From: Arun Sharma <arun.sharma@intel.com>
To: Anthony Liguori <aliguori@us.ibm.com>
Cc: Mike Wray <mike.wray@hp.com>, xen-devel <xen-devel@lists.xensource.com>
Subject: Re: exception looking up device number f,or hda
Date: Fri, 05 Aug 2005 14:20:06 -0700 [thread overview]
Message-ID: <42F3D806.9070201@intel.com> (raw)
In-Reply-To: <42F3CCFE.8010302@us.ibm.com>
Anthony Liguori wrote:
> Hi Arun,
>
> I just pulled the latest unstable and could not reproduce. I have an
> identical scenario (no /dev/hda only /dev/sda due to SATA).
>
> Also, if the devno resolution code fails you should also get a vbd:
> Device not found exception in the logs. The exception you are seeing is
> not actually a problem, it just means that stat has failed.
>
> We should probably cleanup that error message.
>
> Can you post a more complete log file? I think the error is being
> caused somewhere else.
>
Hi Anthony,
You're probably not able to reproduce the problem because you're not
using VMX domains?
We're overloading the semantics of
disk = [ 'file:/var/images/min-el3-i386.img,hda,w' ]
to create a IDE hard drive for a VMX domain. So we want to take
advantage of the config file parsing code from the VBD code path, but we
don't want to do other tasks such as connecting the backend/frontend etc.
The log from a successful VMX domain startup looks like:
[2005-08-05 14:54:41 xend] INFO (XendRoot:126) EVENT> xend.domain.create
['fc3', 2]
[2005-08-05 14:54:42 xend] ERROR (process:37) [block-file] ioctl:
LOOP_SET_FD: Device or resource busy
[2005-08-05 14:54:42 xend] DEBUG (blkif:81) Connecting blkif
<BlkifBackend frontend=2 backend=0 id=0>
[2005-08-05 14:54:42 xend] DEBUG (blkif:89) send_be_create <BlkifBackend
frontend=2 backend=0 id=0>
[2005-08-05 14:54:42 xend] DEBUG (image:293) args: cdrom, val: None
[2005-08-05 14:54:42 xend] DEBUG (image:293) args: boot, val: c
[2005-08-05 14:54:42 xend] DEBUG (image:293) args: fda, val: None
[2005-08-05 14:54:42 xend] DEBUG (image:293) args: fdb, val: None
[2005-08-05 14:54:42 xend] DEBUG (image:293) args: localtime, val: None
[2005-08-05 14:54:42 xend] DEBUG (image:293) args: serial, val: None
[2005-08-05 14:54:42 xend] DEBUG (image:293) args: std-vga, val: None
[2005-08-05 14:54:42 xend] DEBUG (image:293) args: isa, val: None
[2005-08-05 14:54:42 xend] INFO (image:352) spawning device models:
/usr/lib/xen/bin/qemu-dm ['/usr/lib/xen/bin/qemu-dm', '
-vncconnect', '143.183.130.223:5502', '-d', '2', '-p', '19', '-m',
'128', '-boot', 'c', '-hda', '/var/images/min-el3-i386.i
mg', '-vnc', '-k', 'en-us', '-vncport', '5902']
i.e. we were doing things which we really shouldn't be doing, but it
didn't hurt. But it started hurting today :)
In the future, we're also planning to have a VBD driver for unmodified
OSes running on VMX domains and I was thinking of a syntax such as:
disk = [ 'file:/var/images/min-el3-i386.img,vhda,w' ]
Any comments/suggestions on the syntax above to present emulated real
devices as well as virtual devices to an unmodified OS?
-Arun
Complete log for the failed case:
[2005-08-05 14:40:33 xend] INFO (SrvDaemon:300) Xend Daemon started
[2005-08-05 14:40:33 xend] INFO (channel:148) Virq 3 on port 13
[2005-08-05 14:40:33 xend] DEBUG (XendDomainInfo:171)
info={'vcpu_to_cpu': [0, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
-1, -1], 'cpu_time': 20984020436L, 'ssidref': 0, 's
hutdown_reason': 0, 'dying': 0, 'dom': 0, 'running': 1, 'maxmem_kb': -4,
'paused': 0, 'crashed': 0, 'vcpus': 1, 'shutdown':
0, 'mem_kb': 250880, 'cpumap': [-1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0], 'blocked': 0}
[2005-08-05 14:40:33 xend] DEBUG (XendDomainInfo:172) config='None'
[2005-08-05 14:40:33 xend] INFO (SrvServer:63) unix
path=/var/lib/xend/xend-socket
[2005-08-05 14:40:33 xend] INFO (XendRoot:126) EVENT> xend.start 0
[2005-08-05 14:40:33 xend] INFO (process:35) [network] network start
bridge=xen-br0 netdev=eth0 antispoof=no
[2005-08-05 14:40:53 xend] DEBUG (image:135) initDomain: cpu=-1
mem_kb=131204 ssidref=-1 dom=1
[2005-08-05 14:40:53 xend] DEBUG (XendDomainInfo:673) init_domain>
Created domain=1 name=ExampleVMXDomain memory=128
[2005-08-05 14:40:53 xend] INFO (XendRoot:126) EVENT> xend.domain.create
['ExampleVMXDomain', 1]
[2005-08-05 14:40:53 xend] INFO (image:190) buildDomain os=vmx dom=1 vcpus=1
[2005-08-05 14:40:53 xend] INFO (image:259) VMX device model port: 3
[2005-08-05 14:40:53 xend] WARNING (XendDomainInfo:924) Unknown config
field memmap
[2005-08-05 14:40:53 xend] WARNING (XendDomainInfo:924) Unknown config
field device_model
[2005-08-05 14:40:53 xend] WARNING (XendDomainInfo:924) Unknown config
field boot
[2005-08-05 14:40:53 xend] WARNING (XendDomainInfo:924) Unknown config
field vnc
[2005-08-05 14:40:53 xend] WARNING (XendDomainInfo:924) Unknown config
field display
[2005-08-05 14:40:53 xend] DEBUG (blkif:24) exception looking up device
number for hda: [Errno 2] No such file or directory
: '/dev/hda'
[2005-08-05 14:40:53 xend] DEBUG (blkif:449) Destroying blkif domain=1
[2005-08-05 14:40:53 xend] DEBUG (blkif:337) Destroying vbd domain=1 id=0
>>>> xend dies here <<<
[2005-08-05 14:43:32 xend] INFO (SrvDaemon:300) Xend Daemon started
next prev parent reply other threads:[~2005-08-05 21:20 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-08-05 19:34 exception looking up device number f,or hda Arun Sharma
2005-08-05 20:33 ` Anthony Liguori
2005-08-05 21:20 ` Arun Sharma [this message]
2005-08-08 2:21 ` takebe_akio
2005-08-08 5:38 ` exception looking up device number for hda Arun Sharma
2005-08-08 14:48 ` Mark Williamson
2005-08-09 0:06 ` Arun Sharma
2005-08-09 1:25 ` David Hopwood
2005-08-09 1:49 ` Mark Williamson
2005-08-09 2:58 ` Xiaofeng Ling
2005-08-09 18:19 ` Arun Sharma
[not found] ` <mailman.1123552364.22306@unix-os.sc.intel.com>
2005-08-09 19:06 ` Arun Sharma
2005-08-09 19:18 ` Mark Williamson
2005-08-08 8:08 ` exception looking up device number f,or hda Mike Wray
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=42F3D806.9070201@intel.com \
--to=arun.sharma@intel.com \
--cc=aliguori@us.ibm.com \
--cc=mike.wray@hp.com \
--cc=xen-devel@lists.xensource.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.