b=17752
i=yury.umanets
i=tianzy

Disable quota support for CMD case temporary.

i=Johann Lombardi
i=ZhiYong Tian

quota_check shall return an error when a target goes offline

i=Nikita
i=umka

	compact FLD feature.

let dqacq_in_flight() hold lock for qunit
b=16890
i=johann
i=panda

i=cliffw
i=wangyb

lbuild should not ignore the --with-linux option (and instead look for a
kernel-source RPM) specified for patchless builds

Remove extra whitespace at end of lines.

b=12596
i=grev
i=adilger

check striping after setstripe in recovery-small test 18*.
move get_stripe_info() to t-f.

b=16164
i=adilger
i=bobijam

enable OBD_CONNECT_{CKSUM,BRW_SIZE} in mds_lov_connect()

b=16860
i=nathan
i=rread

With AT enabled, the recovery window can be excessively long (6000+
seconds). To address this problem, we no longer use
OBD_RECOVERY_FACTOR when extending the recovery window (the connect
timeout no longer depends on the service time, it is set to
INITIAL_CONNECT_TIMEOUT now) and clients report the old service
time via pb_service_time.

b=17818
i=fanyong
i=pravin

b=12749
i=nathan,deen

Third of three patches for the root squash feature

map root UID/GID to UID/GID set via configuration parameter root_squash
with the exception for NID range set via configuration parameter nosquash_nids

b=12749
i=nathan,deen

Second of three patches for the root squash feature

This patch is to allow configuration parameters to contain spaces
It is accopmpished with
1. in mgs_write_log_params (renamed to mgs_write_log_param)
    assume the incoming string is a single parameter (the while loop is removed)
2. in mgs_write_log_target, call mgs_write_log_params for each parameter
a) new funcion class_get_next_param copies the entire next parameter into a temp buffer
   spaces are included if they are inside quotes (single or double)
   the first space outside of quotes is the terminator
   the quotes are stripped in the copy
   the copy is null-terminated
b) repeatedly call class_get_next_param and mgs_write_log_param in a loop from mgs_write_log_target

b=12749
i=maxim,isaac

First of three patches for the root squash feature

This adds three functions to manipulate with lists of NID ranges:
cfs_parse_nidlist - parses nid range list and
                    creates structure of lists of [lo,hi,stride,net]
                    which can be used to check matching a NID to set of NIDs
cfs_free_nidlist - frees structures allocated on nidlist parsing
cfs_match_nid - check whether a NID matches to set of NIDs
cfs_iswhite - auxiliary function to check if a character is a whitespace

These functions have prefixes cfs_ instead of libcfs_ because of bug #17167.

Example of NID range list: 10.2.0.[1-20/2]@tcp [0,2,6]@elan0 [1,2,4-100/4]@elan1
nidlist syntax:
  <nidlist>         :== <nidrange> [ ' ' <nidrange> ]
  <nidrange>        :== <addrrange> '@' <net>
  <addrrange>       :== '*' |
                        <ipaddr_range> |
                        <numaddr_range>
  <ipaddr_range>    :== <numaddr_range>.<numaddr_range>.<numaddr_range>.<numaddr_range>
  <numaddr_range>   :== <number> |
                        <expr_list>
  <expr_list>       :== '[' <range_expr> [ ',' <range_expr>] ']'
  <range_expr>      :== <number> |
                        <number> '-' <number> |
                        <number> '-' <number> '/' <number>
  <net>             :== <netname> | <netname><number>
  <netname>         :== "lo" | "tcp" | "o2ib" | "cib" | "openib" | "iib" |
                        "vib" | "ra" | "elan" | "gm" | "mx" | "ptl"
  <number>          :== <nonnegative decimal> | <hexadecimal>

Wireshark plugins for dissecting LNET/Lustre 1.6 packets.
(contributed by Stephane Thiell)

update b_som from the proper branch

fix the false qunit_put in qctxt_wait_pending_dqacq()
b=17794
i=tianzy
i=panda

i=Huang hua
i=Rahul

This patch contains llo API interface change. Now it allows to create object
in specified path.
Fixes bug in object create in directory other than "/"
Adds llo unregister api.

Restore the changes I made yesterday which broke build because I used a stale kernel.

remove a line of m4_allow_pattern.
b=17781
r=adilger

b=17371
i=Elena
fail each time setup/cleanup does not work

i=Brian
load_modules fn modprobe.d fix

i=Huang Hua
check_config fn default network type fix

i=umka
i=panda
b=17611

don't override lcm->lcm_name

Rationale is the following:

* 30 sec we need because in 30 sec locks on client will get old enough to _start_ kiling them;
* 10 sec we need because client checks SLV and kills locks every 10 sec;
* 10 sec more we need to make sure that we did not race in worse case scenario and client really had some time to kill locks.

i=green

 - few minor issues fixed

fix "should take longer" problem of test_18 of sanity-quota.sh
b=17773
i=johann
i=panda

change target_handle_dqacq_callback() error handling
b=16890
i=johann
i=panda

Fixed the raid5 patches.
- rebuild policy for rhel5 .21 kernel
- soft lockups fixed

b=17084
r=adilger,jay

b=17786

i=adilger, huanghua

Initialize the request.

support gss on mgc-mgs connection.
b=16054
r=umka
r=vitaly

rewrite part of sptlrpc configuration system to be able to interoperate
with 1.8.
b=15363
r=nathan
r=fanyong

i=Tappro
FAIL_ON_ERROR=false fix: force suits exit 1 if some tests failed

Add "abort_recov" mount option to mount.lustre usage message.

Fix 80-column line overflow.

Add error messages in case of failure in test_27p.
Don't have explicit return inside test_75*, this is already skipped due
to test 75 being listed in ALWAYS_EXCEPT.

Add regression tests for -1 stripe width case.
b=12836

Fix filename formatting to include space for sticky bit.

Fix broken whitespace in error message.