<div dir="ltr">Sure. It will need to be tested appropriately.</div><div class="gmail_extra"><br><div class="gmail_quote">On Thu, Sep 10, 2015 at 4:49 AM, Joseph Qi <span dir="ltr"><<a href="mailto:joseph.qi@huawei.com" target="_blank">joseph.qi@huawei.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hi Junxiao & Sunil,<br>
Your comments would be appreciated.<br>
<br>
Thanks,<br>
Joseph<br>
<div class="HOEnZb"><div class="h5"><br>
On 2015/9/6 21:11, Joseph Qi wrote:<br>
> Comments for dlm_dispatch_work is described below:<br>
> /* Worker function used during recovery. */<br>
><br>
> But actually dlm_worker is used by 4 types of dlm message workers:<br>
> dlm_assert_master_worker<br>
> dlm_deref_lockres_worker<br>
> dlm_request_all_locks_worker<br>
> dlm_mig_lockres_worker<br>
><br>
> And the first 2 are not dlm recovery related. Moreover, it will send<br>
> DLM_ASSERT_MASTER_MSG to all other nodes in dlm_assert_master_worker.<br>
> And it may do a lot of assert master during recovery. In our scenario,<br>
> it is tens of thousands.<br>
> This will delay the recovery because dlm_worker is a single thread<br>
> workqueue and cluster is hanging during dlm recovery.<br>
> So I doubt if we can move the assert master to a new workqueue or just<br>
> use a system workqueue.<br>
> Any suggestions?<br>
><br>
><br>
> _______________________________________________<br>
> Ocfs2-devel mailing list<br>
> <a href="mailto:Ocfs2-devel@oss.oracle.com">Ocfs2-devel@oss.oracle.com</a><br>
> <a href="https://oss.oracle.com/mailman/listinfo/ocfs2-devel" rel="noreferrer" target="_blank">https://oss.oracle.com/mailman/listinfo/ocfs2-devel</a><br>
><br>
><br>
<br>
<br>
</div></div></blockquote></div><br></div>