[LU-833] Readonly patch way too verbose

• Status: Resolved
• Project: Lustre
• Affects Version/s: Lustre 2.1.0
• Type: Bug
• Priority: Blocker
• Reporter: Johann Lombardi (Inactive)
• Assignee: nasf (Inactive)
• Resolution: Fixed
• Severity: 3

Description

When running replay-single.sh/replay-ost-single.sh, the console logs is flooded with messages like this:

Release to readonly device dm-0 (0xfd00000): [inode 18] [block 39994] [count 5] [is_meta 1]
Lustre: Failing over lustre-OST0000
Write to readonly device dm-0 (0xfd00000) bi_flags: f000000000000001, bi_vcnt: 1, bi_idx: 0, bi->size: 4096, bi_cnt: 2, bi_private: ffff8800750ddf00

Although i understand the need of printing debug information to the console, it makes it really difficult to find useful information in the middle of all this noise.
e.g. for just replay-ost-single/test 8:
$ grep "Write to readonly" replay-ost-single.test_8.console.client-25vm8.log | wc -l
538

This additional debug information was added in bugzilla 16680:

commit 2d3aaef4122c11dcb6d892da89522ffa37036136
Author: Fan Yong <[e-mail]>
Date:   Thu Feb 25 00:32:12 2010 -0800

    b=16680 diagnostic patch to verify lustre read-only device mechanism
    
    Diagnostic patch to check whether lustre read-only device mechanism works well or not.
    
    i=alex.zhuravlev
    i=rahul.deshmukh

If we are now confident that the read-only patch works well now, i would like to disable those debug information. If not, it probably deserves more digging.

Comments

[Johann Lombardi (Inactive)]

Fanyong, any thoughts?

[Peter Jones]

FanYong

Could you please have a look at this one?

Thanks

Peter

[nasf (Inactive)]

I agree with Johann to remove the diagnostic patch which was introduced by bug 16680. I have checked the latest master code, part of the patch has been removed already. I will clean the left parts.

[nasf (Inactive)]

The patch for removing such noisy diagnostic information:

http://review.whamcloud.com/#change,2068

[nasf (Inactive)]

The patch has been landed to lustre-2.2