From: Xiubo Li <xiubli@redhat.com>
Date: Wed, 19 May 2021 07:54:12 +0000 (+0800)
Subject: mds: progress the recover queue immediately after the inode is enqueued
X-Git-Tag: v15.2.17~35^2
X-Git-Url: http://git-server-git.apps.pok.os.sepia.ceph.com/?a=commitdiff_plain;h=fba399b123276d00d278acaba661bf1a8cf8faa0;p=ceph.git

mds: progress the recover queue immediately after the inode is enqueued

Fixes: https://tracker.ceph.com/issues/50840
Signed-off-by: "Yan, Zheng" <ukernel@gmail.com>
(cherry picked from commit 616ebc3f04620c22439a62f3af35845fa98e508f)
---

diff --git a/src/mds/Locker.cc b/src/mds/Locker.cc
index 2a587659afd..ff98c8d3cf3 100644
--- a/src/mds/Locker.cc
+++ b/src/mds/Locker.cc
@@ -5415,6 +5415,7 @@ void Locker::file_eval(ScatterLock *lock, bool *need_issue)
   }
   else if (in->state_test(CInode::STATE_NEEDSRECOVER)) {
     mds->mdcache->queue_file_recover(in);
+    mds->mdcache->do_file_recover();
   }
 }