Bug #5704: mon/OSDMonitor.cc: 132: FAILED assert(latest_bl.length() != 0) on cuttlefish - Ceph - Ceph

Actions

Copy link

Bug #5704

closed

mon/OSDMonitor.cc: 132: FAILED assert(latest_bl.length() != 0) on cuttlefish

Added by Sage Weil almost 11 years ago. Updated almost 11 years ago.

Status:

Resolved

Priority:

Immediate

Assignee:

Joao Eduardo Luis

Category:

Target version:

% Done:

Source:

Community (user)

Tags:

Backport:

Regression:

Severity:

3 - minor

Reviewed:

Affected Versions:

ceph-qa-suite:

Pull request ID:

Crash signature (v1):

Crash signature (v2):

Description

     0> 2013-07-22 09:40:12.174661 7fe055f57780 -1 mon/OSDMonitor.cc: In function 'virtual void OSDMonitor::update_from_paxos(bool*)' thread 7fe055f57780 time 2013-07-22 09:40:12.173831
mon/OSDMonitor.cc: 132: FAILED assert(latest_bl.length() != 0)

 ceph version 0.61.5 (8ee10dc4bb73bdd918873f29c70eedc3c7ef1979)
 1: /usr/bin/ceph-mon() [0x5073d6]
 2: (PaxosService::refresh(bool*)+0x19b) [0x4edd4b]
 3: (Monitor::refresh_from_paxos(bool*)+0x57) [0x48e5a7]
 4: (Monitor::init_paxos()+0xf5) [0x48e755]
 5: (Monitor::preinit()+0x6ac) [0x4a4e7c]
 6: (main()+0x1c19) [0x483559]
 7: (__libc_start_main()+0xed) [0x7fe0541b276d]
 8: /usr/bin/ceph-mon() [0x485e7d]
 NOTE: a copy of the executable, or `objdump -rdS <executable>` is needed to interpret this.

stefan saw this and worked around it by setting the latest_full floor at first_committed. not sure why that was needed, though.. need a tarball of a broken mon dir to se what is up