Closing as Fix Released; we fixed this with multipath-tools
(0.4.9-3ubuntu11) in vivid.
Nominating a fix for Trusty too which we should SRU.
** Changed in: multipath-tools (Ubuntu)
Status: Confirmed = Fix Released
** Also affects: multipath-tools (Ubuntu Trusty)
Importance: Undecided
Using the lucid kernel, I was able to reproduce, though it took a lot longer
than
original report implied, easily over an hour, probably more than 3 I wasn't
watching
it the entire time.
multipath-tools 0.4.8-14ubuntu4.10.04.2
kernel 2.6.32-41-generic
udevadm monitor simply stops responding.
** Changed in: multipath-tools (Ubuntu)
Status: Invalid = Confirmed
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to multipath-tools in Ubuntu.
https://bugs.launchpad.net/bugs/585027
Title:
Race condition with dmsetup causes
Retested on precise base and downgraded to multipath-tools 0.4.8-14ubuntu4
on generic 3.2.0.27.29. Ran unit test for several hours, no fault found.
Intend to downgrade to lucid kernel and retest, hopefully it pops then, save
me the trouble of installing lucid somewhere.
--
You received this bug
In examining the ioctl interface to device mapper (kernel), I have not found
any architecture that
assures that when a device mapper device has been marked for deletion, that
this information is
relayed back to userspace so it can block on this deletion. An example will
clarify things.
In SD
Should really be won't fix but I apparently can't set it.
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to multipath-tools in Ubuntu.
https://bugs.launchpad.net/bugs/585027
Title:
Race condition with dmsetup causes 'map already
I don't follow this. Your title mentions 'dmsetup' but your description only
talks about multipathd. What is the problem here? One of the multipath
commands in your loop failing to wait until the associated udev action has
completed before exiting? Or some dmsetup command you are running in
On Mon, Jul 09, 2012 at 03:29:11PM -, Peter Petrakis wrote:
In examining the ioctl interface to device mapper (kernel), I have not found
any architecture that
assures that when a device mapper device has been marked for deletion, that
this information is
relayed back to userspace so it
@Alasdair
It's buried in the description:
We were attempting variants of 'while true; do sudo multipath -F; sudo
multipath -v4 ; done' to create a set of udev add/remove events and
noticed that shortly after starting, no output appeared within the
'udevadm monitor' command.
So it appears that
On Mon, Jul 09, 2012 at 04:33:25PM -, Peter Petrakis wrote:
I'm not familiar with the cookie feature you mentioned, but we appear to have
some of the
capability here. I haven't compared it to upstream yet.
The code issuing the removal ioctl specfies a cookie (a semaphore with a
positive
So when this bug was first reported we were using multipath 0.4.8, it doesn't
appear
to have this cookie mechanism. 0.4.9 does (oneiric and higher); I have been
unable to reproduce the issue with 0.4.9. Going back through the logs...
From 8d63b33d0996e141a2451df552b062b908db15bc Mon Sep 17
I have been looking for a race condition that sometimes causes problems
with dmraid devices. Can you be more specific about the nature of this
race?
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to multipath-tools in ubuntu.
Thanks for the information.
chuck
** Changed in: multipath-tools (Ubuntu)
Importance: Undecided = Medium
** Changed in: multipath-tools (Ubuntu)
Status: New = Triaged
--
Race condition with dmsetup causes 'map already present' messages
https://bugs.launchpad.net/bugs/585027
You
Thank you for taking the time to report this bug and helping to make Ubuntu
better. Please answer these questions:
1. Is this reproducible?
2. If so, what specific steps should we take to recreate this bug? Be as
detailed as possible.
This will help us to find and resolve the problem.
**
Yes, as stated in my original comment, it is reproducible (myself and
Dustin Kirkland), with variants of while true; do sudo multipath -F;
sudo multipath -v4 ; done.
All that is required is a device that multipath will create a
/dev/mapper entry for (use_friendly_names or not) and remove with -F
15 matches
Mail list logo