Whamcloud - gitweb
brian [Thu, 24 Jun 2004 21:56:49 +0000 (21:56 +0000)]
An insanity.sh cfg file for ltest driven testsl (i.e. insanity-multinode.tst).
brian [Thu, 24 Jun 2004 21:55:10 +0000 (21:55 +0000)]
Print df's return status in test_0.
brian [Thu, 24 Jun 2004 21:54:29 +0000 (21:54 +0000)]
Remove an extraneous $LCONF which would execute on the local system.
jacob [Thu, 24 Jun 2004 19:46:57 +0000 (19:46 +0000)]
Correct fix for 2.6 cpu affinity support
adilger [Thu, 24 Jun 2004 18:45:00 +0000 (18:45 +0000)]
Land b1_2_smallfix onto b1_2 (20040624_1212)
- make sure to unset replay flag from failed open requests (b=3440)
- if default stripe count is 0, use OST count for inode size (b=3636)
- update parent mtime/ctime on client for create/unlink (b=2611)
- drop dentry ref in ext3_add_link from open_connect_dentry (b=3266)
adilger [Fri, 18 Jun 2004 18:02:29 +0000 (18:02 +0000)]
Point release for LLNL who is hitting 3430 a lot (didn't make it into 1.2.3).
b=3430
adilger [Fri, 18 Jun 2004 17:55:56 +0000 (17:55 +0000)]
Don't clean up request reference if we had an MDS open error.
b=3430
adilger [Fri, 18 Jun 2004 17:43:09 +0000 (17:43 +0000)]
Update build version, release date.
b=3229
buffalo-pull [Fri, 18 Jun 2004 13:22:51 +0000 (13:22 +0000)]
Before trying to insmod a .o module, try to modprobe one from the current
kernel.
Also, try to insmod a .ko module if a .o module fails, for kernel 2.6.
adilger [Fri, 18 Jun 2004 01:03:17 +0000 (01:03 +0000)]
Remove the ext3-trusted-ea patch from the chaos series. This is already in
the RHEL kernel, and the changes that were lingering from the old trusted EA
patch only served to screw things up.
adilger [Wed, 16 Jun 2004 23:29:32 +0000 (23:29 +0000)]
Patch updates from b1_4 that are not in b1_2.
adilger [Wed, 16 Jun 2004 23:16:09 +0000 (23:16 +0000)]
Add rh-2.4.21-15.EL in which_patch, vanilla-2.4.24
adilger [Wed, 16 Jun 2004 21:56:07 +0000 (21:56 +0000)]
Fix the configurable stack patch to work for both chaos and rhel.
Fix brokenness in the trusted_ea patch, it got the changes twice.
adilger [Wed, 16 Jun 2004 17:38:52 +0000 (17:38 +0000)]
One more RPM build fix.
adilger [Wed, 16 Jun 2004 16:59:22 +0000 (16:59 +0000)]
Land b1_2_smallfix onto b1_2 (20040616_1009)
adilger [Wed, 16 Jun 2004 16:50:40 +0000 (16:50 +0000)]
Land b1_2_smallfix onto b1_2 (20040616_1009)
- allow clients to reconnect during replay (b=1742)
- re-awaken ptlrpcd if new requests arrive during check_set (b=3554)
- fix cond_resched (b=3554)
- only evict unfinished clients after recovery (b=3515)
- allow bulk resend, prevent data loss (b=3570)
- remove extraneous obd dereference causing LASSERT failure (b=3334)
- don't use get_cycles() when creating temp. files on the mds (b=3156)
- hold i_sem when setting i_size in ll_extent_lock() (b=3564)
- fix ksocknal_fmb_callback() error messages (b=2918)
adilger [Thu, 10 Jun 2004 18:54:53 +0000 (18:54 +0000)]
Land b1_2_smallfix onto b1_2 (20040610_1244)
- dynamic ptlrpc request buffer allocation (b=2102)
- don't allow unlinking open directory if it isn't empty (b=2904)
- set MDS/OST threads to umask 0 to not clobber client modes (b=3359)
adilger [Wed, 9 Jun 2004 02:45:35 +0000 (02:45 +0000)]
Land b1_2_smallfix onto b1_2 (20040608_1846)
adilger [Wed, 9 Jun 2004 02:44:52 +0000 (02:44 +0000)]
Land b1_2_smallfix onto b1_2 (20040608_1846)
- clean kiobufs before and after use (b=3485)
- strip trailing '/'s before comparing paths with /proc/mounts (b=3486)
- remove assertions to work around "in-flight rpcs" recovery bug (b=3063)
- change init script to fail more clearly if not run as root (b=1528)
- fix ns_lock/i_sem lock ordering deadlock for kms update (b=3477)
- don't do DNS lookups on NIDs too small for IP addresses (b=3442)
- servers can dump a log evicting a client - lustre.dump_on_timeout=1
buffalo-pull [Thu, 3 Jun 2004 00:30:29 +0000 (00:30 +0000)]
After further discussion with Andreas, this test and error message should be
the correct ones.
buffalo-pull [Wed, 2 Jun 2004 20:35:10 +0000 (20:35 +0000)]
Make test more accurate and update error message again.
buffalo-pull [Wed, 2 Jun 2004 17:41:23 +0000 (17:41 +0000)]
Make the error message a little more clear as to the conditions of the test
being a FAIL.
buffalo-pull [Wed, 2 Jun 2004 15:12:45 +0000 (15:12 +0000)]
Increase TIMEOUT from 10 to 20 as per b3353.
cvs2svn [Wed, 2 Jun 2004 15:06:24 +0000 (15:06 +0000)]
This commit was manufactured by cvs2svn to create branch 'b1_2'.
phil [Wed, 2 Jun 2004 15:06:23 +0000 (15:06 +0000)]
landing b_cmobd_merge on HEAD
b_cmobd_merge should not be used for any further development or testing!
alex [Tue, 1 Jun 2004 15:19:34 +0000 (15:19 +0000)]
- mds->lmv->mdc propagate lower timeout down to import
we need this becase client's timeout should be longer then inter-mds recovery may take
- few ugly hacks that allow client to reconnect with old (before failure took place) UUID
- mds_preprw() and mds_preprw() should use l_dput() instead of f_dput()
- SOCKNAL_IO_TIMEOUT has been set to 20 seconds to make discovery of
stale connection faster
- lconf generates persisten UUIDs for LMV clients
it each new LMV uses fresh UUID, then each recovered MDS looks like
new client (target MDS doesn't recognize it's old because of new UUID).
if target MDS gets restarted, then it'll find more clients in LAST_RCVD
then it actually had
zab [Sat, 29 May 2004 00:17:09 +0000 (00:17 +0000)]
file export-zap-page-range.patch was initially added on branch b_mmap.
zab [Sat, 29 May 2004 00:13:20 +0000 (00:13 +0000)]
file mmap_sanity.c was initially added on branch b_mmap.
adilger [Thu, 27 May 2004 21:39:00 +0000 (21:39 +0000)]
Update ChangeLog for final 1.2.2 release (3409 is back in).
adilger [Thu, 27 May 2004 13:48:30 +0000 (13:48 +0000)]
Add an ldlm_timeout /proc tuneable during debugging of AST timeouts.
If ldlm_bl_to_thread() fails, call ldlm_handle_bl_callback() directly.
b=3267, b=3468
adilger [Thu, 27 May 2004 08:15:51 +0000 (08:15 +0000)]
Fix rename bug for this patch series, whose name sadly doesn't match "*htree*".
b=3417
adilger [Thu, 27 May 2004 06:31:55 +0000 (06:31 +0000)]
Update tags target from b1_4.
adilger [Wed, 26 May 2004 22:49:02 +0000 (22:49 +0000)]
Add rename_many to tests/Makefile.am
b=3417
adilger [Wed, 26 May 2004 19:33:08 +0000 (19:33 +0000)]
Keep b1_2 and b1_4 htree patches consistent.
b=3275
adilger [Wed, 26 May 2004 15:58:30 +0000 (15:58 +0000)]
Add ext3-htree-rename_fix.patch (originally added on b1_2_smallfix, but
"land1.sh" didn't merge it for some reason).
b=3417
adilger [Wed, 26 May 2004 15:58:24 +0000 (15:58 +0000)]
file ext3-htree-rename_fix.patch was initially added on branch b1_2.
adilger [Tue, 25 May 2004 22:59:08 +0000 (22:59 +0000)]
Land b1_2_smallfix onto b1_2 (20040525_1520)
- revert lock cancellation to blocking thread for glimpse ASTs (3409)
- fix rare problem with rename on htree directories (3417)
- 2.6 patch fixes
alex [Tue, 25 May 2004 10:22:10 +0000 (10:22 +0000)]
b=2950
landed on HEAD
alex [Tue, 25 May 2004 09:10:38 +0000 (09:10 +0000)]
b=3063
alex [Mon, 24 May 2004 18:39:04 +0000 (18:39 +0000)]
- lmv_getattr_name() recognizes cross-node name and repeats getattr to proper
MDS node with gotten ino/generation couple
- warning for cross-node rename in lmv_rename(). it isn't implemented yet.
hence, we'd like to know when we hit this path
- mdc_getattr_name() accepts empty name in order to support getattr by fid request
- mds_get_parent_child_locked() forces MDS_INODELOCK_LOOKUP to be used
if found dentry points to another MDS node
- mea_name2idx() and raw_name2idx() checks for last not to be zero and warns if it's
NOTE: this should fix the cause of the test #5 failure
alex [Mon, 24 May 2004 14:42:18 +0000 (14:42 +0000)]
- bugs in fsfilt_ext3_add_dir_entry() and fsfilt_ext3_del_dir_entryI() fixed:
JBD routines must be called under BKL on 2.4
alex [Mon, 24 May 2004 08:57:47 +0000 (08:57 +0000)]
- ll_update_inode() puts number of MDS holding inode to inode->i_dev. this
fix the problem some apps confuses on: they see the same inode number for
different objects. thanks to Andreas for the idea.
alex [Sun, 23 May 2004 16:22:08 +0000 (16:22 +0000)]
- mdt_obj_create() understands a request which is part of cross-node mkdir()
recovery and looks for given ino/generation first. if no ino/generation are
given then it simple recreates new object and returns ino/generation back
to MDS holding a name. leaked inode is subject to rollback mechanism that
is to be implemented yet
- when mds_reint_create() is asked to replay cross-node mkdir() it signals
this to MDS holding dir inode and sends ino/generation couple if possible
NOTE: now cross-node mkdir() survives failure on any MDS node: holding the
name and holding the inode
alex [Fri, 21 May 2004 17:16:44 +0000 (17:16 +0000)]
- extN-wantedi accepts generation as well as ino to create an inode with
given ino/generation. we need this for cross-node mkdir() which creates
inode on remote MDS and adds dir entry then. after that remote MDS may
fail and primary MDS will have to recreate inode with already gotten
ino and generation numbers
- export refcounting bug in lmv fixed
- mdc_obj_create() stores just received ino/generation back to request.
if request replay will happen mdt_obj_create() will use stored numbers
to recreate lost inode
- mds_filter_recovery_request() accepts OST_CREATE request. need to be
discussed though
- mdt_obj_create() rewritten to support recovery phase
- needless mds_lmv_disconnect() in mds_fs_cleanup() removed
- possible bug in retrieve_generation_numbers() fixed: it should understand
cross-node dir entries and put proper mdsnum's
NOTE: please, re-apply extN-wantedi patch
alex [Fri, 21 May 2004 08:43:35 +0000 (08:43 +0000)]
- workaround for name collision in mdt_obj_create() + debug CERROR there
- mds_commitrw() should use l_dput() instead of f_dput()
- removed needless CERROR() from mds_create_local_dentry()
alex [Thu, 20 May 2004 19:48:24 +0000 (19:48 +0000)]
- pdirops patches against vanilla-2.4.20 series
NOTE: vanilla-2.4.20 runs CMD now
alex [Thu, 20 May 2004 13:23:46 +0000 (13:23 +0000)]
- new routine lmv_get_mea_and_update_object() to be called for
MDS's reply with -ERESTART which notifies directory got splitted
- lmv_intent_open() recognizes a dir gets splitted during request,
retrieves mea and repeats the request using proper MDS
- lmv_create() recognizes a dir gets splitted during request,
retrieves mea and repeats the request using proper MDS
- bug fixed in lmv_getattr_name(): it passed wrong namelen to raw_name2idx()
- lmv_obd_create() sets OBD_MD_FLID to flag MDSs to mark created object
unsplittable. we have to differ two possible requests: 1) to create
remote inode for cross-node mkdir(); 2) to create slave objects. last
ones must not be splitted recursively
- mdt_obj_create() has been rewritten to comply just described rules.
also, it takes a lock on newly created inode. this is needed for recovery
- bug fixed in scan_and_distribute(): it tried to open an inode using decimal
number and this caused iopen_lookup() to find alias dentries
- mds_get_lmv_attr() should return right mea size for given conf. in any case
- minor cleanups
adilger [Wed, 19 May 2004 22:42:40 +0000 (22:42 +0000)]
Fixes for mount.lustre.
- refuse to mount the same filesystem twice on same mountpoint (b=3394)
- allow llmount to create routes for mounting behind routers (b=3320)
alex [Wed, 19 May 2004 20:48:31 +0000 (20:48 +0000)]
- lmv_packmd() and lmv_unpackmd() need not be real connected
jacob [Wed, 19 May 2004 19:57:42 +0000 (19:57 +0000)]
file md_path_lookup-2.6-suse.patch was initially added on branch b1_4_smallfix.
adilger [Wed, 19 May 2004 19:48:54 +0000 (19:48 +0000)]
Test to exercise htree rename bug.
b=3417
adilger [Wed, 19 May 2004 19:48:47 +0000 (19:48 +0000)]
file rename_many.c was initially added on branch b1_2.
adilger [Wed, 19 May 2004 18:26:13 +0000 (18:26 +0000)]
Land b1_2_smallfix onto b1_2 (20040519_1118)
- loop device shouldn't call sync_dev() for nul device (3092)
- push lock cancellation to blocking thread for glimpse ASTs (3409)
- don't call osc_set_data_with_check() for TEST_LOCK matches (3159)
alex [Wed, 19 May 2004 10:29:25 +0000 (10:29 +0000)]
- name var was used being uninitialized
alex [Wed, 19 May 2004 09:00:24 +0000 (09:00 +0000)]
- raw_name2idx declaration to avoid warnings
- ldlm_cli_enqueue() takes 2nd number from resource name into account
to recognize that returned lock changed. we need this because MDS
may return LOOKUP lock with the same ino, but another generation.
the case is very simple mkdir a; then touch a/b; remount; lookup a/b;
a and b my live on different MDSs and have the same ino
- lmv_handle_remote_inode() changes intent from IT_LOOKUP to IT_GETATTR:
caller expect to find attributes in reply
- lmv_intent_lookup() chooses right MDS to revalidate the name
- lmv_create() chooses right MDS where to forward request to
- lmv_link() chooses right mds where to forward request to
- lmv_unlink() chooses right mds where to forward request to
- lmv_readpage() removes . and .. from all the pages, but ones that come
from master MDS for given directory
- lmv_obd_create_single() requests creation of single directory on given MDS
- mdt_obj_create() creates a directory for new mkdir() semantic
- each time new llite connects to MDS, the latter try to connect to own LMV
- mds_pack_inode2body() returns nlink=1 for directories
- mds_lmv_postsetup() pass valid cookiesize to lmv and down to the stack
- mds_reint_create() distributes new dir's inodes over the cluster
- bug in mds_create_local_dentry() fixed: rmdir may cause FIDS/ removal
- mds_get_parents_childer_lock() recognizes cross-ref dentries
adilger [Wed, 19 May 2004 07:13:13 +0000 (07:13 +0000)]
Land b1_2_2766 onto b1_2 (20040518_2110)
- clear page cache after eviction (2766)
- resynchronize MDS->OST in background (2824)
Running test 5 in replay-ost-single.sh (still disabled) still fails in UML,
I'm not sure if this is bug 2766-related or something else.
zab [Tue, 18 May 2004 21:02:58 +0000 (21:02 +0000)]
file ext3-extents-asyncdel-2.4.20-rh.patch was initially added on branch b1_4_mballoc.
zab [Tue, 18 May 2004 19:56:16 +0000 (19:56 +0000)]
file ext3-extents-asyncdel-2.4.24.patch was initially added on branch b1_4_mballoc.
phil [Mon, 17 May 2004 11:52:09 +0000 (11:52 +0000)]
smash the HEAD with the contents of b_cmd. HEAD_PRE_CMD_SMASH and
cMD_PRE_CMD_SMASH are the relevant tags.
The pre-smashing contents of HEAD will be committed by yury into b_cmd as a
merge of the two branches. These will be stabilized on b_cmd and eventually
landed onto HEAD.
rread [Fri, 14 May 2004 20:17:45 +0000 (20:17 +0000)]
file loop-sync-2.4.21-suse.patch was initially added on branch b1_2_smallfix.
green [Thu, 13 May 2004 19:36:10 +0000 (19:36 +0000)]
b=2657
Added group lock tests. Those are not built by default.
adilger [Thu, 13 May 2004 16:12:15 +0000 (16:12 +0000)]
Land b_smallfix onto HEAD (20040512_1806)
b=2094, b=3138, b=3335, b=3274, b=3293, b=2862, b=3147, b=2350, b=3313
b=3265, b=3315, b=3301, b=3325, b=3329, b=3125, b=3098
eeb [Thu, 13 May 2004 14:44:36 +0000 (14:44 +0000)]
file .cvsignore was initially added on branch b_openib.
eeb [Thu, 13 May 2004 13:45:32 +0000 (13:45 +0000)]
file openibnal_cb.c was initially added on branch b_openib.
eeb [Thu, 13 May 2004 13:45:31 +0000 (13:45 +0000)]
file openibnal.h was initially added on branch b_openib.
eeb [Thu, 13 May 2004 13:45:30 +0000 (13:45 +0000)]
file openibnal.c was initially added on branch b_openib.
eeb [Thu, 13 May 2004 13:45:29 +0000 (13:45 +0000)]
file autoMakefile.am was initially added on branch b_openib.
eeb [Thu, 13 May 2004 13:45:28 +0000 (13:45 +0000)]
file Makefile.in was initially added on branch b_openib.
adilger [Thu, 13 May 2004 09:09:27 +0000 (09:09 +0000)]
Land b1_2_smallfix onto b1_2 (20040512_1444)
b=2350, b=3313, b=3265, b=3315, b=3301, b=3325, b=3329, b=3125, b=3098
wangchao [Thu, 13 May 2004 08:16:53 +0000 (08:16 +0000)]
a small fix on lmc to update its reference and description
efelix [Thu, 13 May 2004 03:07:37 +0000 (03:07 +0000)]
removing patch for 2.6.5 as i have moved on.
efelix [Wed, 12 May 2004 23:18:35 +0000 (23:18 +0000)]
New files for vanilla 2.6.6 linux kernel. also a new ldiskfs for 2.6.6
yury [Wed, 12 May 2004 10:27:04 +0000 (10:27 +0000)]
Default mount options stuff.
wangchao [Wed, 12 May 2004 06:52:49 +0000 (06:52 +0000)]
b=2693
r=Robert
Update the regex of conf-sanity.sh test 13 to parse mds uuid from xml file properly.
buffalo-pull [Wed, 12 May 2004 01:34:42 +0000 (01:34 +0000)]
- tagging RC_CURRENT
zab [Tue, 11 May 2004 00:24:10 +0000 (00:24 +0000)]
file llite_mmap.c was initially added on branch b_mmap.
zab [Mon, 10 May 2004 22:56:34 +0000 (22:56 +0000)]
- trivial fox for llog compile warning
rread [Mon, 10 May 2004 19:12:23 +0000 (19:12 +0000)]
Give multiop a chance to open the file on fast machines.
wangchao [Sun, 9 May 2004 09:15:28 +0000 (09:15 +0000)]
Small fix on the description of some lctl commands.
tianying [Sun, 9 May 2004 09:12:03 +0000 (09:12 +0000)]
a small fix to the reference beyond the buffer
wangchao [Sun, 9 May 2004 07:02:32 +0000 (07:02 +0000)]
Correct the description and usage of "lctl conn_list".
buffalo-pull [Fri, 7 May 2004 05:46:05 +0000 (05:46 +0000)]
- tagging RC_CURRENT
ericm [Wed, 5 May 2004 17:53:22 +0000 (17:53 +0000)]
minor fix for build on amd64.
efelix [Wed, 5 May 2004 15:37:42 +0000 (15:37 +0000)]
Adding patch set for vanilla 2.6.5, had to port over one of the suse patched from bitkeeper for o_direct so that the suse patches would go in a little cleaner.
yury [Wed, 5 May 2004 11:40:01 +0000 (11:40 +0000)]
landing smfs.
buffalo-pull [Wed, 5 May 2004 01:34:10 +0000 (01:34 +0000)]
- tagging RC_CURRENT
zab [Tue, 4 May 2004 21:16:56 +0000 (21:16 +0000)]
b=3119
Clear page->private while we're adding it to struct page. Make
page->private errors really assertions as they're very bad and
shouldn't happen.
zab [Tue, 4 May 2004 20:55:15 +0000 (20:55 +0000)]
b=3119
Assert when page->private isn't what we expect.
adilger [Tue, 4 May 2004 17:43:02 +0000 (17:43 +0000)]
b=3285
adilger [Tue, 4 May 2004 17:38:59 +0000 (17:38 +0000)]
b=3043
adilger [Tue, 4 May 2004 17:38:27 +0000 (17:38 +0000)]
Return the right error code.
adilger [Tue, 4 May 2004 17:18:38 +0000 (17:18 +0000)]
Zero i_rdev for non-device files
b=3147
adilger [Tue, 4 May 2004 09:51:02 +0000 (09:51 +0000)]
Add newline to quiet warning.
zab [Tue, 4 May 2004 00:29:39 +0000 (00:29 +0000)]
- derive graph colors from rpc names. this gets hard to look at with
many rpcs, but at least it works.
- also raise the graph a bit to account for lots more types of rpcs.
need to get rid of a lot of the whitespace between bars..
buffalo-pull [Sat, 1 May 2004 02:57:26 +0000 (02:57 +0000)]
- tagging RC_CURRENT
nic [Fri, 30 Apr 2004 22:37:40 +0000 (22:37 +0000)]
fixup LASSERTF for liblustre
buffalo-pull [Fri, 30 Apr 2004 03:19:32 +0000 (03:19 +0000)]
- tagging RC_CURRENT
ccooper [Fri, 30 Apr 2004 02:03:23 +0000 (02:03 +0000)]
- updated tinderbox config info for rh series on x86
adilger [Thu, 29 Apr 2004 23:10:43 +0000 (23:10 +0000)]
Make the "/sbin/mount.lustre: No such device" error message less cryptic.
ccooper [Thu, 29 Apr 2004 17:35:50 +0000 (17:35 +0000)]
- use sles8 iopen patch