- Dec 12, 2008
-
-
Yury Umanets authored
r=johann,shadow - handles -EINTR while waiting for lock + race with bl_ast; - adds sanityN.sh test_34 for testing/reproducing this situation.
-
Elena Gryaznova authored
i=Vladimir.Saveliev test_55 typo fix
-
tianzy authored
add more test info to sanity-quota.sh b=17925 i=johann
-
Terry Rutledge authored
-
- Dec 11, 2008
-
-
Andrew Perepechko authored
b=17938 i=Johann Lombardi i=Vitaly Fertman fdatasync crash test
-
Andrew Perepechko authored
b=17938 i=Vitaly Fertman i=Johann Lombardi fdatasync crash test
-
Johann Lombardi authored
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 Lombardi authored
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 Lombardi authored
b=12596 i=grev i=rread fix file path in compare_stripe_info.
-
Johann Lombardi authored
b=12596 i=grev i=rread fix file path in compare_stripe_info
-
tianzy authored
add ll_recover_lost_found_objs to lustre manual b=16911
-
tianzy authored
add ll_recover_lost_found_objs to lustre manual b=16911
-
Alexey Lyashkov authored
if client will evicted at this time - we start invalidate thread without reference to import and import can be freed at same time. Branch HEAD b=17802 i=nathan i=johann
-
Alexey Lyashkov authored
if client will evicted at this time - we start invalidate thread without reference to import and import can be freed at same time. Branch b1_6 b=17802 i=nathan i=johann
-
huanghua authored
b=17288 att=20766 i=adilger i=rahul.deshmukh i=nathan.rutman ignore direct "trusted.{lma, link}" setting on client and server.
-
- Dec 10, 2008
-
-
girish authored
b=17465 i=adilger i=johann
-
girish authored
b=17052 i=nathan.rutman i=eric.mei
-
girish authored
b=17052 i=nathan.rutman i=eric.mei
-
Elena Gryaznova authored
i=H.Huang run lsof fix
-
Elena Gryaznova authored
i=H.Huang run lsof fix
-
kalpak authored
b=17855 i=huanghua, pravin Revert patch as this breaks CMD. CMD uses child operations
-
kalpak authored
b=16855 i=nathan Fix small bug in search_fsname()
-
- Dec 09, 2008
-
-
Alexander Zarochentsev authored
b=17461 i=r.read i=tappro avoid nesting of obd_uncommitted_replies and rs_lock locks in ptlrpc_handle_rs.
-
Hongchao Zhang authored
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
-
Hongchao Zhang authored
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
-
Alexey Lyashkov authored
Branch HEAD b=14951 i=umka
-
Alexey Lyashkov authored
Branch HEAD b=16492 i=umka
-
Alexey Lyashkov authored
Branch b1_6 b=17918 i=green
-
Alexey Lyashkov authored
Branch HEAD b=17918 i=green
-
Eric Mei authored
add debugging level for user space lgss_keyring.
-
Terry Rutledge authored
-
Andrew Perepechko authored
i=Alexander Zarochentsev i=Johann Lombardi add a trailing zero to avoid oops from proc entry creation routines
-
Andrew Perepechko authored
i=Alexander Zarochentsev i=Johann Lombardi add a trailing zero to avoid oops from proc entry creation routines
-
- Dec 08, 2008
-
-
Nathan Rutman authored
CMD seems to add an unnamed link
-
Eric Mei authored
fix calculation of sleeping time. b=13659 r=nathan
-
Eric Mei authored
fix calculation of sleeping time. b=13659 r=nathan
-
Johann Lombardi authored
b=17379 Print value of symlink depth before checking it.
-
Johann Lombardi authored
b=17379 Print value of symlink depth before checking it.
-
- Dec 07, 2008
-
-
kalpak authored
b=17855 i=h.huang i=tappro mdo_{create,unlink} should use parent object's operations
-
- Dec 06, 2008
-
-
Yong Fan authored
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.
-