Re: [Patch] blkiomon: repost with some fixes and improvements

Re: [Patch] blkiomon: repost with some fixes and improvements

[Alan D. Brunelle]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1382 bytes --]

T�r�k Edwin wrote:
> On 2008-07-21 21:56, T�r�k Edwin wrote:
>> On 2008-07-21 21:45, Alan D. Brunelle wrote:
>>   
>>> Hi Edwin -
>>>
>>> With the patches sent out today (kernel & application), you can then use
>>> the updated script attached here. It only asks for getrq & sleeprq
>>> traces - so it will cut down a lot, but most likely there will still be
>>> a lot of getrq's (in particular).
>>>
>>> Will now have some time to look at the more general issue concerning how
>>> to see the effects of the sleeprq's...
>>>   
>>>     
>>   
> 
> I think it would be useful to somehow "dump" the contents of both queues
> when a sleep occurs, at least
> the rw_flags, the pid, and the size of the operation.
> Maybe that could be done somehow via blktrace on-demand? [it already has
> all the relayfs communication ...]
> 
> P.S.: I tried to blktrace all queueing events and save that to the disk,
> but no matter what buffer size I used,
> it kept telling me it has dropped events :(
> So the only way to know the contents of the queue is for the kernel to
> give it to me on-demand.

Hi Edwin -

That's really strange: I very rarely have that problem - I find if I
increase the number and size of buffers, those problems go away. (But, I
tend to have a lot of memory to play with too...)

You are dumping to another disk (other than the one you are tracing), right?

Alan