Details
-
Bug
-
Resolution: Fixed
-
Critical
-
Lustre 2.9.0, Lustre 2.10.1, Lustre 2.10.4
-
3
-
9223372036854775807
Description
Actually exactly sure of if this is bug, as a doubt filing this bug. Please adviced. Here is use case or more details about it.
Say I have 4 osts on my setup and created pool with 2 osts assign to it
lctl pool_new lustre.pool ; lctl pool_add lustre.pool lustre-OST0000 lustre-OST0001 ; mkdir /mnt/lustre/test_pool lfs setstripe -c -1 -p lustre.pool /mnt/lustre/test_pool
Now I can create file in that by following two ways
case1: lfs setstripe -c 3 -p lustre.pool /mnt/lustre/test_pool/test1 case2: lfs setstripe -c 3 /mnt/lustre/test_pool/test2
Here in case 1 we trim the strip count to 2 as pool has 2 osts
But in case 2 we allow stripe count 3
So question is case2 is expected behaviour or bug (as test_pool is assigned to pool) ?
Following text taken from admin guide.
Several lfs commands can be run with OST pools. Use the lfs setstripe command to associate a directory with an OST pool. This causes all new regular files and directories in the directory to be created in the pool. The lfs command can be used to list pools in a file system and OSTs in a named pool.
Attachments
Issue Links
- duplicates
-
LU-7660 FS default striping settings only honored on MDT 0
- Resolved
- is related to
-
LU-10369 running "lfs setstripe -c3 /mnt/lustre/pooldir/newfile" doesn't inherit pool
- Resolved
-
LU-7813 default pool not inherited when specifying any layout parameter
- Resolved
- is related to
-
LU-10350 ost-pools test 1n fails with 'failed to write to /mnt/lustre/d1n.ost-pools/file: 1'
- Resolved
- mentioned in
-
Page Loading...
-
Page Loading...
-
Page Loading...
-
Page Loading...
-
Page Loading...
-
Page Loading...
-
Page Loading...