All of lore.kernel.org
 help / color / mirror / Atom feed
* [BUG] libxl: error: libxl_device.c:301:libxl__device_disk_set_backend: no suitable backend for disk
@ 2016-02-03 23:53 Alex Braunegg
  2016-02-04 11:25 ` Ian Campbell
  0 siblings, 1 reply; 6+ messages in thread
From: Alex Braunegg @ 2016-02-03 23:53 UTC (permalink / raw)
  To: xen-devel

Hi all,

I have been testing the Xen 4.6.0 packages - however I have found the
following issue with attempting to perform the following commands:

	[root@mynas-s5000xvn ~]# xl cd-eject test2 hdc         
	libxl: error: libxl_device.c:301:libxl__device_disk_set_backend: no
suitable backend for disk hdc
	[root@mynas-s5000xvn ~]#

Verbose logging of the issue presents the following:

----------------------------------------------------

libxl: debug: libxl.c:2841:libxl_cdrom_insert: ao 0xe68c30: create:
how=(nil) callback=(nil) poller=0xe68cc0
libxl: debug: libxl_device.c:269:libxl__device_disk_set_backend: Disk
vdev=hdc spec.backend=unknown
libxl: debug: libxl_device.c:215:disk_try_backend: Disk vdev=hdc, backend
phy unsuitable as phys path not a block device
libxl: debug: libxl_device.c:298:libxl__device_disk_set_backend: Disk
vdev=hdc, using backend qdisk
libxl: debug: libxl_device.c:269:libxl__device_disk_set_backend: Disk
vdev=hdc spec.backend=phy
libxl: debug: libxl_device.c:215:disk_try_backend: Disk vdev=hdc, backend
phy unsuitable as phys path not a block device
libxl: error: libxl_device.c:301:libxl__device_disk_set_backend: no suitable
backend for disk hdc
libxl: debug: libxl_event.c:1853:libxl__ao_create_fail: ao 0xe68c30: create
fail
libxl: debug: libxl_event.c:1843:libxl__ao__destroy: ao 0xe68c30: destroy
xc: debug: hypercall buffer: total allocations:20 total releases:20
xc: debug: hypercall buffer: current allocations:0 maximum allocations:2
xc: debug: hypercall buffer: cache current size:2
xc: debug: hypercall buffer: cache hits:12 misses:2 toobig:6

----------------------------------------------------

There was a thread on this in 2013
(http://lists.xen.org/archives/html/xen-devel/2013-03/msg00806.html) however
I have looked at the disk configuration, and there is no phy: / raw: / file:
prefix to the disk specification.

The VM configuration is as follows:

----------------------------------------------------

builder='hvm'
memory = 512
shadow_memory = 8
uuid = '4789b84d-3cdd-409f-9899-88bb13652176'
name = 'test2'
vif = [ 'bridge=br0, mac=00:16:3e:f1:48:8c' ]
disk = [
'/dev/zvol/storage0/xen/test2/disk_sda,,hda','/storage0/data-shares/iso/Cent
OS-6.5-x86_64-minimal.iso,,hdc,cdrom' ]
boot='dc'
sdl=0
vnc=1
vncconsole=1
vnclisten='0.0.0.0'
stdvga=1
serial='pty'
usbdevice='tablet'
vncpasswd='asdf1234'
vncdisplay=0
localtime=1
audio='1'
soundhw='ac97'

----------------------------------------------------

Additionally, if I attempt to insert the ISO file (which I expect it to fail
as I have not ejected it yet), I get some additional information which may
be beneficial in diagnosing the eject issue:

----------------------------------------------------
[root@mynas-s5000xvn ~]# xl -vvvv cd-insert test2 hdc
/storage0/data-shares/iso/CentOS-6.5-x86_64-minimal.iso
libxl: debug: libxl.c:2841:libxl_cdrom_insert: ao 0x101cc30: create:
how=(nil) callback=(nil) poller=0x101ccc0
libxl: debug: libxl_device.c:269:libxl__device_disk_set_backend: Disk
vdev=hdc spec.backend=unknown
libxl: debug: libxl_device.c:215:disk_try_backend: Disk vdev=hdc, backend
phy unsuitable as phys path not a block device
libxl: debug: libxl_device.c:298:libxl__device_disk_set_backend: Disk
vdev=hdc, using backend qdisk
libxl: debug: libxl_device.c:269:libxl__device_disk_set_backend: Disk
vdev=hdc spec.backend=phy
libxl: debug: libxl_qmp.c:706:libxl__qmp_initialize: connected to
/var/run/xen/qmp-libxl-37
libxl: debug: libxl_qmp.c:296:qmp_handle_response: message type: qmp
libxl: debug: libxl_qmp.c:555:qmp_send_prepare: next qmp command: '{
    "execute": "qmp_capabilities",
    "id": 1
}
'
libxl: debug: libxl_qmp.c:296:qmp_handle_response: message type: return
libxl: debug: libxl_qmp.c:555:qmp_send_prepare: next qmp command: '{
    "execute": "eject",
    "id": 2,
    "arguments": {
        "device": "ide-5632"
    }
}
'
libxl: debug: libxl_qmp.c:296:qmp_handle_response: message type: error
libxl: error: libxl_qmp.c:287:qmp_handle_error_response: received an error
message from QMP server: Device 'ide-5632' is locked
libxl: debug: libxl_event.c:1853:libxl__ao_create_fail: ao 0x101cc30: create
fail
libxl: debug: libxl_event.c:1843:libxl__ao__destroy: ao 0x101cc30: destroy
xc: debug: hypercall buffer: total allocations:26 total releases:26
xc: debug: hypercall buffer: current allocations:0 maximum allocations:2
xc: debug: hypercall buffer: cache current size:2
xc: debug: hypercall buffer: cache hits:17 misses:2 toobig:7
[root@mynas-s5000xvn ~]#
----------------------------------------------------


Any suggestions is greatly appreciated.


Best regards,

Alex

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [BUG] libxl: error: libxl_device.c:301:libxl__device_disk_set_backend: no suitable backend for disk
  2016-02-03 23:53 [BUG] libxl: error: libxl_device.c:301:libxl__device_disk_set_backend: no suitable backend for disk Alex Braunegg
@ 2016-02-04 11:25 ` Ian Campbell
  2016-02-04 19:19   ` Alex Braunegg
                     ` (2 more replies)
  0 siblings, 3 replies; 6+ messages in thread
From: Ian Campbell @ 2016-02-04 11:25 UTC (permalink / raw)
  To: Alex Braunegg, xen-devel

On Thu, 2016-02-04 at 10:53 +1100, Alex Braunegg wrote:
> 
> There was a thread on this in 2013
> (http://lists.xen.org/archives/html/xen-devel/2013-03/msg00806.html) however
> I have looked at the disk configuration, and there is no phy: / raw: / file:
> prefix to the disk specification.

Not sure what is going on but I just wanted to mention that phy: etc
prefixes are considered legacy. See http://xenbits.xen.org/docs/4.6-testing
/misc/xl-disk-configuration.txt for the actual syntax and how phy: is
mapped.

Ian.

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [BUG] libxl: error: libxl_device.c:301:libxl__device_disk_set_backend: no suitable backend for disk
  2016-02-04 11:25 ` Ian Campbell
@ 2016-02-04 19:19   ` Alex Braunegg
  2016-02-08 21:48   ` Alex Braunegg
  2016-02-10  1:03   ` Alex Braunegg
  2 siblings, 0 replies; 6+ messages in thread
From: Alex Braunegg @ 2016-02-04 19:19 UTC (permalink / raw)
  To: 'Ian Campbell', xen-devel

Hi Ian,

As per http://xenbits.xen.org/docs/4.6-testing/misc/xl-disk-configuration.txt the disk specification conforms to:

	disk = [ '/dev/vg/guest-volume,,hda', '/root/image.iso,,hdc,cdrom' ]

Is there any additional information you need to replicate this issue?

Best regards,

Alex


-----Original Message-----
From: Ian Campbell [mailto:ian.campbell@citrix.com] 
Sent: Thursday, 4 February 2016 10:26 PM
To: Alex Braunegg; xen-devel@lists.xen.org
Subject: Re: [Xen-devel] [BUG] libxl: error: libxl_device.c:301:libxl__device_disk_set_backend: no suitable backend for disk

On Thu, 2016-02-04 at 10:53 +1100, Alex Braunegg wrote:
> 
> There was a thread on this in 2013
> (http://lists.xen.org/archives/html/xen-devel/2013-03/msg00806.html) however
> I have looked at the disk configuration, and there is no phy: / raw: / file:
> prefix to the disk specification.

Not sure what is going on but I just wanted to mention that phy: etc
prefixes are considered legacy. See http://xenbits.xen.org/docs/4.6-testing
/misc/xl-disk-configuration.txt for the actual syntax and how phy: is
mapped.

Ian.

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [BUG] libxl: error: libxl_device.c:301:libxl__device_disk_set_backend: no suitable backend for disk
  2016-02-04 11:25 ` Ian Campbell
  2016-02-04 19:19   ` Alex Braunegg
@ 2016-02-08 21:48   ` Alex Braunegg
  2016-02-09 12:00     ` Roger Pau Monné
  2016-02-10  1:03   ` Alex Braunegg
  2 siblings, 1 reply; 6+ messages in thread
From: Alex Braunegg @ 2016-02-08 21:48 UTC (permalink / raw)
  To: 'Ian Campbell', xen-devel

Hi Ian,

Any update in looking at this issue?

Best regards,

Alex

-----Original Message-----
From: Alex Braunegg [mailto:alex.braunegg@gmail.com] 
Sent: Friday, 5 February 2016 6:19 AM
To: 'Ian Campbell'; 'xen-devel@lists.xen.org'
Subject: RE: [Xen-devel] [BUG] libxl: error: libxl_device.c:301:libxl__device_disk_set_backend: no suitable backend for disk

Hi Ian,

As per http://xenbits.xen.org/docs/4.6-testing/misc/xl-disk-configuration.txt the disk specification conforms to:

	disk = [ '/dev/vg/guest-volume,,hda', '/root/image.iso,,hdc,cdrom' ]

Is there any additional information you need to replicate this issue?

Best regards,

Alex


-----Original Message-----
From: Ian Campbell [mailto:ian.campbell@citrix.com] 
Sent: Thursday, 4 February 2016 10:26 PM
To: Alex Braunegg; xen-devel@lists.xen.org
Subject: Re: [Xen-devel] [BUG] libxl: error: libxl_device.c:301:libxl__device_disk_set_backend: no suitable backend for disk

On Thu, 2016-02-04 at 10:53 +1100, Alex Braunegg wrote:
> 
> There was a thread on this in 2013
> (http://lists.xen.org/archives/html/xen-devel/2013-03/msg00806.html) however
> I have looked at the disk configuration, and there is no phy: / raw: / file:
> prefix to the disk specification.

Not sure what is going on but I just wanted to mention that phy: etc
prefixes are considered legacy. See http://xenbits.xen.org/docs/4.6-testing
/misc/xl-disk-configuration.txt for the actual syntax and how phy: is
mapped.

Ian.

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [BUG] libxl: error: libxl_device.c:301:libxl__device_disk_set_backend: no suitable backend for disk
  2016-02-08 21:48   ` Alex Braunegg
@ 2016-02-09 12:00     ` Roger Pau Monné
  0 siblings, 0 replies; 6+ messages in thread
From: Roger Pau Monné @ 2016-02-09 12:00 UTC (permalink / raw)
  To: Alex Braunegg, 'Ian Campbell', xen-devel

El 8/2/16 a les 22:48, Alex Braunegg ha escrit:
> Hi Ian,
> 
> Any update in looking at this issue?

Hello,

I've been able to reproduce this issue and I will prepare a fix. I'm
going to Cc you on the patch if you don't mind, so that you can test it.

Roger.

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [BUG] libxl: error: libxl_device.c:301:libxl__device_disk_set_backend: no suitable backend for disk
  2016-02-04 11:25 ` Ian Campbell
  2016-02-04 19:19   ` Alex Braunegg
  2016-02-08 21:48   ` Alex Braunegg
@ 2016-02-10  1:03   ` Alex Braunegg
  2 siblings, 0 replies; 6+ messages in thread
From: Alex Braunegg @ 2016-02-10  1:03 UTC (permalink / raw)
  To: 'Ian Campbell', xen-devel

Hi Ian

Is there anything I can do to help here at all to provide further details? I am happy to do any local side diagnosing on this issue.

When performing validation testing of Xen code from a developers point of view - what packages / versions are these tests performed against? Perhaps there is something local to my system that is contributing to xl not being able to issue the cd-eject / cd-insert command properly. Maybe a kernel config option relating to Xen?

Kernel / Xen / Qemu package list below:

[root@mynas-s5000xvn ~]# rpm -qa | grep xen
mynas-xen-drivers-1.4.0-201601271353.el6.x86_64
xen-runtime-4.6.0-12.el6.x86_64
xen-doc-4.6.0-11.el6.x86_64
xen-hypervisor-4.6.0-12.el6.x86_64
xen-ocaml-4.6.0-12.el6.x86_64
xen-4.6.0-12.el6.x86_64
xen-licenses-4.6.0-12.el6.x86_64
xen-libs-4.6.0-12.el6.x86_64

[root@mynas-s5000xvn ~]# rpm -qa | grep qemu
qemu-img-0.12.1.2-2.479.el6_7.4.x86_64

[root@mynas-s5000xvn ~]# rpm -qa | grep kernel
kernel-firmware-4.3.4-2.el6.x86_64
kernel-4.3.4-1.el6.x86_64
kernel-4.3.4-2.el6.x86_64
dracut-kernel-004-388.el6.noarch
kernel-3.14.58-3.el6.x86_64

[root@mynas-s5000xvn ~]# uname -a
Linux mynas-s5000xvn.localdomain 4.3.4-2.el6.x86_64 #1 SMP Tue Feb 9 10:38:50 EST 2016 x86_64 x86_64 x86_64 GNU/Linux
[root@mynas-s5000xvn ~]# 

[mockbuild@buildserver SOURCES]$ cat config-4.3.4-x86_64 | grep XEN
CONFIG_XEN=y
CONFIG_XEN_DOM0=y
CONFIG_XEN_PVHVM=y
CONFIG_XEN_512GB=y
CONFIG_XEN_SAVE_RESTORE=y
CONFIG_XEN_DEBUG_FS=y
CONFIG_XEN_PVH=y
CONFIG_PCI_XEN=y
CONFIG_XEN_PCIDEV_FRONTEND=y
CONFIG_XEN_BLKDEV_FRONTEND=y
CONFIG_XEN_BLKDEV_BACKEND=y
CONFIG_XEN_SCSI_FRONTEND=m
CONFIG_NETXEN_NIC=m
CONFIG_XEN_NETDEV_FRONTEND=m
CONFIG_XEN_NETDEV_BACKEND=m
CONFIG_INPUT_XEN_KBDDEV_FRONTEND=y
CONFIG_HVC_XEN=y
CONFIG_HVC_XEN_FRONTEND=y
CONFIG_TCG_XEN=m
CONFIG_XEN_WDT=m
CONFIG_XEN_BALLOON=y
CONFIG_XEN_SELFBALLOONING=y
CONFIG_XEN_BALLOON_MEMORY_HOTPLUG=y
CONFIG_XEN_BALLOON_MEMORY_HOTPLUG_LIMIT=512
CONFIG_XEN_SCRUB_PAGES=y
CONFIG_XEN_DEV_EVTCHN=y
CONFIG_XEN_BACKEND=y
CONFIG_XENFS=y
CONFIG_XEN_COMPAT_XENFS=y
CONFIG_XEN_SYS_HYPERVISOR=y
CONFIG_XEN_XENBUS_FRONTEND=y
CONFIG_XEN_GNTDEV=y
CONFIG_XEN_GRANT_DEV_ALLOC=y
CONFIG_SWIOTLB_XEN=y
CONFIG_XEN_TMEM=m
CONFIG_XEN_PCIDEV_BACKEND=y
CONFIG_XEN_SCSI_BACKEND=m
CONFIG_XEN_PRIVCMD=y
CONFIG_XEN_ACPI_PROCESSOR=y
CONFIG_XEN_MCE_LOG=y
CONFIG_XEN_HAVE_PVMMU=y
CONFIG_XEN_EFI=y
CONFIG_XEN_AUTO_XLATE=y
CONFIG_XEN_ACPI=y
CONFIG_XEN_SYMS=y
CONFIG_XEN_HAVE_VPMU=y


I had the same issue with xen-4.4.1 in regards to xl & the cd-insert / cd-eject commands. The work around for me was to use the following:

	"/usr/sbin/xm block-configure $selectedVM file:$isoFile hdc:cdrom r"

Which would mount the ISO no issue in the host OS, however xm is depreciated and there is no equivalent command in xl - hence trying to get the correct command to actually work.

Best regards,

Alex

-----Original Message-----
From: Alex Braunegg [mailto:alex.braunegg@gmail.com] 
Sent: Tuesday, 9 February 2016 8:49 AM
To: 'Ian Campbell'; 'xen-devel@lists.xen.org'
Subject: RE: [Xen-devel] [BUG] libxl: error: libxl_device.c:301:libxl__device_disk_set_backend: no suitable backend for disk

Hi Ian,

Any update in looking at this issue?

Best regards,

Alex

-----Original Message-----
From: Alex Braunegg [mailto:alex.braunegg@gmail.com] 
Sent: Friday, 5 February 2016 6:19 AM
To: 'Ian Campbell'; 'xen-devel@lists.xen.org'
Subject: RE: [Xen-devel] [BUG] libxl: error: libxl_device.c:301:libxl__device_disk_set_backend: no suitable backend for disk

Hi Ian,

As per http://xenbits.xen.org/docs/4.6-testing/misc/xl-disk-configuration.txt the disk specification conforms to:

	disk = [ '/dev/vg/guest-volume,,hda', '/root/image.iso,,hdc,cdrom' ]

Is there any additional information you need to replicate this issue?

Best regards,

Alex


-----Original Message-----
From: Ian Campbell [mailto:ian.campbell@citrix.com] 
Sent: Thursday, 4 February 2016 10:26 PM
To: Alex Braunegg; xen-devel@lists.xen.org
Subject: Re: [Xen-devel] [BUG] libxl: error: libxl_device.c:301:libxl__device_disk_set_backend: no suitable backend for disk

On Thu, 2016-02-04 at 10:53 +1100, Alex Braunegg wrote:
> 
> There was a thread on this in 2013
> (http://lists.xen.org/archives/html/xen-devel/2013-03/msg00806.html) however
> I have looked at the disk configuration, and there is no phy: / raw: / file:
> prefix to the disk specification.

Not sure what is going on but I just wanted to mention that phy: etc
prefixes are considered legacy. See http://xenbits.xen.org/docs/4.6-testing
/misc/xl-disk-configuration.txt for the actual syntax and how phy: is
mapped.

Ian.

^ permalink raw reply	[flat|nested] 6+ messages in thread

end of thread, other threads:[~2016-02-10  1:03 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2016-02-03 23:53 [BUG] libxl: error: libxl_device.c:301:libxl__device_disk_set_backend: no suitable backend for disk Alex Braunegg
2016-02-04 11:25 ` Ian Campbell
2016-02-04 19:19   ` Alex Braunegg
2016-02-08 21:48   ` Alex Braunegg
2016-02-09 12:00     ` Roger Pau Monné
2016-02-10  1:03   ` Alex Braunegg

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.