Re: [Qemu-devel] QMP: RFC: I/O error info & query-stop-reason

[Kevin Wolf]

Am 02.06.2011 20:09, schrieb Luiz Capitulino:
> On Thu, 02 Jun 2011 13:00:04 -0500
> Anthony Liguori <address@hidden> wrote:
> 
>> On 06/02/2011 12:57 PM, Luiz Capitulino wrote:
>>> On Wed, 01 Jun 2011 16:35:03 -0500
>>> Anthony Liguori<address@hidden>  wrote:
>>>
>>>> On 06/01/2011 04:12 PM, Luiz Capitulino wrote:
>>>>> Hi there,
>>>>>
>>>>> There are people who want to use QMP for thin provisioning. That's, the 
>>>>> VM is
>>>>> started with a small storage and when a no space error is triggered, more 
>>>>> space
>>>>> is allocated and the VM is put to run again.
>>>>>
>>>>> QMP has two limitations that prevent people from doing this today:
>>>>>
>>>>> 1. The BLOCK_IO_ERROR doesn't contain error information
>>>>>
>>>>> 2. Considering we solve item 1, we still have to provide a way for clients
>>>>>      to query why a VM stopped. This is needed because clients may miss 
>>>>> the
>>>>>      BLOCK_IO_ERROR event or may connect to the VM while it's already 
>>>>> stopped
>>>>>
>>>>> A proposal to solve both problems follow.
>>>>>
>>>>> A. BLOCK_IO_ERROR information
>>>>> -----------------------------
>>>>>
>>>>> We already have discussed this a lot, but didn't reach a consensus. My 
>>>>> solution
>>>>> is quite simple: to add a stringfied errno name to the BLOCK_IO_ERROR 
>>>>> event,
>>>>> for example (see the "reason" key):
>>>>>
>>>>> { "event": "BLOCK_IO_ERROR",
>>>>>      "data": { "device": "ide0-hd1",
>>>>>                "operation": "write",
>>>>>                "action": "stop",
>>>>>                "reason": "enospc", }
>>>>
>>>> you can call the reason whatever you want, but don't call it stringfied
>>>> errno name :-)
>>>>
>>>> In fact, just make reason "no space".
>>>
>>> You mean, we should do:
>>>
>>>    "reason": "no space"
>>>
>>> Or that we should make it a boolean, like:
>>>
>>>   "no space": true
>>
>>
>> Do we need reason in BLOCK_IO_ERROR if query-block returns this information?
> 
> True, no.
> 
>>> I'm ok with either way. But in case you meant the second one, I guess
>>> we should make "reason" a dictionary so that we can group related
>>> information when we extend the field, for example:
>>>
>>>   "reason": { "no space": false, "no permission": true }

Splitting up enums into a number of booleans looks like a bad idea to
me. It makes things more verbose than they should be, and even worse, it
implies that more than one field could be true.

If this new schema thing doesn't support proper enums, that's something
that should be changed.

>>
>> Why would we ever have "no permission"?
> 
> It's an I/O error. I have a report from a developer who was getting
> the BLOCK_IO_ERROR event and had to debug qemu to know the error cause,
> it turned out to be no permission.

And I want to add that it's a PITA to handle bug report when the only
message you get from qemu is "something went wrong". Sorry, that's not
useful at all. I want to see the real error reason (and at least for
debugging this means, I want to see the errno value/string).

Finding out that it was -EACCES in fact cost me (and QA who ran into the
problem) much more time than it should have. It's simply too much that
you need to attach gdb to find out what really happened.

Kevin