Open Stack

Hi Pete

> necessary to have the expiration period in your patch. Why not
> update the hashes as soon as the loss is found in any given suffix?

Without expiration period, object-auditor update try to all suffix hash values
and it will take too long time for object-auditor, I guess.
I thught that it is useful for operators to adjust frequency by set
expiration period
in config file.
This is the reason why I added expiration period in this patch.
But if without this functionality, patch will be more simple.
So it may remain a matter of debate.

Best regards,
   You Yamagata

2012/10/26 Pete Zaitcev <zaitcev at redhat.com>:
> On Thu, 25 Oct 2012 15:43:53 +0900
> 山縣陽 <bi.yamagata at gmail.com> wrote:
>
>> Object-replicator checks the suffix hash value between own and remote node's one
>> and if the value is different, object replicator execute rsync.
>>
>> if object files are unexpectedly lost, suffix hash values will be not updated.
>> So object-replicator can't detect the chage, as a result rsync is not triggered.
>
> Thanks for the explanation. I understand now why replicator is not
> the place to fix this problem and why auditor must be involved
> (it invalidates the hash in the node that suffered the loss so that
> the replicator on the node with the data may detect the situation
> and push the missing file). However, I do not follow why it is
> necessary to have the expiration period in your patch. Why not
> update the hashes as soon as the loss is found in any given suffix?
>
> -- Pete