* Journaler::_flush
@ 2017-11-26 15:19 陶冬冬
2017-11-27 9:11 ` Journaler::_flush Yan, Zheng
0 siblings, 1 reply; 2+ messages in thread
From: 陶冬冬 @ 2017-11-26 15:19 UTC (permalink / raw)
To: Ceph Development; +Cc: Yan, Zheng, pdonnell
Hi Zheng,
i have one question about Journaler flush,
suppose there are two in flight flushes ?
first is flushing (a, b), second is flushing (b+1, c)
let's assume these two range belong to two adjacent objects.
is that possible that the second flush got the reply first ?
if that so, the safe_pos would be wrong ?
Thanks,
Dongdong
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: Journaler::_flush
2017-11-26 15:19 Journaler::_flush 陶冬冬
@ 2017-11-27 9:11 ` Yan, Zheng
0 siblings, 0 replies; 2+ messages in thread
From: Yan, Zheng @ 2017-11-27 9:11 UTC (permalink / raw)
To: 陶冬冬; +Cc: Ceph Development, Patrick Donnelly
On Sun, Nov 26, 2017 at 11:19 PM, 陶冬冬 <tdd21151186@gmail.com> wrote:
> Hi Zheng,
>
> i have one question about Journaler flush,
> suppose there are two in flight flushes ?
> first is flushing (a, b), second is flushing (b+1, c)
> let's assume these two range belong to two adjacent objects.
> is that possible that the second flush got the reply first ?
yes
> if that so, the safe_pos would be wrong ?
>
no. check Journaler::_finish_flush
> Thanks,
> Dongdong
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2017-11-27 9:11 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2017-11-26 15:19 Journaler::_flush 陶冬冬
2017-11-27 9:11 ` Journaler::_flush Yan, Zheng
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.