Project

General

Profile

Actions

Bug #12711

closed

mds get damaged

Added by Zheng Yan over 8 years ago. Updated over 8 years ago.

Status:
Resolved
Priority:
Urgent
Assignee:
-
Category:
-
Target version:
-
% Done:

0%

Source:
other
Tags:
Backport:
Regression:
No
Severity:
3 - minor
Reviewed:
Affected Versions:
ceph-qa-suite:
Component(FS):
Labels (FS):
Pull request ID:
Crash signature (v1):
Crash signature (v2):

Description

http://pulpito.ceph.com/teuthology-2015-08-10_23:08:02-kcephfs-master-testing-basic-multi/1010323/

It's easy to reproduce this bug locally.

2015-08-17 20:59:33.389204 7f05327fc700 10 mds.0.journaler(ro) try_read_entry at 375040156 read 375040156~936 (have 16981327)
2015-08-17 20:59:33.389230 7f05327fc700 -1 mds.0.journaler(ro) try_read_entry: decode error from _is_readable
2015-08-17 20:59:33.389232 7f05327fc700  0 mds.0.log _replay journaler got error -22, aborting
2015-08-17 20:59:33.389233 7f05327fc700 10 mds.0.journaler(ro) reread_head

In my case, it fails at osdc/Journaler.cc:1276

  if (read_buf.length() >= *need) {
    if (format >= JOURNAL_FORMAT_RESILIENT) {
      ::decode(entry_sentinel, p);
      if (entry_sentinel != sentinel) {
        throw buffer::malformed_input("Invalid sentinel");
        ^^^^^^^
      }
    }

    ::decode(entry_size, p);
  } else {
    return false;
  }

Actions #1

Updated by Zheng Yan over 8 years ago

  • Status changed from New to Fix Under Review
Actions #2

Updated by Greg Farnum over 8 years ago

  • Status changed from Fix Under Review to 7

merged in commit:3cfb7e4ccc08a67ceec73ee684049320c75e9bb2

Actions #3

Updated by Zheng Yan over 8 years ago

  • Status changed from 7 to Resolved
Actions

Also available in: Atom PDF