- Oct 11, 2007
-
-
Elena Gryaznova authored
i=Nathan i=Scjody use multiop instead of exec
-
Elena Gryaznova authored
i=Nathan i=Adilger make CLEANUP and SETUP usage more consistent
-
- Oct 10, 2007
-
-
Jinshan Xiong authored
Add a new module llite_lloop in rpm package.
-
- Oct 09, 2007
-
-
Andreas Dilger authored
Add missing connect flag in obd_connect_names for flags only in HEAD.
-
Johann Lombardi authored
b=13659 i=nathan enable D_OTHER (= D_ADAPTTO) for AT test.
-
Johann Lombardi authored
b=11686 i=adilger i=green Move changelog entry from lustre/ChangeLog to lnet/ChangeLog.
-
- Oct 08, 2007
-
-
Johann Lombardi authored
b=11686 i=adilger i=green Severity : enhancement Bugzilla : 11686 Description: Console message flood Details : Make cdls ratelimiting more tunable by adding several tunable in procfs /proc/sys/lnet/console_{min,max}_delay_centisecs and /proc/sys/lnet/console_backoff.
-
komaln authored
r=Nathan, Adilger This patch changes the default stripe policy, when changed on root of the filesystem using 'lfs setstripe' command.
-
Yang Sheng authored
b=13751 i=shadow i=johann Update target & which_patch file for RHEL5 2.6.18-8.1.14.el5.
-
Bobi Jam authored
b=12744 i=johann, adilger NR_CPUS/smp_num_cpus -> num_possible_cpus().
-
- Oct 06, 2007
-
-
Walter Poxon authored
i=adilger
-
Walter Poxon authored
i=adilger
-
Walter Poxon authored
i=adilger i=green
-
Andreas Dilger authored
Move block device configuration commands out of "testing (DANGEROUS)" section of the lctl help. b=5498
-
Andreas Dilger authored
Description: enable data checksumming by default Details : enable checksum by default, allow --disable-checksum configure option and "-o nochecksum" mount option. Checksums can also be disabled at runtime via $LPROC/osc/*/checksums. b=13579 i=nathan i=johann
-
Andreas Dilger authored
Description: enable data checksumming by default Details : enable checksum by default, allow --disable-checksum configure option and "-o nochecksum" mount option. Checksums can also be disabled at runtime via $LPROC/osc/*/checksum_pages. b=13579 i=nathan i=johann Minor fixes to get sanity.sh running.
-
Andreas Dilger authored
Use a 64-bit variable for RPC XID in case of async write error. b=2066 i=liuy i=nathan
-
Andreas Dilger authored
Fix up acceptance-small.sh to take multiple OSTs into account when calculating free space. Don't setstripe on the root directory. b=13587 i=scjody i=nathan
-
- Oct 05, 2007
-
-
Elena Gryaznova authored
i=Nathan i=Anserper change relative pathname to absolute
-
- Oct 04, 2007
-
-
deen authored
Original patch by green. b=10919 i=adilger i=tianzy
-
komaln authored
r=Adilger, Shadow Prevent large kmalloc in lprocfs_alloc_stats() for per-client statistics.
-
Alexey Lyashkov authored
b=13723
-
- Oct 03, 2007
-
-
Alexey Lyashkov authored
-
- Oct 02, 2007
-
-
Alexey Lyashkov authored
b=11710 i=green i=johann
-
Walter Poxon authored
i=green i=adilger
-
- Oct 01, 2007
-
-
Walter Poxon authored
i=green i=adilger
-
- Sep 30, 2007
-
-
Yang Sheng authored
b=13452 i=scjody i=mjmac Kernel config & target file for vanilla kernel-2.6.18.8.
-
- Sep 28, 2007
-
-
Jinshan Xiong authored
-
Yang Sheng authored
b=13518 i=shadow i=johann Modify the vm-tunables-rhel4.patch for RHEL4 kernel-2.6.9-55.0.6.
-
Jinshan Xiong authored
-
- Sep 27, 2007
-
-
Nathan Rutman authored
typo
-
Nathan Rutman authored
i=adilger limit connection switch max to obd_timeout, don't ignore the req timeout of the first reconnecting client
-
mjmac authored
* reverting broken commit (b1_6 does not build with this commit)
-
kalpak authored
i=andreas, green If LAST_ID value on disk is smaller than the objects existing in the O/0/d* directories, it indicates disk corruption and causes an LBUG(). If the object is 0-length, then we should use the existing object. This will help to avoid a full fsck in most cases.
-
kalpak authored
r=bzzz, kalpak jbd-stats patch for rhel5 and 2.6.18.8-vanilla
-
Alexey Lyashkov authored
b=13334 i=green i=johann
-
bwzhou authored
b=12207 add parentheses to eliminate compiling warnings in the original patch
-
Andreas Dilger authored
Quiet overly verbose messages when trying to set block device tunables. There are many reasons why these might fail (e.g. LVM, older kernel, etc). b=11230 i=nathan
-
bwzhou authored
b=12207 i=johann i=shadow hide trusted.lov to the users and export it internally as lustre.lov
-
Jinshan Xiong authored
-