Whamcloud - gitweb
anserper [Wed, 17 Dec 2008 15:17:08 +0000 (15:17 +0000)]
b=17905
i=Johann Lombardi
i=Alexey Lyahskov
zero dchild for not to oops on dereferncing ERR_PTR
tappro [Wed, 17 Dec 2008 10:15:49 +0000 (10:15 +0000)]
- open reconstruct for cases when request xid is less than last_xid
b:17939
i:bzzz, rread
fanyong [Wed, 17 Dec 2008 02:45:52 +0000 (02:45 +0000)]
Branch b_release_1_8_0
b=15691
i=johann
i=h.huang
Diagnostic patch
grev [Tue, 16 Dec 2008 20:57:26 +0000 (20:57 +0000)]
b=17839
i=Brian
cmd3-11 ported to acc-sm t-f
-- new RECOVERY_MDS_SCALE
-- rundbench modified
-- new t-f functions, run loads scripts
grev [Tue, 16 Dec 2008 20:47:28 +0000 (20:47 +0000)]
b=17661
i=Brian
run mpi tests as MPI_USER
alex [Mon, 15 Dec 2008 19:33:42 +0000 (19:33 +0000)]
b=17576
- temporal workaround: retry in case of allocation failure, use highmem
tianzy [Mon, 15 Dec 2008 14:53:33 +0000 (14:53 +0000)]
Branch b_release_1_8_0
fix a build error caused by the patch of 16542
b=18027
i=tianzy
i=girish.shilamkar
tianzy [Mon, 15 Dec 2008 07:58:14 +0000 (07:58 +0000)]
Branch b_release_1_8_0
meta blocks ignored can lead to occasional -EDQUOT occasionally
b=16542
i=johann
i=alex.zhuravlev
zhanghc [Sun, 14 Dec 2008 15:14:39 +0000 (15:14 +0000)]
b=17495
in osc_precreated, check OSCC_FLAG_NOSPC before
checking last_id/next_id, remove unnecessary
spin_lock on oscc_lock at the same time.
i=Andreas
i=Nathan.Rutman
kalpak [Sat, 13 Dec 2008 13:06:16 +0000 (13:06 +0000)]
b=17898
i=adilger
i=grev
reset IFS to original value in test_130
yury [Sat, 13 Dec 2008 11:51:15 +0000 (11:51 +0000)]
b=17645
r=johann,shadow
- handles -EINTR while waiting for lock + race with bl_ast;
- adds sanityN.sh test_34 for testing/reproducing this situation.
kalpak [Sat, 13 Dec 2008 11:17:48 +0000 (11:17 +0000)]
b=17908
i=adilger
i=girish
stop kmmpd if mount fails
grev [Fri, 12 Dec 2008 18:20:58 +0000 (18:20 +0000)]
b=17987
i=Vladimir.Saveliev
test_55 typo fix
tianzy [Fri, 12 Dec 2008 06:00:40 +0000 (06:00 +0000)]
Branch b_release_1_8_0
add more test info to sanity-quota.sh
b=17925
i=johann
anserper [Thu, 11 Dec 2008 15:11:17 +0000 (15:11 +0000)]
b=17938
i=Vitaly Fertman
i=Johann Lombardi
Initizalize rqset for not to oops in callback
johann [Thu, 11 Dec 2008 14:38:05 +0000 (14:38 +0000)]
Branch b_release_1_8_0
b=16522
i=adilger
i=wangdi
enable OBD_CONNECT_MDT flag when connecting from the MDS so that
the OSTs know that the MDS "UUID" can be reused for the same export
from a different NID, so we do not need to wait for the export to be
evicted.
johann [Thu, 11 Dec 2008 14:16:09 +0000 (14:16 +0000)]
Branch b_release_1_8_0
b=12596
i=grev
i=rread
fix file path in compare_stripe_info.
tianzy [Thu, 11 Dec 2008 08:12:22 +0000 (08:12 +0000)]
Branch b_release_1_8_0
add ll_recover_lost_found_objs to lustre manual
b=16911
shadow [Thu, 11 Dec 2008 05:51:47 +0000 (05:51 +0000)]
client_disconnect_export vs connect request race.
if client will evicted at this time - we start invalidate
thread without reference to import and import can be freed
at same time.
Branch b_release_1_8_0
b=17802
i=nathan
i=johann
girish [Wed, 10 Dec 2008 17:31:54 +0000 (17:31 +0000)]
Fix lost unlinked ev in reply_in_callback (Isaac)
b=17052
i=nathan.rutman
i=eric.mei
tappro [Wed, 10 Dec 2008 15:04:59 +0000 (15:04 +0000)]
- objids gap handling
b:17666
grev [Wed, 10 Dec 2008 09:48:59 +0000 (09:48 +0000)]
b=17827
i=H.Huang
run lsof fix
huanghua [Wed, 10 Dec 2008 08:37:05 +0000 (08:37 +0000)]
Branch b_release_1_8_0
b=17288
att=20710
i=adilger
i=rahul.deshmukh
filter out "trusted.lma" setting on client and server.
anserper [Tue, 9 Dec 2008 16:03:27 +0000 (16:03 +0000)]
b=17917
i=Alexander Zarochentsev
i=Johann Lombardi
fix proc entries list termination to avoid oops
ericm [Mon, 8 Dec 2008 19:22:28 +0000 (19:22 +0000)]
branch: b_release_1_8_0
fix calculation of sleeping time.
b=13659
r=nathan
huanghua [Sat, 6 Dec 2008 06:59:32 +0000 (06:59 +0000)]
Branch b_release_1_8_0
i=tappro
i=yury.umanets
i=adilger
wait a bit before client umount if vfsmnt still has references.
fanyong [Sat, 6 Dec 2008 04:09:05 +0000 (04:09 +0000)]
Branch b_release_1_8_0
b=15691
i=h.huang
i=johann
If there are too many locks on client-side, then some locks taken by
statahead maybe dropped automatically before the real "revalidate"
using them. Under such case the statahead should fail.
cvs2svn [Fri, 5 Dec 2008 18:38:22 +0000 (18:38 +0000)]
This commit was manufactured by cvs2svn to create branch 'b_release_1_8_0'.
stinson1947 [Fri, 5 Dec 2008 18:38:21 +0000 (18:38 +0000)]
Updated for another build. b_release_1_8_0 created for this build.
johann [Fri, 5 Dec 2008 15:13:41 +0000 (15:13 +0000)]
Branch b1_8_gate
b=17490
i=bzzz
i=kalpak
correctly handle device paths using a subdirectory in /dev when
creating the per-device procfs directory under /proc/fs/ldiskfs.
johann [Fri, 5 Dec 2008 15:13:39 +0000 (15:13 +0000)]
Branch b1_8_gate
b=17490
i=bzzz
i=kalpak
correctly handle device paths using a subdirectory in /dev when
creating the per-device procfs directory under /proc/fs/ldiskfs.
johann [Fri, 5 Dec 2008 15:13:38 +0000 (15:13 +0000)]
Branch b1_8_gate
b=17490
i=bzzz
i=kalpak
correctly handle device paths using a subdirectory in /dev when
creating the per-device procfs directory under /proc/fs/ldiskfs.
johann [Fri, 5 Dec 2008 15:13:35 +0000 (15:13 +0000)]
Branch b1_8_gate
b=17490
i=bzzz
i=kalpak
correctly handle device paths using a subdirectory in /dev when
creating the per-device procfs directory under /proc/fs/ldiskfs.
yury [Fri, 5 Dec 2008 11:06:58 +0000 (11:06 +0000)]
b=17758
r=johann,shadow
- Do NOT call server_deregister_mount() here. This leads to
inability cleanup cleanly and free lsi and other stuff when
mgs calls server_put_mount() in error handling case.
grev [Thu, 4 Dec 2008 17:45:26 +0000 (17:45 +0000)]
b=17747
i=Nathan
run_one: print PASS/FAIL depending on new TEST_FAILED var
huanghua [Thu, 4 Dec 2008 00:31:54 +0000 (00:31 +0000)]
Branch b1_8_gate
b=17589
revert the previous patch because that is rejected by alexey.lyashkov.
grev [Wed, 3 Dec 2008 18:20:01 +0000 (18:20 +0000)]
b=17853
i=Adilger
check_config fix for NETTYPE=ptl
vitaly [Wed, 3 Dec 2008 18:11:19 +0000 (18:11 +0000)]
Branch b1_8_gate
b=17644
i=green
i=adilger
1 extra rpc in flight for high priority request
vitaly [Wed, 3 Dec 2008 18:11:18 +0000 (18:11 +0000)]
Branch b1_8_gate
b=17644
i=green
i=adilger
1 extra rpc in flight for high priority request
vitaly [Wed, 3 Dec 2008 18:11:16 +0000 (18:11 +0000)]
Branch b1_8_gate
b=17644
i=green
i=adilger
1 extra rpc in flight for high priority request
vitaly [Wed, 3 Dec 2008 18:11:15 +0000 (18:11 +0000)]
Branch b1_8_gate
b=17644
i=green
i=adilger
1 extra rpc in flight for high priority request
anserper [Wed, 3 Dec 2008 17:29:16 +0000 (17:29 +0000)]
Branch b1_8_gate
b=17770
i=Elena Gryaznova
avoid using quota_usr/quota_2usr groups
anserper [Wed, 3 Dec 2008 17:29:15 +0000 (17:29 +0000)]
Branch b1_8_gate
b=17770
i=Elena Gryaznova
avoid using quota_usr/quota_2usr groups
anserper [Wed, 3 Dec 2008 17:29:14 +0000 (17:29 +0000)]
Branch b1_8_gate
b=17770
i=Elena Gryaznova
avoid using quota_usr/quota_2usr groups
anserper [Wed, 3 Dec 2008 17:29:13 +0000 (17:29 +0000)]
Branch b1_8_gate
b=17770
i=Elena Gryaznova
avoid using quota_usr/quota_2usr groups
anserper [Wed, 3 Dec 2008 17:29:12 +0000 (17:29 +0000)]
Branch b1_8_gate
b=17770
i=Elena Gryaznova
avoid using quota_usr/quota_2usr groups
vitaly [Wed, 3 Dec 2008 16:51:48 +0000 (16:51 +0000)]
Branch b1_8_gate
b=17748
i=grev
i=adilger
sanityN test issue is fixed
anserper [Wed, 3 Dec 2008 16:38:18 +0000 (16:38 +0000)]
Branch b1_8_gate
b=17371
i=Andreas Dilger
move the cleanup/setup test to the end of the series
zhanghc [Wed, 3 Dec 2008 15:43:36 +0000 (15:43 +0000)]
b=16432
fix mgs_setparam, which will return -EINVAL when param
related to llite is set by MDT or OST
i=johann
i=nathan.rutman
grev [Wed, 3 Dec 2008 15:39:50 +0000 (15:39 +0000)]
b=13584
i=Scjody
test_99a fix: use $TMP as working dir
shadow [Wed, 3 Dec 2008 07:39:44 +0000 (07:39 +0000)]
shrink LOV EAs before replying
Branch b1_8_gate
b=16693
i=shadow
i=johann
shadow [Wed, 3 Dec 2008 07:39:42 +0000 (07:39 +0000)]
shrink LOV EAs before replying
Branch b1_8_gate
b=16693
i=shadow
i=johann
shadow [Wed, 3 Dec 2008 04:21:40 +0000 (04:21 +0000)]
don't resend llog cancels,
fix resend requests for ldlm imports.
Branch b1_8_gate
b=17695
i=umka
i=tappro
shadow [Wed, 3 Dec 2008 04:21:38 +0000 (04:21 +0000)]
don't resend llog cancels,
fix resend requests for ldlm imports.
Branch b1_8_gate
b=17695
i=umka
i=tappro
shadow [Wed, 3 Dec 2008 04:21:36 +0000 (04:21 +0000)]
don't resend llog cancels,
fix resend requests for ldlm imports.
Branch b1_8_gate
b=17695
i=umka
i=tappro
shadow [Wed, 3 Dec 2008 04:21:35 +0000 (04:21 +0000)]
don't resend llog cancels,
fix resend requests for ldlm imports.
Branch b1_8_gate
b=17695
i=umka
i=tappro
huanghua [Wed, 3 Dec 2008 02:26:59 +0000 (02:26 +0000)]
Branch b1_8_gate
b=17589
i=yong.fan
i=yury.umanets
wait a bit while umount client if there is still ref count on mnt.
tappro [Tue, 2 Dec 2008 20:10:22 +0000 (20:10 +0000)]
- update server boot epoch if recovery is not needed too
b:17765
i:adilger,shadow
tappro [Tue, 2 Dec 2008 20:10:21 +0000 (20:10 +0000)]
- update server boot epoch if recovery is not needed too
b:17765
i:adilger,shadow
zam [Tue, 2 Dec 2008 19:43:21 +0000 (19:43 +0000)]
Branch b1_8_gate
b=17742
i=h.huang
i=alexey.lyashkov
Add necessary spin-lock protection to dump_exports()
grev [Tue, 2 Dec 2008 14:31:36 +0000 (14:31 +0000)]
b=17267
i=Adilger
fsx fix: limit bound file size by 1/2 RAM
yury [Tue, 2 Dec 2008 12:17:28 +0000 (12:17 +0000)]
b=17813
r=shadow
- take into account the limit on the host because the higher limit the longer it takes to kill some locks.
grev [Mon, 1 Dec 2008 21:19:14 +0000 (21:19 +0000)]
b=16897
i=Adilger
i=Tappro
test_70b fix: use do_nodes instead of loop to run rundbench
grev [Mon, 1 Dec 2008 09:27:46 +0000 (09:27 +0000)]
b=17696
i=Adilger
skip liblustre for different client/mds versions
grev [Mon, 1 Dec 2008 09:27:45 +0000 (09:27 +0000)]
b=17696
i=Adilger
skip liblustre for different client/mds versions
tianzy [Mon, 1 Dec 2008 03:52:18 +0000 (03:52 +0000)]
Branch b1_8_gate
fix an error in the test_18b of sanity-quota.sh
b=17832
i=tianzy
zhanghc [Sun, 30 Nov 2008 16:12:47 +0000 (16:12 +0000)]
branch=b1_6
b=17031
during refreshing locks waiting its I/O to complete,
take current service time into account, not only using
the timeout gotten by ldlm_get_enqueue_timeout
i=Andreas
i=Nathan.Rutman
zhanghc [Sun, 30 Nov 2008 16:12:46 +0000 (16:12 +0000)]
branch=b1_6
b=17031
during refreshing locks waiting its I/O to complete,
take current service time into account, not only using
the timeout gotten by ldlm_get_enqueue_timeout
i=Andreas
i=Nathan.Rutman
zhanghc [Sun, 30 Nov 2008 16:12:44 +0000 (16:12 +0000)]
branch=b1_6
b=17031
during refreshing locks waiting its I/O to complete,
take current service time into account, not only using
the timeout gotten by ldlm_get_enqueue_timeout
i=Andreas
i=Nathan.Rutman
maxim [Sat, 29 Nov 2008 13:38:48 +0000 (13:38 +0000)]
b=17087
i=rread
i=isaac
Landing the patch making ksocklnd more os-neutral (normalizing names of
lock primitives, moving linux-spcific code to socklnd_lib-linux.c)
maxim [Sat, 29 Nov 2008 13:38:46 +0000 (13:38 +0000)]
b=17087
i=rread
i=isaac
Landing the patch making ksocklnd more os-neutral (normalizing names of
lock primitives, moving linux-spcific code to socklnd_lib-linux.c)
maxim [Sat, 29 Nov 2008 13:38:45 +0000 (13:38 +0000)]
b=17087
i=rread
i=isaac
Landing the patch making ksocklnd more os-neutral (normalizing names of
lock primitives, moving linux-spcific code to socklnd_lib-linux.c)
maxim [Sat, 29 Nov 2008 13:38:44 +0000 (13:38 +0000)]
b=17087
i=rread
i=isaac
Landing the patch making ksocklnd more os-neutral (normalizing names of
lock primitives, moving linux-spcific code to socklnd_lib-linux.c)
maxim [Sat, 29 Nov 2008 13:38:43 +0000 (13:38 +0000)]
b=17087
i=rread
i=isaac
Landing the patch making ksocklnd more os-neutral (normalizing names of
lock primitives, moving linux-spcific code to socklnd_lib-linux.c)
maxim [Sat, 29 Nov 2008 13:38:42 +0000 (13:38 +0000)]
b=17087
i=rread
i=isaac
Landing the patch making ksocklnd more os-neutral (normalizing names of
lock primitives, moving linux-spcific code to socklnd_lib-linux.c)
maxim [Sat, 29 Nov 2008 13:38:41 +0000 (13:38 +0000)]
b=17087
i=rread
i=isaac
Landing the patch making ksocklnd more os-neutral (normalizing names of
lock primitives, moving linux-spcific code to socklnd_lib-linux.c)
grev [Fri, 28 Nov 2008 18:40:18 +0000 (18:40 +0000)]
b=17761
i=Adilger
i=Johann
test_6 fix
anserper [Fri, 28 Nov 2008 10:24:25 +0000 (10:24 +0000)]
Branch b1_8_gate
b=17772
i=Johann Lombardi
i=ZhiYong Tian
lov_quota_check shall return an error when a target goes offline
tianzy [Fri, 28 Nov 2008 07:00:59 +0000 (07:00 +0000)]
Branch b1_8_gate
let dqacq_in_flight() hold lock for qunit
b=16890
i=johann
i=panda
aVS: ----------------------------------------------------------------------
brian [Thu, 27 Nov 2008 21:56:57 +0000 (21:56 +0000)]
b=16133
i=cliffw
i=wangyb
lbuild should not ignore the --with-linux option (and instead look for a
kernel-source RPM) specified for patchless builds
johann [Thu, 27 Nov 2008 10:48:44 +0000 (10:48 +0000)]
Branch b1_8_gate
b=12596
i=grev
i=adilger
check striping after setstripe in recovery-small test 18*.
move get_stripe_info() to t-f.
johann [Thu, 27 Nov 2008 10:48:43 +0000 (10:48 +0000)]
Branch b1_8_gate
b=12596
i=grev
i=adilger
check striping after setstripe in recovery-small test 18*.
move get_stripe_info() to t-f.
johann [Thu, 27 Nov 2008 10:48:42 +0000 (10:48 +0000)]
Branch b1_8_gate
b=12596
i=grev
i=adilger
check striping after setstripe in recovery-small test 18*.
move get_stripe_info() to t-f.
adilger [Wed, 26 Nov 2008 23:15:34 +0000 (23:15 +0000)]
Branch b1_8_gate
Fix 80-column line wrapping and complex type casting for V3 EAs.
adilger [Wed, 26 Nov 2008 23:09:21 +0000 (23:09 +0000)]
Branch b1_8_gate
Remove extraneous whitespace.
tappro [Wed, 26 Nov 2008 21:03:29 +0000 (21:03 +0000)]
- fix liblustre build
b:17717
yangsheng [Wed, 26 Nov 2008 15:14:22 +0000 (15:14 +0000)]
Branch b1_8_gate
b=17691
i=adilger, johann
Handle the 1.6 lustre-msg in interop case.
yangsheng [Wed, 26 Nov 2008 15:14:20 +0000 (15:14 +0000)]
Branch b1_8_gate
b=17691
i=adilger, johann
Handle the 1.6 lustre-msg in interop case.
tianzy [Wed, 26 Nov 2008 08:23:33 +0000 (08:23 +0000)]
Branch b1_8_gate
fix the false qunit_put in qctxt_wait_pending_dqacq()
b=17794
i=tianzy
i=panda
tappro [Wed, 26 Nov 2008 08:05:10 +0000 (08:05 +0000)]
- check that fs is version-capable and avoid VBR if not
b:17717
i:adilger,alex
tappro [Wed, 26 Nov 2008 08:05:06 +0000 (08:05 +0000)]
- check that fs is version-capable and avoid VBR if not
b:17717
i:adilger,alex
tappro [Wed, 26 Nov 2008 08:05:02 +0000 (08:05 +0000)]
- check that fs is version-capable and avoid VBR if not
b:17717
i:adilger,alex
jxiong [Wed, 26 Nov 2008 07:11:19 +0000 (07:11 +0000)]
Restore the changes I made yesterday which broke build because I used a stale kernel.
jxiong [Wed, 26 Nov 2008 07:11:18 +0000 (07:11 +0000)]
Restore the changes I made yesterday which broke build because I used a stale kernel.
jxiong [Wed, 26 Nov 2008 07:11:17 +0000 (07:11 +0000)]
Restore the changes I made yesterday which broke build because I used a stale kernel.
adilger [Tue, 25 Nov 2008 23:58:25 +0000 (23:58 +0000)]
Branch b1_8_gate
Ignore generated file.
anserper [Tue, 25 Nov 2008 23:12:01 +0000 (23:12 +0000)]
Branch b1_8_gate
b=17371
i=Elena
fail each time cleanup/setup failed
grev [Tue, 25 Nov 2008 20:10:53 +0000 (20:10 +0000)]
b=17326
i=Brian
load_modules fn modprobe.d fix
grev [Tue, 25 Nov 2008 19:22:14 +0000 (19:22 +0000)]
b=17477
i=Huanh Hua
check_config fn default network type fix
yangsheng [Tue, 25 Nov 2008 12:44:39 +0000 (12:44 +0000)]
Branch b1_8_gate
b=17630
Add a comment to explain the change.