[Xen-bugs] [Bug 743] Unable to start more than a few domU's on RHEL4 amd64 (Error: Backend device not found)

[bugzilla-daemon@xxxxxxxxxxxxxxxxxxx]

http://bugzilla.xensource.com/bugzilla/show_bug.cgi?id=743


andrew.warfield@xxxxxxxxxxxx changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |andrew.warfield@xxxxxxxxxxxx




------- Comment #10 from andrew.warfield@xxxxxxxxxxxx  2006-08-28 18:20 -------
(In reply to comment #7)
> I tried using the blktap driver. First, it is slow; maybe because the
> user-space tools spew a lot of messages to syslog by default. 
> 
> With the blktap driver, after I start around 5 VM's (3 vbd's each), I get
> random failures:
> "Error: Device 0 (vif) could not be connected. Hotplug scripts not working"
> OR
> "Kernel panic - not syncing: VFS: Unable to mount root fs on
> unknown-block(0,0)"
> 
> 
> From syslog:
> ---
> Aug 28 19:32:26 compute-5-24 BLKTAPCTRL:        PID: [1135]
> Aug 28 19:32:26 compute-5-24 logger: /etc/xen/scripts/vif-bridge: Successful
> vif-bridge offline for vif38.0, bridge xenbr0.
> Aug 28 19:32:26 compute-5-24 BLKTAPCTRL: Write_msg called: CTLMSG_PARAMS,
> sending [aio:/usr0/VM/vm7/vm7-fs.img, /usr0/VM/vm7/vm7-fs.img]
> Aug 28 19:32:26 compute-5-24 BLKTAPCTRL: Generated cookie, 40944
> Aug 28 19:32:26 compute-5-24 BLKTAPCTRL: Received CTLMSG_PID_RSP
> Aug 28 19:32:26 compute-5-24 TAPDISK: Tapdisk: Received msg, len 59, type 1,
> UID 40944, drivertype 0
> Aug 28 19:32:26 compute-5-24 BLKTAPCTRL: Read_msg failure - CTLMSG_IMG
> Aug 28 19:32:26 compute-5-24 TAPDISK: Received CTLMSG_PARAMS:
> [/usr0/VM/vm7/vm7-fs.img]
> Aug 28 19:32:26 compute-5-24 BLKTAPCTRL: Unmapping vbd
> Aug 28 19:32:26 compute-5-24 TAPDISK: Loaded driver: name [tapdisk_aio], type
> [0]
> Aug 28 19:32:26 compute-5-24 BLKTAPCTRL: Write_msg called: CTLMSG_CLOSE
> Aug 28 19:32:26 compute-5-24 TAPDISK: XXX: block-aio
> open('/usr0/VM/vm7/vm7-fs.img')
> Aug 28 19:32:26 compute-5-24 BLKTAPCTRL: DEL_DISKTYPE: Freeing entry
> Aug 28 19:32:26 compute-5-24 TAPDISK: Couldn't get fd for AIO poll support. 
> This is probably because your kernel does not have the aio-poll patch applied.
> --
> 
> I have EPOLL compiled into the kernel. I am not sure what other "aio-poll
> patch" the user-space tool is complaining about. I am not sure if this is the
> cause for the failures.
> 

The error that you see about aio-poll indicates that you probably aren't
running the unstable dom0 kernel, which has a patch to make AIO and normal IO
both work through poll(), and the error is printed in response to this patch
being missing.  Are you completely sure that you are running the kernel from
xen-unstable?

Also, can you explain how you are starting your VMs?  Are you firing them all
up at once using a script, or are you doing them one-by-one?

thanks,
a.


-- 
Configure bugmail: 
http://bugzilla.xensource.com/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.

_______________________________________________
Xen-bugs mailing list
Xen-bugs@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-bugs