Project

General

Profile

Actions
Copy link

Bug #9434

closed

rbd rm hangs

Added by Yi Li over 9 years ago. Updated over 9 years ago.

Status:
Can't reproduce
Priority:
Normal
Assignee:
-
Category:
-
Target version:
-
% Done:

0%

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

Description

I'm using ceph 0.71(maybe it's a little old)
I did some performance measurements on ceph these days, but I met some problems:
? The fio using ioengine=rbd sometimes became blocked. As I aways run hundreds of fio processes for each measurement, there were always some processes blocked so that the whole work couldn't be finished.

read-64k-1: (g=0): rw=read, bs=64K-64K/64K-64K/64K-64K, ioengine=rbd, iodepth=32
fio-2.1.10
Starting 1 process
rbd engine: RBD version: 0.1.8
Jobs: 1 (f=0): [I(1)] [0.0% done] [0KB/0KB/0KB /s] [0/0/0 iops] [eta 03m:20s]

? I tried to use rbd rm to remove the images, but it was blocked too

-sh-4.1$ for i in `sudo rbd ls`; do sudo rbd rm $i; done
Removing image: 100% complete...
Removing image: 100% complete...
Removing image: 6% complete...

But if I used Python API, this wouldn't happen..

Is it a rbd bug? In newer version(i.e. 0.86), is it fixed?
If it is, I will consider to upgrade the cluster.

Thanks!

Actions
Copy link

Also available in: Atom PDF