https://jira.whamcloud.com This file is an XML representation of an issue en-us 9.4.14 940014 05-12-2023 https://jira.whamcloud.com/browse/LU-1571 Lustre <p>In recovery, it should not update last xid for open replay, otherwise it will mess the last xid in the last rcvd, which make the following resend (after replay) failed.</p> LU-1571

It should not update last xid for replay open.

Bug Blocker Resolved Fixed Oleg Drokin Di Wang Wed, 27 Jun 2012 03:31:14 +0000 Mon, 17 Mar 2014 17:51:25 +0000 Sun, 24 Feb 2013 19:25:08 +0000 Lustre 2.4.0 Lustre 2.4.0 Lustre 2.1.5 0 1 <p><a href="http://review.whamcloud.com/3195" class="external-link" target="_blank" rel="nofollow noopener">http://review.whamcloud.com/3195</a></p> <p>Considering following case. 1. client open the file(with xid_1), keep the replay open req on the client. 2. client mkdir(with xid_2), server finish the create and drop reply, but update last_rcvd_xid with xid_2.</p> <p>MDT restart, init last_rcvd_xid with xid_2</p> <p>1. client replay open req. then update last_rcvd_xid with xid_1</p> <p>2. then replay finish, resend mkdir req. but last_rcvd_xid has been changed to xid_1. so it can not recognize the resend.</p> <p>We need to update last_rcvd_xid only if open transno is bigger than one in last_rcvd, otherwise that means open wasn't last request from client and doesn't need reconstruct. Please check also <a href="https://jira.whamcloud.com/browse/LU-971" title="Re-sent open requests cause mod_count leakages" class="issue-link" data-issue-key="LU-971"><del>LU-971</del></a> bug, there was problem with that so we update last_rcvd data for open too but in memory only, looks like it causes <a href="https://jira.whamcloud.com/browse/LU-1571" title="It should not update last xid for replay open." class="issue-link" data-issue-key="LU-1571"><del>LU-1571</del></a> now.</p> <p>Landed for 2.4</p> Duplicate LU-2277 Development Rank 1|hzv2yv: Rank (Obsolete) 3980 Severity