Whamcloud - gitweb
grev [Thu, 12 Feb 2009 18:49:24 +0000 (18:49 +0000)]
b=17839
i=Brian
recovery-mds-scale for SLOW=no
grev [Thu, 12 Feb 2009 16:14:01 +0000 (16:14 +0000)]
revert b=17839 commit
grev [Thu, 12 Feb 2009 15:27:38 +0000 (15:27 +0000)]
b=17839
i=Brian
recovery-mds-scale for SLOW=no
bobijam [Thu, 12 Feb 2009 06:41:10 +0000 (06:41 +0000)]
Branch b18
b=18015
sles9 doesn't define SLAB_DESTROY_BY_RCU.
girish [Wed, 11 Feb 2009 18:07:07 +0000 (18:07 +0000)]
b=18417
o=h.huang
i=girish
Fix typo from last commit which caused test-framework to exit wrongly
shadow [Wed, 11 Feb 2009 08:48:43 +0000 (08:48 +0000)]
update client to support vanila kernels up to 2.6.27.
Branch b1_8
b=14250
i=green
i=deen
girish [Wed, 11 Feb 2009 06:54:52 +0000 (06:54 +0000)]
b=18417
i=kalpak
i=girish
o=adilger
cleanup correctly after sanity 102 tests
jxiong [Tue, 10 Feb 2009 07:11:36 +0000 (07:11 +0000)]
b=15428
r=adilger
Fix a crash issue for raid5 zerocopy
fanyong [Tue, 10 Feb 2009 04:57:51 +0000 (04:57 +0000)]
Branch b1_8
b=18294
i=tianzy
i=panda
Diagnostic patch to catch what's the garbage left in the system when quota error.
tianzy [Tue, 10 Feb 2009 04:28:48 +0000 (04:28 +0000)]
Branch b1_8
add an entry under /proc to support what policy(RR or QOS) will be used.
b=18334
i=nathan
i=adilger
grev [Tue, 10 Feb 2009 00:24:07 +0000 (00:24 +0000)]
b=18219
i=Adilger
test_70b one client fix
grev [Mon, 9 Feb 2009 23:42:19 +0000 (23:42 +0000)]
b=18366
i=Johann
test_5 ENOSPC fix
johann [Mon, 9 Feb 2009 15:19:09 +0000 (15:19 +0000)]
Branch b1_8
b=18374
i=johann (patch from Tappro)
i=shadow
Return only valid cookies, empty slot for failed ones.
This fixes a memory corruption causing random oops.
johann [Mon, 9 Feb 2009 15:17:18 +0000 (15:17 +0000)]
Branch b1_8
b=18374
i=shadow
i=tappro
Wait for mds_fl_synced to be set during recovery before allowing clients
to reconnect. This is done to make sure that llog is properly set up.
johann [Mon, 9 Feb 2009 15:16:00 +0000 (15:16 +0000)]
Branch b1_8
b=18067
i=johann
i=adilger
Change oap magic LASSERT into LASSERTF to aid debugging.
anserper [Sun, 8 Feb 2009 23:44:26 +0000 (23:44 +0000)]
kernel support clarification
green [Sun, 8 Feb 2009 23:14:31 +0000 (23:14 +0000)]
b=16823
r=adilger,shadow
Update offset calculations for 32 bit platforms
pravins [Sun, 8 Feb 2009 17:24:39 +0000 (17:24 +0000)]
b=15134
i=adilger
i=h.huang
interop patch for lu_dirent attribute handling.
anserper [Sun, 8 Feb 2009 11:26:09 +0000 (11:26 +0000)]
warning clarificationlustre-core.m4
shadow [Fri, 6 Feb 2009 12:30:33 +0000 (12:30 +0000)]
resolve lots races and use after free.
Branch b1_8
b=18114
i=jc.lafoucriere
i=panda
bobijam [Fri, 6 Feb 2009 02:20:38 +0000 (02:20 +0000)]
Branch b1_8
b=17011
i=tom.wang
i=joahnn
remove redundant brw_stats on obdfilter.
huanghua [Fri, 6 Feb 2009 02:11:31 +0000 (02:11 +0000)]
Branch b1_8
b=3633
i=robert.read
i=oleg.drokin
i=tappro
clear the "hold for replay" flag for open request once a file is closed
huanghua [Thu, 5 Feb 2009 12:30:29 +0000 (12:30 +0000)]
Branch b1_8
b=11824
i=rahul.deshmukh
i=yong.fan
interop - enable recovery between 1.8 client and 2.0 server.
shadow [Thu, 5 Feb 2009 11:19:01 +0000 (11:19 +0000)]
handle mds reconnect more fastly.
Branch b1_8
b=18304
i=johann
i=green
bobijam [Thu, 5 Feb 2009 02:09:43 +0000 (02:09 +0000)]
Branch b1_8
b=18225
o=Nic Henke(nic@cray.com)
i=johann
i=adilger
use obd_brw_async in echo_client_kbrw.
bobijam [Wed, 4 Feb 2009 01:42:30 +0000 (01:42 +0000)]
Branch b1_8
b=18015
i=joahnn
i=sheng.yang
add SLAB_DESTROY_BY_RCU to ldlm_lock_slab's flags for ldlm_locks are freed
with RCU mechanism.
girish [Tue, 3 Feb 2009 19:24:43 +0000 (19:24 +0000)]
b=18417
i=adilger
i=kalpak
missed hunk from last commit
girish [Tue, 3 Feb 2009 19:09:40 +0000 (19:09 +0000)]
b=18417
i=adilger
i=kalpak
remove star tests from sanity.sh
grev [Tue, 3 Feb 2009 13:34:16 +0000 (13:34 +0000)]
b=17899
i=Johann
test_53* fix: rm dirs used at start
tianzy [Tue, 3 Feb 2009 06:45:03 +0000 (06:45 +0000)]
Branch b1_8
use QUOTA_REQ_RETURNED to replace -EAGAIN when a quota req is finished.
b=18307
i=johann
i=panda
fanyong [Tue, 3 Feb 2009 02:20:02 +0000 (02:20 +0000)]
Branch b1_8
b=18135
i=tianzy
i=andrew.perepechko
Wake up all wait threads on the "lq_waitq" when "lqc_import" is invalid.
shadow [Mon, 2 Feb 2009 17:33:32 +0000 (17:33 +0000)]
don't call obd_disconnect under lov_lock.
Branch b1_8
b=17310
i=johann
i=rread
grev [Mon, 2 Feb 2009 13:50:08 +0000 (13:50 +0000)]
remove debug line
grev [Mon, 2 Feb 2009 12:53:26 +0000 (12:53 +0000)]
b=16818
i=Nathan
get rid of set_obd_timeout(), improve test_26*
kalpak [Mon, 2 Feb 2009 09:05:05 +0000 (09:05 +0000)]
b=17895
i=adilger
i=alex
add files for last commit
kalpak [Mon, 2 Feb 2009 08:59:35 +0000 (08:59 +0000)]
b=17895
i=adilger
i=alex
While using HA for Lustre servers with Linux RAID, it is possible that MMP will not detect multiple mounts. To make this work we need to unplug the device queue in RAID when the MMP block is being written. Also while reading the MMP block, we should read it from disk and not the cached one.
huanghua [Thu, 29 Jan 2009 08:34:57 +0000 (08:34 +0000)]
Branch b1_8
b=18318
i=rahul
i=pravin
reserve larger reply buffer for close and unlink:
2.0 mds wire protocol returns capabilities.
shadow [Wed, 28 Jan 2009 09:32:44 +0000 (09:32 +0000)]
fix typo
shadow [Wed, 28 Jan 2009 09:12:00 +0000 (09:12 +0000)]
don't lose wakeups for imp_recovery_waitq
Branch b1_8
b=18154
i=nathan
i=rread
girish [Wed, 28 Jan 2009 03:52:30 +0000 (03:52 +0000)]
b=16895
i=adilger (o=johann)
i=girish
dont assert on lvfs_check_rdonly()
yury [Tue, 27 Jan 2009 14:25:03 +0000 (14:25 +0000)]
- added comment about LDLM_FL_* flags at Adilger's request;
- ordered LDLM_FL_* flags correctly.
yury [Tue, 27 Jan 2009 13:17:48 +0000 (13:17 +0000)]
b=17282
r=johann,shadow
- simplifies a lot lru resize memory presssure handling. No estimation pass is needed right now as it was too CPU consuming.
johann [Mon, 26 Jan 2009 13:59:40 +0000 (13:59 +0000)]
Branch b1_8
b=18304
i=shadow
i=nathan
MSG_CONNECT_INITIAL is not set on the initial MDS->OST connect.
As a conseqence, the patch from bug 18224 is not operational
and the MDS export cannot be reused on the OSTs until it gets
evicted.
johann [Mon, 26 Jan 2009 13:42:42 +0000 (13:42 +0000)]
Branch b1_8
b=18307
i=adilger
i=shadow
send OST_CONNECT to OST_REQUEST_PORTAL instead of OST_IO_PORTAL (was used
for compatibility with releases < 1.4.6)
yangsheng [Sat, 24 Jan 2009 18:11:17 +0000 (18:11 +0000)]
Branch b1_8
b=15593
i=nikita, shadow
Force to remove the suid in interop environment.
yangsheng [Sat, 24 Jan 2009 17:56:44 +0000 (17:56 +0000)]
Branch b1_8
b=18091
i=adilger, tappro
Clean up IOC_{INIT, PACK, UNPACK} macros.
grev [Sat, 24 Jan 2009 09:18:00 +0000 (09:18 +0000)]
b=18045
i=Adilger
-- get rid of Hendrix rates values
-- create NUM_FILES -le IFree
-- use mdsrate for cleanup instead of rm
huanghua [Sat, 24 Jan 2009 02:07:36 +0000 (02:07 +0000)]
Branch b1_8
b=17777
i=robert.read
as Robert suggests, get version with more precise match.
yangsheng [Fri, 23 Jan 2009 17:38:19 +0000 (17:38 +0000)]
Branch b1_8
b=17464
Fixed previous commit.
huanghua [Fri, 23 Jan 2009 16:45:56 +0000 (16:45 +0000)]
Branch b1_8
b=17777
i=grev
i=adilger
i=robert.read
test scripts improvement to enable interop testing between 1.8 and 2.0
girish [Fri, 23 Jan 2009 16:39:55 +0000 (16:39 +0000)]
b=15524
i=huanghua
i=kalpak
fix test 102f
shadow [Fri, 23 Jan 2009 14:31:58 +0000 (14:31 +0000)]
add workaround for leak space in ESTALE case.
Branch b1_8
b=16417
grev [Fri, 23 Jan 2009 13:27:32 +0000 (13:27 +0000)]
b=18279
i=Nathan
AT tests cleanup: run setup/cleanup only if AT tests run
yangsheng [Fri, 23 Jan 2009 10:24:38 +0000 (10:24 +0000)]
Branch b1_8
b=17464
i=johann, bobijam
Swab the lov_mds_md when ioctl() be called directly in liblustre.
tianzy [Fri, 23 Jan 2009 09:50:54 +0000 (09:50 +0000)]
Branch b1_8
when there is any inactive ost, refuse to do quotacheck
b=18151
i=johann
i=andrew.perepechko
tianzy [Fri, 23 Jan 2009 06:21:47 +0000 (06:21 +0000)]
Branch b1_8
fix the test issue of test_12 in sanity-quota.sh
b=18151
i=johann
i=panda
vitaly [Thu, 22 Jan 2009 20:52:17 +0000 (20:52 +0000)]
Branch b1_8
b=17748
i=grev
i=adilger
sanityN test fix
girish [Wed, 21 Jan 2009 18:38:33 +0000 (18:38 +0000)]
b=16895
i=adilger (o=johann)
i=girish
don't LASSERT on read-only device
huanghua [Wed, 21 Jan 2009 17:32:45 +0000 (17:32 +0000)]
Branch b1_8
b=18267
i=yong.fan
i=rahul.deshmukh
pass default group value, that is 0, to ost:
this is for such interop: 1.8 client + 1.8 mds + 2.0 ost.
yury [Wed, 21 Jan 2009 16:46:20 +0000 (16:46 +0000)]
b=17310
r=johann
- tmp fix for long connect rpc causing this assert in umount time
fanyong [Wed, 21 Jan 2009 08:56:10 +0000 (08:56 +0000)]
Branch b1_8
b=17800
i=adilger
i=huanghua
Add missing ll_rpc_opcode_table entries for "MDS_WRITEPAGE" and "MDS_IS_SUBDIR".
fanyong [Wed, 21 Jan 2009 03:26:52 +0000 (03:26 +0000)]
Branch b1_8
b=15691
i=huanghua
i=yep
Another diagnostic patch to collect more log when error.
tianzy [Wed, 21 Jan 2009 02:56:51 +0000 (02:56 +0000)]
Branch b1_8
take refcount of export when quotacheck is going on
b=18126
i=shadow
i=panda
adilger [Tue, 20 Jan 2009 19:22:32 +0000 (19:22 +0000)]
Branch b1_8
Minor updates to lustre_idl.h to make it largely match the version in HEAD.
A number of new #defines from HEAD are added to avoid potential future
conflicts, along with whitespace cleanups.
b=17800
shadow [Tue, 20 Jan 2009 13:02:28 +0000 (13:02 +0000)]
always handle mds connect as reconnect.
Branch b1_8
b=18224
i=johhan
i=green
i=adilger
i=wangdi
shadow [Tue, 20 Jan 2009 08:57:54 +0000 (08:57 +0000)]
don't confuse mds_finish_transno() with PTR_ERR(-ENOENT)
Branch b1_8
b=18238
i=green
i=umka
adilger [Tue, 20 Jan 2009 05:16:10 +0000 (05:16 +0000)]
Branch b1_8
Minor updates to lustre_idl.h to make it largely match the version in HEAD.
A number of new #defines from HEAD are added to avoid potential future
conflicts, along with whitespace cleanups.
b=17800
kalpak [Tue, 20 Jan 2009 04:43:42 +0000 (04:43 +0000)]
b=17942
i=adilger
i=girish
update fiemap structures and API to match upstream
fanyong [Tue, 20 Jan 2009 01:38:17 +0000 (01:38 +0000)]
Branch b1_8
b=18182
i=tianzy
i=panda
Set stripe count as 1 and stripte index as 0 to make sure the file is created on first ost.
adilger [Tue, 20 Jan 2009 01:00:13 +0000 (01:00 +0000)]
Branch b1_8
Minor updates to lustre_idl.h to make it largely match the version in HEAD.
A number of new #defines from HEAD are added to avoid potential future
conflicts, along with whitespace cleanups.
b=17800
shadow [Mon, 19 Jan 2009 21:19:04 +0000 (21:19 +0000)]
release lock refernce only after add page to pages list.
Branch b1_8
b=17972
i=shadow
i=johann
johann [Mon, 19 Jan 2009 12:28:40 +0000 (12:28 +0000)]
Branch b1_8
b=18044
i=bobijam
i=vitaly
copy DISCARD_DATA flag if cp ast includes bl ast.
yangsheng [Thu, 15 Jan 2009 02:31:35 +0000 (02:31 +0000)]
Branch b1_8
b=18056
i=adilger, tappro
Fix warnings during automake-1.9.
zhanghc [Tue, 13 Jan 2009 15:41:20 +0000 (15:41 +0000)]
b=17887
1, print the actual time to wait for
lock cancellation in "waiting_locks_callback"
2, compare resource in "ost_rw_hpreq_lock_match"
3, replace "LDLM_ERROR" with "LDLM_WARN" if a lock
is completed after waiting more than obd_timeout
i=vitaly.fertman
i=johann
tappro [Tue, 13 Jan 2009 07:06:37 +0000 (07:06 +0000)]
- warning instead of assert when gap is found in objids without recovery. That
may happen due to inode reuse problem
b:18168
i:shadow, adilger
shadow [Sun, 11 Jan 2009 08:26:18 +0000 (08:26 +0000)]
wait until mds<>ost connected.
Branch b1_8
b=18150
i=grev
shadow [Sun, 11 Jan 2009 07:16:06 +0000 (07:16 +0000)]
resolve compile issue after 16919 landing.
b=16919
green [Sat, 10 Jan 2009 06:47:14 +0000 (06:47 +0000)]
b=18016
r=deen,wangdi
page index type in ll_read_ahead_page should be pgoff_t, not int
girish [Fri, 9 Jan 2009 18:06:07 +0000 (18:06 +0000)]
Change sanity.sh to test "tar --xattr"
b=15524
i=adilger
i=kalpak.shah
green [Fri, 9 Jan 2009 04:03:08 +0000 (04:03 +0000)]
b=16919
r=adilger,shadow
Implement replay of bulk write RPCs, allow servers to reply to write RPCs before
all the metadata was finally committed to disk
girish [Thu, 8 Jan 2009 18:09:15 +0000 (18:09 +0000)]
Add configure option to allow repackaging of e2fsprogs as ldiskfsprogs. (by Jim Garlick)
b=17953
i=adilger
i=girish.shilamkar
yangsheng [Thu, 8 Jan 2009 06:26:53 +0000 (06:26 +0000)]
Branch b1_8
b=17968
i=yangsheng, deen
Fixed various compiler warnings on ppc/ppc64 sles9.
Author: Jim Garlick(LLNL)
vs [Wed, 7 Jan 2009 23:37:57 +0000 (23:37 +0000)]
b=17694
i=adilger
run the test in separate directory
adilger [Wed, 7 Jan 2009 20:55:55 +0000 (20:55 +0000)]
Branch b1_8
Use __GFP_HIGHMEM so that the pages can use all of the available memory
on 32-bit machines.
b=17576
i=adilger (alex.zhuravlev original patch)
i=nikita
zhanghc [Wed, 7 Jan 2009 08:04:18 +0000 (08:04 +0000)]
b=17710
uncommit the patch for further investigation
i=Andreas
tappro [Tue, 6 Jan 2009 07:41:21 +0000 (07:41 +0000)]
- fix for 18148
b:18148
i:shadow,bzzz
adilger [Tue, 6 Jan 2009 03:34:26 +0000 (03:34 +0000)]
Branch b1_8
Remove obsolete test.
yury [Mon, 5 Jan 2009 18:09:35 +0000 (18:09 +0000)]
- synchronize test with HEAD version and rename it to test_35 (just like in HEAD);
- OBD_RACE in all execution branches in ldlm_lock_decref_internal() to balance OBD_RACE in bl_ast
and prevent hanging.
yury [Wed, 31 Dec 2008 16:10:56 +0000 (16:10 +0000)]
b=18127
- reformat in test 47 after test 46 which uses 6 OST
grev [Tue, 30 Dec 2008 12:13:55 +0000 (12:13 +0000)]
b=17797
i=H.Huang
test_57b remote mds fix
zhanghc [Tue, 30 Dec 2008 06:23:19 +0000 (06:23 +0000)]
b=17710
add "sys.ldlm_timeout" as an option to the
MKFS.LUSTRE for setting "ldlm_timeout"
i=Johann
i=Nathan.Rutman
shadow [Mon, 29 Dec 2008 09:46:46 +0000 (09:46 +0000)]
send ACTIVATE event only if connect finished and import have state FULL.
Branch b1_8
i=addilger
i=tappro
i=green
tianzy [Mon, 29 Dec 2008 07:19:04 +0000 (07:19 +0000)]
Branch b1_8
fix test problems in test_18 of sanity-quota.sh
b=17985
i=shadow
shadow [Mon, 29 Dec 2008 06:15:08 +0000 (06:15 +0000)]
revert patch from bug 16417
yangsheng [Sun, 28 Dec 2008 16:18:21 +0000 (16:18 +0000)]
Branch b1_8
b=17153
i=adilger, tappro
Remove 2.4 compatibility.
tappro [Thu, 25 Dec 2008 07:32:11 +0000 (07:32 +0000)]
b:18103
i:adilger, bzzz
shadow [Wed, 24 Dec 2008 21:20:45 +0000 (21:20 +0000)]
checking maximal acl size before set.
Branch b1_8
b=17636
i=johann
i=deen
yangsheng [Wed, 24 Dec 2008 17:45:20 +0000 (17:45 +0000)]
Branch b1_8
b=17153
i=adilger, tappro
Remove 2.4 compatibility from build process.
anserper [Wed, 24 Dec 2008 17:30:48 +0000 (17:30 +0000)]
b=18083
i=Yong Fan
i=Andrew Perepechko
Enforce op. quota file format for MDS too
yury [Wed, 24 Dec 2008 13:00:31 +0000 (13:00 +0000)]
r=shadow
- check for OBD_FAIL_OBD_LLOG_SETUP fail_loc regardless ->lop_setup definition. This is required because
after changes in mds_ost llog setup, which is now done on behalf of separate thread, we need to make
llog_setup() fail in context of mds start thread to make the test actualy work. And those llogs which
get initialized in mds start thread do not have ->lop_setup defined.