!243 [sync] PR-238: mon: fix mds metadata lost in one case

From: @openeuler-sync-bot 
Reviewed-by: @liuqinfei 
Signed-off-by: @liuqinfei

0004-fix-mds-metadata-lost-in-one-case.patch

@@ -0,0 +1,42 @@
+From 56cbf3f0716b556c815487d719abe86021125925 Mon Sep 17 00:00:00 2001
+From: shimin <shimin@kuaishou.com>
+Date: Wed, 10 Apr 2024 09:10:04 +0800
+Subject: [PATCH] mon:fix mds metadata lost in one case
+In most cases, peon's pending_metadata is inconsistent with mon's db.
+When a peon turns into leader, and at the same time a active mds stops,
+the new leader may flush wrong mds metadata into db. So we meed to
+update mds metadata from db at every fsmap change.
+
+This phenomenon can be reproduce like this:
+A Cluster with 3 mon and 3 mds (one active, other two standby), 6 osd.
+step 1. stop two standby mds;
+step 2. restart all mon; (make pending_medata consistent with db)
+step 3. start other two mds
+step 4. stop leader mon
+step 5. run "ceph mds metadata" command to check mds metadata
+step 6. stop active mds
+step 7. run "ceph mds metadata" command to check mds metadata again
+
+In step 7, we would find mds metadata lost.
+
+Fixes: https://tracker.ceph.com/issues/63166
+Signed-off-by: shimin <shimin@kuaishou.com>
+---
+ src/mon/MDSMonitor.cc | 1 +
+ 1 file changed, 1 insertion(+)
+
+diff --git a/src/mon/MDSMonitor.cc b/src/mon/MDSMonitor.cc
+index 4b27d828c..0ac5060f7 100644
+--- a/src/mon/MDSMonitor.cc
++++ b/src/mon/MDSMonitor.cc
+@@ -136,6 +136,7 @@ void MDSMonitor::update_from_paxos(bool *need_bootstrap)
+ 	   << ", my e " << get_fsmap().epoch << dendl;
+   ceph_assert(version > get_fsmap().epoch);
+ 
++  load_metadata(pending_metadata);
+   load_health();
+ 
+   // read and decode
+-- 
+2.27.0
+

ceph.spec

@@ -174,7 +174,7 @@
 #################################################################################
 Name:		ceph
 Version:	18.2.2
-Release:	2
+Release:	3
 %if 0%{?fedora} || 0%{?rhel} || 0%{?openEuler}
 Epoch:		2
 %endif
@@ -195,6 +195,7 @@ Source0:	%{?_remote_tarball_prefix}ceph-18.2.2.tar.gz
 Patch1: 0001-modify-xsimd-source-to-local-and-set-cxx17-for-arrow.patch
 Patch2: 0002-fix-compilation-with-cython3.patch
 Patch3: 0003-fix-CVE-2023-46159.patch
+Patch4: 0004-fix-mds-metadata-lost-in-one-case.patch
 
 %if 0%{?suse_version}
 # _insert_obs_source_lines_here
@@ -2632,6 +2633,9 @@ exit 0
 %{_datadir}/snmp/mibs
 
 %changelog
+* Wed Apr 10 2024 cenhuilin <cenhuilin@kylinos.cn> - 2:18.2.2-3
+- mon: fix mds metadata lost in one case
+
 * Fri Mar 22 2024 lizhipeng <qiuxinyidian@gmail.com> - 2:18.2.2-2
 - fix CVE-2023-46159