Project

General

Profile

Actions

Bug #9533

closed

kcephfs: fail to send requests initiated during mds restart

Added by Sage Weil over 9 years ago. Updated over 9 years ago.

Status:
Duplicate
Priority:
Urgent
Assignee:
-
Category:
fs/ceph
Target version:
-
% Done:

0%

Source:
Development
Tags:
Backport:
Regression:
Severity:
3 - minor
Reviewed:
Affected Versions:
ceph-qa-suite:
Crash signature (v1):
Crash signature (v2):

Description

mds sees


2014-09-18 19:50:26.603346 7fc9a60cd700  1 -- 10.214.134.10:6800/59746 <== client.2722685 10.214.137.23:0/614492051 15043 ==== client_request(client.2722685:399496 getattr Xs #10003502c71 2014-09-18 18:03:27.000000 RETRY=1) v2 ==== 130+0+0 (2208529155 0 0) 0x10f3da00 con 0x5bc7840

then ~400 caps messages and then
2014-09-18 19:50:26.703888 7fc9a60cd700  1 -- 10.214.134.10:6800/59746 <== client.2722685 10.214.137.23:0/614492051 15444 ==== client_request(client.2722685:399676 lookup #1/teuthology-archive 2014-09-18 19:50:26.000000) v2 ==== 148+0+0 (416039997 0 0) 0x3772500 con 0x5bc7840

the intervening requests are all hung on the client
399520  mds0    getattr  #1
399521  mds0    getattr  #1
399522  mds0    getattr  #1
399523  mds0    getattr  #100034fef3e
399524  mds0    getattr  #1
399525  mds0    getattr  #1
399526  mds0    getattr  #1
399527  mds0    getattr  #1
399528  mds0    getattr  #1
399529  mds0    getattr  #1
399530  mds0    getattr  #1
399531  mds0    getattr  #1
399532  mds0    getattr  #100034feefb
...
399667  mds0    getattr  #1
399668  mds0    getattr  #1
399669  mds0    getattr  #1
399670  mds0    getattr  #1
399671  mds0    getattr  #1
399672  mds0    getattr  #1
399673  mds0    getattr  #1
399674  mds0    getattr  #1
399675  mds0    getattr  #1

and dump_ops_in_flight shows nothing.

Actions #1

Updated by Sage Weil over 9 years ago

  • Status changed from New to Duplicate

this was an old bug, patch was missing from running kernel.

ceph: fix kick_requests()
Actions

Also available in: Atom PDF