Whamcloud - gitweb
fs/lustre-release.git
16 years agoBranch b1_6
bobijam [Mon, 10 Mar 2008 02:03:33 +0000 (02:03 +0000)]
Branch b1_6
b=14321
o=Brian Behlendorf(behlendorf1@llnl.gov)
i=bobijam
i=shadow

Description: lustre_mgs: operation 101 on unconnected MGS
Details    : When MGC is disconnected from MGS long enough, MGS will evict the
             MGC, and late on MGC cannot successfully connect to MGS and a lot
             of the error messages complaining that MGS is not connected.

16 years agob=14471
komaln [Sat, 8 Mar 2008 20:12:59 +0000 (20:12 +0000)]
b=14471
r=Adilger
Changed recovery-small.sh to use 'lctl {get,set}_param'

16 years agoBranch b1_6
adilger [Fri, 7 Mar 2008 20:59:49 +0000 (20:59 +0000)]
Branch b1_6
Make description of bug 14775 more clear.

16 years agoBranch b1_6
anserper [Fri, 7 Mar 2008 19:08:07 +0000 (19:08 +0000)]
Branch b1_6
b=15036
i=johann
i=tianzy

update local vars types to handle 64-bit data

16 years agoBranch b1_6
anserper [Fri, 7 Mar 2008 19:05:46 +0000 (19:05 +0000)]
Branch b1_6
b=15036
i=johann
i=tianzy

update local vars types to handle with 64-bit limits

16 years agob=14471
komaln [Fri, 7 Mar 2008 11:05:37 +0000 (11:05 +0000)]
b=14471

A few corrections in earlier patch to change sanity.sh to use 'lctl {get,set}_param'

16 years agoForget to initialize the li_group.
jxiong [Fri, 7 Mar 2008 04:29:14 +0000 (04:29 +0000)]
Forget to initialize the li_group.

16 years agob=13557 (author Adilger)
grev [Thu, 6 Mar 2008 21:38:32 +0000 (21:38 +0000)]
b=13557 (author Adilger)

print the stripecount as a signed value

16 years agob=14957
grev [Thu, 6 Mar 2008 17:49:58 +0000 (17:49 +0000)]
b=14957
i=Adilger

test_23a fix

16 years agoBranch: b1_6
wangdi [Thu, 6 Mar 2008 08:11:55 +0000 (08:11 +0000)]
Branch: b1_6
b=14483
detect stride IO mode in read-ahead.
i=adilger, nikita

16 years agob=14971
komaln [Thu, 6 Mar 2008 05:29:32 +0000 (05:29 +0000)]
b=14971
r=Adilger
to solve RHEL5/ia64 build error related to 'lctl {get,set}_param' patch

16 years agoClient not clear own cache if answer to reconnect is lost.
shadow [Thu, 6 Mar 2008 04:19:28 +0000 (04:19 +0000)]
Client not clear own cache if answer to reconnect is lost.
b=14775
i=green
i=nathan
i=johann

16 years agob=11300
jxiong [Thu, 6 Mar 2008 03:03:53 +0000 (03:03 +0000)]
b=11300
r=oleg,adilger

Fix interval tree issues at the customer's side.

16 years agoBranch b1_6
bobijam [Thu, 6 Mar 2008 02:24:08 +0000 (02:24 +0000)]
Branch b1_6
b=13715
i=johann
i=nathan
i=adilger

protect accessing imp->generation w/ imp_lock.

16 years agoBranch b1_6
bobijam [Thu, 6 Mar 2008 02:09:09 +0000 (02:09 +0000)]
Branch b1_6
b=12584
i=adilger

There is a issue with getfacl which just literally find the symbol link
target with `pwd`+[symbol content], regardless where the link file is
located.

This fix avoids the issue by chdir to the symbol link's dir before getting it's
acl list.

16 years agoWhen the failover node is the primary node, it is possible
deen [Wed, 5 Mar 2008 18:09:51 +0000 (18:09 +0000)]
When the failover node is the primary node, it is possible
to have two identical connections in imp_conn_list. We must
compare not conn's pointers but NIDs, otherwise we can defeat
connection throttling.

b=14774
i=adilger
i=johann

16 years agoBranch b1_6
bobijam [Wed, 5 Mar 2008 07:24:49 +0000 (07:24 +0000)]
Branch b1_6
b=14873
i=johann
i=shadow

protect competitive adding to obd->obd_nid_stats.

16 years agob=13677
green [Wed, 5 Mar 2008 06:22:46 +0000 (06:22 +0000)]
b=13677
r=eeb,adilger

wait for async rpc completion in liblustre

16 years agob=13297 (comment 36)
grev [Tue, 4 Mar 2008 19:25:40 +0000 (19:25 +0000)]
b=13297 (comment 36)

add 27u to ALWAYS_EXCEPT list

16 years agoBranch b1_6
adilger [Mon, 3 Mar 2008 21:16:40 +0000 (21:16 +0000)]
Branch b1_6
Fix incorrect build version.

16 years agoBranch b1_6
adilger [Mon, 3 Mar 2008 17:19:47 +0000 (17:19 +0000)]
Branch b1_6
Update build version for b1_6.

16 years agob=14748
green [Mon, 3 Mar 2008 02:33:47 +0000 (02:33 +0000)]
b=14748
r=adilger,jay

When traversing extent wating list queue for PR lock and meeting same or wider
lock without AST_SENT, we are guaranteed there is no other one like this down
the line, so we can return immediatelly

16 years agob=14471
komaln [Sun, 2 Mar 2008 20:23:26 +0000 (20:23 +0000)]
b=14471
r=Adilger
Changed conf-sanity.sh to use 'lctl {get,set}_param'

16 years agob=14471
komaln [Sun, 2 Mar 2008 19:23:42 +0000 (19:23 +0000)]
b=14471
r=Adilger
Changed sanityN.sh to use 'lctl {get,set}_param'

16 years agob=14471
komaln [Sun, 2 Mar 2008 19:19:15 +0000 (19:19 +0000)]
b=14471
r=Adilger

Changed sanity.sh to use 'lctl {get,set}_param'

16 years agoFix for sanity quota failure. (From Kalpak)
girish [Thu, 28 Feb 2008 10:40:16 +0000 (10:40 +0000)]
Fix for sanity quota failure. (From Kalpak)
acc-sm passed.
b=13487

16 years agoBranch b1_6
tianzy [Thu, 28 Feb 2008 07:07:46 +0000 (07:07 +0000)]
Branch b1_6
problem:
when a quota request is delayed or dropped, the quota slave who has sent it
will wait for it forever.
solution:
When a quota request time out, dqacq_interpret will hanle it in case the
situation above happened.
b=14840
i=johann
i=andrew.perepechko

16 years agoBranch b1_6
yangsheng [Thu, 28 Feb 2008 06:34:21 +0000 (06:34 +0000)]
Branch b1_6
b=13397
i=shadow
i=johann

Add kernel config for vanilla-2.6.22.14.

16 years agoBranch b1_6
tianzy [Thu, 28 Feb 2008 05:33:09 +0000 (05:33 +0000)]
Branch b1_6
After an unnecessary block in osts for waiting in-flight quota request is
deleted, a few places should be fixed in order to work with it.
b=14783
i=johann
i=andrew.perepechko

16 years agob=14978
grev [Wed, 27 Feb 2008 21:42:42 +0000 (21:42 +0000)]
b=14978
i=Adilger

check whether proc files exist; error if they do not exist

16 years agoBranch: b1_6
wangdi [Wed, 27 Feb 2008 20:45:14 +0000 (20:45 +0000)]
Branch: b1_6
b=13464
Set import disconnection before mgc reconnection, and refine the previous
13464 fix.
i=nathan,adiger

16 years agob=14229
grev [Wed, 27 Feb 2008 19:38:14 +0000 (19:38 +0000)]
b=14229
i=Adilger

fix test_59 to fail if write/read failed

16 years agob=14928
grev [Wed, 27 Feb 2008 19:20:11 +0000 (19:20 +0000)]
b=14928
i=Adilger

don't re-write mtab, wait it will be updated

16 years agob=14528
grev [Wed, 27 Feb 2008 19:06:35 +0000 (19:06 +0000)]
b=14528
i=Adilger

add wiretest to lustre rpm

16 years agob=13571
grev [Tue, 26 Feb 2008 19:33:18 +0000 (19:33 +0000)]
b=13571
i=Adilger

fs2' tests cleanup; renumbering:
- dup test_23 -> 23a and 23b
- 33->33a; 33a->33b

16 years agoBranch:b1_6
wangdi [Tue, 26 Feb 2008 01:45:40 +0000 (01:45 +0000)]
Branch:b1_6
Missing clear "clear" entry in the filter setup error handler.
b=14370
i=adilger,ericm

16 years agoBranch b1_6
bwzhou [Mon, 25 Feb 2008 11:36:22 +0000 (11:36 +0000)]
Branch b1_6
b=13285
r=johann, deen

print messages to console about locks.

16 years agoBranch b1_6
yangsheng [Sat, 23 Feb 2008 16:20:27 +0000 (16:20 +0000)]
Branch b1_6
b=14958
i=adilger
i=yangsheng

Change CONFIG_NR_CPUS from 8 to 32 in a several config file.

16 years agoBranch b1_6
anserper [Fri, 22 Feb 2008 23:09:55 +0000 (23:09 +0000)]
Branch b1_6
b=14852
i=shadow
i=adilger

additional id checks

16 years agoAdding total number of sampled request for an MDS node in snmp support.
deen [Fri, 22 Feb 2008 22:03:12 +0000 (22:03 +0000)]
Adding total number of sampled request for an MDS node in snmp support.

b=14729
i=tappro

16 years agoBranch b1_6
anserper [Fri, 22 Feb 2008 21:07:34 +0000 (21:07 +0000)]
Branch b1_6
b=14899
i=adilger

fix missing \n (which caused error messages)

16 years agoBranch b1_6
adilger [Fri, 22 Feb 2008 19:28:17 +0000 (19:28 +0000)]
Branch b1_6
Make the mi_cbdata match the HEAD version of the statahead patch.
i=fanyong
b=11401

16 years agob=14471
komaln [Fri, 22 Feb 2008 10:23:18 +0000 (10:23 +0000)]
b=14471
r=Adilger, Wangdi
to add lctl {get,set}_param functionality.

16 years agoBranch b1_6
adilger [Fri, 22 Feb 2008 00:01:43 +0000 (00:01 +0000)]
Branch b1_6
Fix comment slightly.
Add a bit more information if debugging is enabled.

16 years agoBranch b1_6
adilger [Thu, 21 Feb 2008 23:51:21 +0000 (23:51 +0000)]
Branch b1_6
Trivial whitespace cleanup.

16 years agoBranch b1_6
johann [Thu, 21 Feb 2008 13:28:18 +0000 (13:28 +0000)]
Branch b1_6
b=14844
i=adilger
i=deen

fix compile issue on PPC. add configure check for asm-ppc/segment.h.

16 years ago2.6.18.8 vanilla series should use sd_iostats-2.6-rhel5.patch
girish [Thu, 21 Feb 2008 12:51:56 +0000 (12:51 +0000)]
2.6.18.8 vanilla series should use sd_iostats-2.6-rhel5.patch
i=adilger
i=bzzz
b=14891

16 years agoBranch b1_6
yangsheng [Thu, 21 Feb 2008 08:29:56 +0000 (08:29 +0000)]
Branch b1_6
b=13397
i=adilger
i=shadow

Add server support for vanilla-2.6.22.14 kernel.
--Landed the kernel patches.
--ldiskfs patches already landed.

16 years agoBranch b1_6
huanghua [Thu, 21 Feb 2008 02:02:05 +0000 (02:02 +0000)]
Branch b1_6
b=14149
a=15414
i=yong.fan
i=yury.umanets

define OBD_CONNECT_FID flag.

16 years agob=14944
grev [Wed, 20 Feb 2008 19:01:24 +0000 (19:01 +0000)]
b=14944
i=Adilger

test_38 cleanup

16 years agoBranch b1_6
bobijam [Wed, 20 Feb 2008 02:17:38 +0000 (02:17 +0000)]
Branch b1_6
b=14529
i=adilger
i=panda

Description: MDS or OSS nodes crash due to stack overflow
Details    : Code changes in 1.6.4 increased the stack usage of some functions.
             In some cases, in conjunction with device drivers that use a lot
             of stack the MDS (or possibly OSS) service threads could overflow
             the stack.  One change which was identified to consume additional
             stack has been reworked to avoid the extra stack usage.

16 years agob=14842
grev [Tue, 19 Feb 2008 23:10:50 +0000 (23:10 +0000)]
b=14842
i=Adilger

move IOZONE=no after $NAME.sh included for SLOW=no default

16 years agob=14842
grev [Tue, 19 Feb 2008 16:25:02 +0000 (16:25 +0000)]
b=14842
i=Adilger

more fixes to reduce acc-sm time; SLOW=no is default now.

16 years agoBetter handle error messages in extents code.
deen [Mon, 18 Feb 2008 14:43:36 +0000 (14:43 +0000)]
Better handle error messages in extents code.

b=14864
i=adilger
i=deen

16 years agoBranch b1_6
vitaly [Mon, 18 Feb 2008 12:43:20 +0000 (12:43 +0000)]
Branch b1_6
b=14379
i=green
i=tappro

increase the amount of buffers for ELC RPCs when there is no conflicting
locks but we cancel locks from lru.

16 years agoBranch b1_6
johann [Mon, 18 Feb 2008 11:12:25 +0000 (11:12 +0000)]
Branch b1_6
b=13223
i=alexey.lyashkov
i=yong.fan

check for -ENODEV in ptlrpc_replay_interpret()

16 years agoBranch b1_6
johann [Mon, 18 Feb 2008 09:34:19 +0000 (09:34 +0000)]
Branch b1_6
b=13400

fix indenting and remove trailing whitespaces in the quota code.

16 years agoBranch b1_6
johann [Mon, 18 Feb 2008 08:56:07 +0000 (08:56 +0000)]
Branch b1_6
b=14648
i=adilger
i=bobijam

when available, use the Adler-32 algorithm instead of CRC32 for
page checksums.

16 years agoBranch b1_6
johann [Mon, 18 Feb 2008 08:07:21 +0000 (08:07 +0000)]
Branch b1_6
b=13805
i=adilger
i=bobijam

add support for several checksum algorithms. Currently, CRC32 and
Adler-32 are supported. The checksum type can be changed on the fly
through /proc/fs/lustre/osc/*/checksum_type.

16 years agoBranch b1_6
yangsheng [Mon, 18 Feb 2008 05:57:44 +0000 (05:57 +0000)]
Branch b1_6
b=14876
i=adilger
i=johann

Kernel security update for RHEL5 2.6.18-53.1.13.el5.

16 years agoBranch b1_6
yangsheng [Mon, 18 Feb 2008 05:52:45 +0000 (05:52 +0000)]
Branch b1_6
b=14858
i=adilger
i=johann

Update SLES10 SP1 kernel to 2.6.16.54-0.2.5.

16 years agoBranch b1_6
johann [Sun, 17 Feb 2008 19:11:45 +0000 (19:11 +0000)]
Branch b1_6
b=14845
i=shadow
i=deen

fix duplicated definitions between asm-ppc64/types.h and lustre_types.h on PPC.

16 years agoBranch b1_6
johann [Sun, 17 Feb 2008 09:32:07 +0000 (09:32 +0000)]
Branch b1_6
b=14618

Downgrad the MPT Fusion driver in the RHEL4 kernel from 3.02.99.00 to 3.02.73 until
the problem regression is fixed.

16 years agob=13766
yury [Sat, 16 Feb 2008 16:23:37 +0000 (16:23 +0000)]
b=13766
r=vitaly,shadow
- fixes abd cleanups in lru resize. Main idea is to avoid dropping SLV much in various situations/loads. Completely reworked memory pressure handler. Fixes in SLV calculations, etc. Please see bug 13766 for more details.

16 years agoBranch b1_6
bwzhou [Fri, 15 Feb 2008 09:53:49 +0000 (09:53 +0000)]
Branch b1_6
b=13380
r=adilger, johann

fix a potential deadlock condition in netpoll code of the sles9 kernel

16 years agoBranch b1_6
bwzhou [Fri, 15 Feb 2008 09:51:09 +0000 (09:51 +0000)]
Branch b1_6
b=14763
r=adilger

fix a problem with conf-sanity

16 years agofile netpoll_xmit_lock-2.6-suse.patch was added on branch b1_6 on 2008-02-15 09:53...
bwzhou [Thu, 14 Feb 2008 08:51:46 +0000 (08:51 +0000)]
file netpoll_xmit_lock-2.6-suse.patch was added on branch b1_6 on 2008-02-15 09:53:48 +0000

16 years agob=13726
grev [Thu, 14 Feb 2008 00:54:21 +0000 (00:54 +0000)]
b=13726
i=Adilger

fix locale for tests where exact output is required

16 years agoBranch b1_6
anserper [Wed, 13 Feb 2008 23:50:29 +0000 (23:50 +0000)]
Branch b1_6
b=14591
i=tianzy
i=andrew.perepechko

16 years agotemporarily remove "sticked" file
anserper [Wed, 13 Feb 2008 23:46:24 +0000 (23:46 +0000)]
temporarily remove "sticked" file

16 years agoBranch b1_6
anserper [Wed, 13 Feb 2008 23:34:25 +0000 (23:34 +0000)]
Branch b1_6
b=14591
i=tianzy
i=andrew.perepechko

kernel patch to avoid oops in dquot code (refer to https://bugzilla.lustre.org/attachment.cgi?id=15058&action=edit)

16 years agoBranch b1_6
anserper [Wed, 13 Feb 2008 21:54:36 +0000 (21:54 +0000)]
Branch b1_6
b=13285
i=johann
i=deen

handle the NULL return case from ptlrpc_prep_req()  (refer to https://bugzilla.lustre.org/attachment.cgi?id=13101&action=edit)

16 years agoBranch b1_6
anserper [Wed, 13 Feb 2008 21:05:42 +0000 (21:05 +0000)]
Branch b1_6
b=13285
i=johann
i=deen

fixes some return code issues (refer to https://bugzilla.lustre.org/attachment.cgi?id=15278&action=edit)

16 years agoBranch b1_6
tianzy [Wed, 13 Feb 2008 11:51:34 +0000 (11:51 +0000)]
Branch b1_6
Reset error code if we are retrying to insert the qdata into the hashtable
to avoid later assertion.
b=14785
i=johann
i=andrew.perepechko

16 years agoPorting raid5 improvements to rhel5 kernels.
jxiong [Wed, 13 Feb 2008 11:51:09 +0000 (11:51 +0000)]
Porting raid5 improvements to rhel5 kernels.
b=13648
r=alex,andreas

16 years agoBranch b1_6
tianzy [Wed, 13 Feb 2008 11:38:02 +0000 (11:38 +0000)]
Branch b1_6
In filter_commitrw_write, lqs is freed between lquota_chkquota and
lquota_pending_commit which led to "quota_pending_commit()) there are too many
blocks!"
b=14784
i=Johann
i=andrew.perepechko

16 years agoBranch b1_6
tianzy [Wed, 13 Feb 2008 07:59:51 +0000 (07:59 +0000)]
Branch b1_6
an unnecessary block in osts for waiting in-flight quota request
b=14783
i=Johann
i=andrew.perepechko

16 years agoBranch b1_6
yangsheng [Wed, 13 Feb 2008 07:05:55 +0000 (07:05 +0000)]
Branch b1_6
b=14720
i=scjody
i=johann

Update RHEL5 kernel to 2.6.18-53.1.6.el5.

16 years agoBranch b1_6
bobijam [Wed, 13 Feb 2008 01:55:24 +0000 (01:55 +0000)]
Branch b1_6
b=13173
i=shadow

make "lctl lov_getconfig" handle the case which "default_stripe_count == -1" by
returning -1 instead of 65535.

16 years agob=14620
grev [Tue, 12 Feb 2008 22:08:19 +0000 (22:08 +0000)]
b=14620
re-vert att. 14680

16 years agob=14330
grev [Tue, 12 Feb 2008 10:47:16 +0000 (10:47 +0000)]
b=14330
i=Adilger

RUNAS dbench and iozone

16 years agob=14330 (comment 14)
grev [Fri, 8 Feb 2008 14:54:59 +0000 (14:54 +0000)]
b=14330 (comment 14)
i=Adilger

remove obsolete cfg scripts

16 years agob=14330
grev [Fri, 8 Feb 2008 14:53:21 +0000 (14:53 +0000)]
b=14330
i=Adilger

remove obsolete insanity-ltest.sh

16 years agob=14330
grev [Fri, 8 Feb 2008 11:57:27 +0000 (11:57 +0000)]
b=14330
i=Adilger
move RUNAS to local.sh; run BONNIE as RUNAS_ID user

b=14474
i=Adilger
don't allow sanity to see MOUNT2; remove chown of MOUNT

16 years agoBranch b1_6
anserper [Thu, 7 Feb 2008 15:47:06 +0000 (15:47 +0000)]
Branch b1_6
b=14421

Changelog update

16 years agoBranch b1_6
anserper [Thu, 7 Feb 2008 10:22:56 +0000 (10:22 +0000)]
Branch b1_6
b=14421
i=nikita.danilov
i=alexey.lyashkov

Removes unnecessary PageDirty check (dirty page is perfectly valid here).

16 years agob=12651
grev [Wed, 6 Feb 2008 18:55:03 +0000 (18:55 +0000)]
b=12651
i=Adilger

skip test_78 if we have no enough space; make the test file slightly smaller

16 years agoBranch b1_6
yangsheng [Tue, 5 Feb 2008 12:53:21 +0000 (12:53 +0000)]
Branch b1_6
b=14793
i=adilger
i=johann

Update RHEL4 kernel to 2.6.9-67.0.4.

16 years agoDon't allow skipping OSTs if index has been specified.
shadow [Tue, 5 Feb 2008 11:52:01 +0000 (11:52 +0000)]
Don't allow skipping OSTs if index has been specified.
b=13196
i=adilger
i=green

16 years agoPatch to remove the warning of compiling for interval_tree.c
jxiong [Tue, 5 Feb 2008 08:38:56 +0000 (08:38 +0000)]
Patch to remove the warning of compiling for interval_tree.c
b=11300
r=shadow

16 years agoJBD2_HAS_COMPAT_FEATURE was used instead of JBD2_HAS_INCOMPAT_FEATURE for
girish [Mon, 4 Feb 2008 16:54:30 +0000 (16:54 +0000)]
JBD2_HAS_COMPAT_FEATURE was used instead of JBD2_HAS_INCOMPAT_FEATURE for
checking _INCOMPAT_ASYNC_COMMIT.
Fixed it.
b=10657

16 years agoBranch b1_6
yangsheng [Mon, 4 Feb 2008 09:30:00 +0000 (09:30 +0000)]
Branch b1_6
b=14714

Fixed previous commit error.

16 years agoBranch b1_6
tianzy [Mon, 4 Feb 2008 07:23:16 +0000 (07:23 +0000)]
Branch b1_6
handle lustre.fail_loc issue when run sanity-quota.sh on different nodes
b=14700
i=johann
i=andrew.perepechko

16 years agoBranch b1_6
yangsheng [Mon, 4 Feb 2008 02:44:15 +0000 (02:44 +0000)]
Branch b1_6
b=14714
i=yangsheng
i=scjody

Use d_move_locked & d_rehash_cond instead of __d_move & __d_rehash in kernel patches.

16 years agoCheck in a fix from Johann Lombardi about missing get_bh().
girish [Sat, 2 Feb 2008 11:51:04 +0000 (11:51 +0000)]
Check in a fix from Johann Lombardi about missing get_bh().
And few small changes as per the upstream patch.
b=10657

16 years agoUsing interval tree for scalable handling of many extent locks.
jxiong [Thu, 31 Jan 2008 08:02:20 +0000 (08:02 +0000)]
Using interval tree for scalable handling of many extent locks.
b=11300
r=vitaly,nikita

16 years agob=14184
komaln [Thu, 31 Jan 2008 05:41:48 +0000 (05:41 +0000)]
b=14184
r=Adilger, Wangdi
To detail ldlm_enqueue and mds_reint stats.

16 years agoBranch b1_6
bwzhou [Thu, 31 Jan 2008 05:36:58 +0000 (05:36 +0000)]
Branch b1_6
b=14721
r=johann, scjody

Quiet "excessive revalidate_it loops" message

16 years agoBranch b1_6
bobijam [Thu, 31 Jan 2008 02:24:11 +0000 (02:24 +0000)]
Branch b1_6
b=13999
o=Philippe Bernadat(philippe.bernadat@hp.com)
i=johann
i=shadow

Don't attempt to wake up evictor thread if it hasn't started yet.

16 years agob=12499
grev [Thu, 31 Jan 2008 02:03:25 +0000 (02:03 +0000)]
b=12499
run sysctl properly (found by Huang Hua)