[libvirt] [PATCH 5/5] qemu: Add better message for some invalid block I/O settings

Martin Kletzander mkletzan at redhat.com
Wed Jan 25 14:07:03 UTC 2017


On Wed, Jan 25, 2017 at 07:38:26AM -0500, John Ferlan wrote:
>
>
>On 01/25/2017 04:16 AM, Martin Kletzander wrote:
>> For example when both total_bytes_sec and total_bytes_sec_max are set,
>> but the former gets cleaned due to new call setting, let's say,
>> read_bytes_sec, we end up with this weird message for the command:
>>
>>  $ virsh blkdeviotune fedora vda --read-bytes-sec 3000
>>  error: Unable to change block I/O throttle
>>  error: unsupported configuration: value 'total_bytes_sec_max' cannot be set if 'total_bytes_sec' is not set
>>
>> So let's make it more descriptive.  This is how it looks after the change:
>>
>>  $ virsh blkdeviotune fedora vda --read-bytes-sec 3000
>>  error: Unable to change block I/O throttle
>>  error: unsupported configuration: cannot reset 'total_bytes_sec' when 'total_bytes_sec_max' is set
>>
>> Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1344897
>>
>> Signed-off-by: Martin Kletzander <mkletzan at redhat.com>
>> ---
>>  src/qemu/qemu_driver.c | 46 +++++++++++++++++++++++++++++++---------------
>>  1 file changed, 31 insertions(+), 15 deletions(-)
>>
>
>At least this is easier with those macros... I'm fine with the error
>message adjustments here - although I thought QEMU checked the various
>inconsistencies and messaged based on that (perhaps something I checked
>when using the qemu command line instead of via virsh - cannot recall now).
>

It does, but we don't handle all types of errors very well.  Also it's
better to error out earlier, what if some version of QEMU will set half
of the settings before failing?

Anyway, QEMU checked that, but it ended up like this (can be
triggerredfor example by setting *bytes_sec to more than
*bytes_sec_max):

 $ virsh blkdeviotune vm2 vda --total-bytes-sec 3000
 error: Unable to change block I/O throttle
 error: internal error: Unexpected error

See the BZ for details.  I didn't want to make the commit messages
longer than the (self-describing) patches themselves.

>ACK,
>
>John
>
>oh and before I forget... Could you please update:
>
>https://bugzilla.redhat.com/show_bug.cgi?id=1336564
>
>to indicate which commit fixes the issues for group name. Thanks and
>sorry for the mess.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: Digital signature
URL: <http://listman.redhat.com/archives/libvir-list/attachments/20170125/3b1ce39f/attachment-0001.sig>


More information about the libvir-list mailing list