Project

General

Profile

Bug #25072

ceph-deploy on Ubuntu with mimic adds all new/replaced disks (osd node) in down state

Added by Greg Smith over 5 years ago. Updated over 5 years ago.

Status:
New
Priority:
Normal
Assignee:
-
Target version:
-
% Done:

0%

Source:
Tags:
Backport:
Regression:
No
Severity:
3 - minor
Reviewed:
Affected Versions:
ceph-qa-suite:
ceph-deploy
Crash signature (v1):
Crash signature (v2):

Description

When a disk failed and trying to replace it or when trying to add a new disk in Ubuntu on mimic with ceph-deploy, it's added with a down state and will not rebuild.
Error in the osd log is: failed to load OSD map for epoch 642, got 0 bytes.

ceph-osd.32.log View (810 KB) Greg Smith, 07/24/2018 06:37 AM

History

#1 Updated by Greg Smith over 5 years ago

4> 2018-07-24 09:34:31.648 7f897f94a700 3 osd.32 0 handle_osd_map epochs [3143,3143], i have 0, src has [3143,3831]
-3> 2018-07-24 09:34:31.648 7f897f94a700 -1 osd.32 0 failed to load OSD map for epoch 3142, got 0 bytes
-2> 2018-07-24 09:34:31.648 7f898d744700 5 -
192.168.31.134:6800/3748 >> 192.168.31.21:6800/11325 conn(0x55c7db5f0000 :-1 s=STATE_OPEN_MESSAGE_READ_FOOTER_AND_DISPATCH pgs=225763 cs=1 l=1). rx mgr.5367581 seq 1 0x55c7db02fc20 mgrconfigure(period=5, threshold=5) v2
1> 2018-07-24 09:34:31.648 7f898df45700 5 - 192.168.31.134:6800/3748 >> 192.168.31.21:6789/0 conn(0x55c7db491800 :-1 s=STATE_OPEN_MESSAGE_READ_FOOTER_AND_DISPATCH pgs=111109 cs=1 l=1). rx mon.0 seq 13 0x55c7db3b9e40 osd_map(3144..3183 src has 3143..3831) v4
0> 2018-07-24 09:34:31.648 7f897f94a700 -1 /build/ceph-13.2.0/src/osd/OSD.h: In function 'OSDMapRef OSDService::get_map(epoch_t)' thread 7f897f94a700 time 2018-07-24 09:34:31.654024
/build/ceph-13.2.0/src/osd/OSD.h: 828: FAILED assert(ret)

ceph version 13.2.0 (79a10589f1f80dfe21e8f9794365ed98143071c4) mimic (stable)

Also available in: Atom PDF