[LU-3129] Failure on test suite sanity test_17m: e2fsck should not report error upon short/long symlink MDT: rc=4 Created: 08/Apr/13  Updated: 22/Apr/13  Resolved: 22/Apr/13

Status: Resolved
Project: Lustre
Component/s: None
Affects Version/s: Lustre 2.4.0
Fix Version/s: Lustre 2.4.0

Type: Bug Priority: Blocker
Reporter: Maloo Assignee: Di Wang
Resolution: Fixed Votes: 0
Labels: MB, dne

Severity: 3
Rank (Obsolete): 7601

 Description   

This issue was created by maloo for sarah <sarah@whamcloud.com>

This issue relates to the following test suite run: https://maloo.whamcloud.com/test_sets/a02cc9b2-9ec5-11e2-975f-52540035b04c.

The sub-test test_17m failed with the following error:

e2fsck should not report error upon short/long symlink MDT: rc=4

Aftter upgrade from 1.8.9 to 2.4, adding a new MDT, hit following error. I found LU-2097 has the similar error while it has been landed:

== sanity test 17m: run e2fsck against MDT which contains short/long symlink == 18:20:11 (1365211211)
create 512 short and long symlink files under /mnt/lustre/d0.sanity/d17m
erase them
Waiting for local destroys to complete
recreate the 512 symlink files with a shorter string
stop and checking mds1: e2fsck -fnvd /dev/sdb1
Stopping /mnt/mds1 (opts:-f) on fat-amd-1
fat-amd-1: e2fsck 1.42.6.wc2 (10-Dec-2012)
fat-amd-1: [QUOTA WARNING] Usage inconsistent for ID 0:actual (5300224, 1209) != expected (5107712, 1086)
fat-amd-1: [QUOTA WARNING] Usage inconsistent for ID 0:actual (5296128, 1208) != expected (5103616, 1085)
Pass 1: Checking inodes, blocks, and sizes
Pass 2: Checking directory structure
Pass 3: Checking directory connectivity
Unconnected directory inode 31 (/ROOT/d0.sanity/???)
Connect to /lost+found? no

'..' in ... (31) is /ROOT/d0.sanity (30), should be <The NULL inode> (0).
Fix? no

Unconnected directory inode 32 (/ROOT/d0.sanity/???)
Connect to /lost+found? no

'..' in ... (32) is /ROOT/d0.sanity (30), should be <The NULL inode> (0).
Fix? no

Pass 4: Checking reference counts
Inode 30 ref count is 18, should be 20.  Fix? no

Inode 31 ref count is 2, should be 1.  Fix? no

Inode 32 ref count is 2, should be 1.  Fix? no

Pass 5: Checking group summary information
Update quota info for quota type 0? no

Update quota info for quota type 1? no


lustre-MDT0000: ********** WARNING: Filesystem still has errors **********


        1222 inodes used (0.12%, out of 1000184)
          73 non-contiguous files (6.0%)
           1 non-contiguous directory (0.1%)
             # of inodes with ind/dind/tind blocks: 2/0/0
      147815 blocks used (29.56%, out of 500000)
           0 bad blocks
           1 large file

         117 regular files
          72 directories
           0 character device files
           0 block device files
           0 fifos
           5 links
        1024 symbolic links (512 fast symbolic links)
           0 sockets
------------
        1218 files
Starting mds1: -o user_xattr,acl  /dev/sdb1 /mnt/mds1
Started lustre-MDT0000
 sanity test_17m: @@@@@@ FAIL: e2fsck should not report error upon  short/long symlink MDT: rc=4 
  Trace dump:


 Comments   
Comment by Jodi Levi (Inactive) [ 09/Apr/13 ]

Di,
Is this a DNE only issue?

Comment by Andreas Dilger [ 09/Apr/13 ]

Di, does this problem also exist for regular DNE filesystems (i.e. not upgraded ones)? I suspect there is something wrong with the agent inode on disk.

Comment by Di Wang [ 09/Apr/13 ]

Andreas, it only exists for upgraded ones, but it is DNE issue. here is the patch http://review.whamcloud.com/#change,5997

Comment by Jodi Levi (Inactive) [ 17/Apr/13 ]

Landed in 2.4

Comment by Mikhail Pershin [ 22/Apr/13 ]

Please check test_32b failures

Comment by Mikhail Pershin [ 22/Apr/13 ]

Several recent conf-sanity tests failed with 'test_32b failed with 1' and reason is exceeded disk quota. This ticket is the only one changing test_32 recently so I suspect it is responsible. The patch description says:

Fix DNE upgrade tests to use a temporary loop device for adding
new MDT, instead of MDTDEV2, which might affect other tests.

and I assume that temporary loop device might be small in size. Please take a look, these failures are 25% of recent runs.

Comment by Di Wang [ 22/Apr/13 ]

Hmm, that change is only about DNE upgrade, which is 32c, nothing related with 32b.

Comment by Di Wang [ 22/Apr/13 ]

ok, I will close this one, please reopen if you find sth still wrong. Thanks

Generated at Sat Feb 10 01:31:13 UTC 2024 using Jira 9.4.14#940014-sha1:734e6822bbf0d45eff9af51f82432957f73aa32c.