into
the string map caps, and the machines can store it in their local structs
however they like.
Note to self: In future, do not attempt to extend mirrored structs without
great cause. Or else commit to making them properly-encapsulated objects.
Revert "auth: Add an auth_uid to AuthTicket. Still to do: copy it around"
This reverts commit
62c8ceba870eae66e26b4796e3adc468e0165028 .
Revert "auth: misc printing and initialization fixes"
This reverts commit
ae8ed243dff02a83cc33c2bf4f4e3715715c482b .
Revert "authtool: set generated key to specific uid if one is given"
This reverts commit
2ab6e75c9b0b2b1f0965ad9ce6ea6a701b6e0098 .
Revert "auth: constant for unknown/unsecured user"
This reverts commit
ea01f18305f5cd79fe8ff69fcbe4dc4b5f13c2f1 .
Revert "auth: auth_uid needs to be in AuthCapsInfo as well."
This reverts commit
d499cf5c0524b2266170f9fe1b7b53451b84aa73 .
Revert "auth: Add a uid field to EntityAuth; make it a required feature"
This reverts commit
14052d4b12a7efb178af5147d761a1589c23e413 .
Conflicts:
src/include/ceph_fs.h
Sage Weil [Thu, 4 Mar 2010 00:55:12 +0000 (16:55 -0800)]
todo
Sage Weil [Tue, 2 Mar 2010 23:41:23 +0000 (15:41 -0800)]
init-ceph: return success error code on missing ceph.conf if no -c
This will avoid errors when installing/removing the unconfigured
package.
Sage Weil [Tue, 2 Mar 2010 23:35:49 +0000 (15:35 -0800)]
mds: fix snapid tests in encode_inodestat
Snapid should always be non-zero; assert as much.
This fixes bug where valid was set to false when stating dir
inodes on non-auth mds's, which in turn returned no caps to
clients and, in the case of cfuse, caused a crash.