From mboxrd@z Thu Jan  1 00:00:00 1970
From: Roland Paterson-Jones <roland@rolandpj.com>
Subject: Re: Re: [Xen-devel] tap:qcow causes dom-U to hang in 3.0.3
Date: Tue, 14 Nov 2006 12:50:14 +0200
Message-ID: <45599F66.6050008@rolandpj.com>
References: <4551EEC3.3010308@rolandpj.com>	<20061108151133.GE3507@leeni.uk.xensource.com>	<4552DF8D.6060600@rolandpj.com>	<ab5e402899c3a334735d24815f36eaac@cam.ac.uk>	<455485FF.7040809@rolandpj.com>	<1a16a93e920ce8db7c8ed8cd908a876b@cam.ac.uk>
	<45599614.3040103@rolandpj.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Return-path: <xen-users-bounces@lists.xensource.com>
In-Reply-To: <45599614.3040103@rolandpj.com>
List-Unsubscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-users>,
	<mailto:xen-users-request@lists.xensource.com?subject=unsubscribe>
List-Post: <mailto:xen-users@lists.xensource.com>
List-Help: <mailto:xen-users-request@lists.xensource.com?subject=help>
List-Subscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-users>,
	<mailto:xen-users-request@lists.xensource.com?subject=subscribe>
Sender: xen-users-bounces@lists.xensource.com
Errors-To: xen-users-bounces@lists.xensource.com
To: Roland Paterson-Jones <roland@rolandpj.com>
Cc: Julian Chesterfield <jac90@cam.ac.uk>, Xen Devel <xen-devel@lists.xensource.com>, xen-users@lists.xensource.com
List-Id: xen-devel@lists.xenproject.org

Roland Paterson-Jones wrote:

> Julian Chesterfield wrote:
>
>>
>> There's a bunch of DPRINTFs in the code which send debug output to 
>> syslog. You can play with syslog settings to switch on the debug 
>> messages and direct them to a log file.
>>
> Here is another clue from syslog. Below is the full log, but the hang 
> seems to accur when the following log messages are triggered (there's 
> a bunch of them at the end):
>
> Nov 14 11:59:47 dom0-0-50-45-5d-6a-bc TAPDISK: AIO_LOCK or 
> iocb_free_count (0) failed[869040]
>
And when I try to 'xm destroy' the instance after it hangs, I get a slew 
of the following :

Nov 14 12:41:35 dom0-0-50-45-5d-6a-bc kernel: blk_tap: invalid kernel 
buffer -- could not remap it
Nov 14 12:41:35 dom0-0-50-45-5d-6a-bc kernel: blk_tap: Reached Fail_flush
Nov 14 12:41:35 dom0-0-50-45-5d-6a-bc kernel: blk_tap: BLKTAP_INVALID_HANDLE

This is the inverse of the error messages I see when I build the 
user-land tools myself. In the above it's the kernel buffer that can't 
be re-mapped. With my user-land tools it's the user buffer that can't be 
remapped. I wonder (but have no idea) whether this is related to PAE at all.

Roland