bk commit into 5.1 tree (tomas:1.2022) BUG#13987

[tomas]

Below is the list of changes that have just been committed into a local
5.1 repository of tomas. When tomas does a push these changes will
be propagated to the main repository and, within 24 hours after the
push, to the public repository.
For information on how to access the public repository
see http://dev.mysql.com/doc/mysql/en/installing-source-tree.html

ChangeSet
  1.2022 06/05/22 15:58:26 tomas@stripped +2 -0
    Bug #18591 Repeated "show" and "all status" calls to ndb_mgm causes mgmd to crash
    - not having lock can cause "block number list" to become corrupt => crash
    - also may result in 2 threads receiving same block no => possble cause of
    Bug #13987 Cluster: Loss of data nodes can cause high CPU usage from ndb_mgmd

  storage/ndb/src/ndbapi/SignalSender.cpp
    1.7 06/05/22 15:58:17 tomas@stripped +2 -0
      Bug #18591 Repeated "show" and "all status" calls to ndb_mgm causes mgmd to crash
      - not having lock can cause "block number list" to become corrupt => crash
      - also may result in 2 threads receiving same block no => possble cause of
      Bug #13987 Cluster: Loss of data nodes can cause high CPU usage from ndb_mgmd

  storage/ndb/src/mgmsrv/MgmtSrvr.cpp
    1.91 06/05/22 15:58:17 tomas@stripped +1 -0
      Bug #18591 Repeated "show" and "all status" calls to ndb_mgm causes mgmd to crash
      - not having lock can cause "block number list" to become corrupt => crash
      - also may result in 2 threads receiving same block no => possble cause of
      Bug #13987 Cluster: Loss of data nodes can cause high CPU usage from ndb_mgmd

# This is a BitKeeper patch.  What follows are the unified diffs for the
# set of deltas contained in the patch.  The rest of the patch, the part
# that BitKeeper cares about, is below these diffs.
# User:	tomas
# Host:	poseidon.ndb.mysql.com
# Root:	/home/tomas/wl2325-alcatel

--- 1.90/storage/ndb/src/mgmsrv/MgmtSrvr.cpp	2006-05-08 14:55:15 +02:00
+++ 1.91/storage/ndb/src/mgmsrv/MgmtSrvr.cpp	2006-05-22 15:58:17 +02:00
@@ -2375,6 +2375,7 @@
 MgmtSrvr::abortBackup(Uint32 backupId)
 {
   SignalSender ss(theFacade);
+  ss.lock(); // lock will be released on exit
 
   bool next;
   NodeId nodeId = 0;

--- 1.6/storage/ndb/src/ndbapi/SignalSender.cpp	2005-09-14 09:39:46 +02:00
+++ 1.7/storage/ndb/src/ndbapi/SignalSender.cpp	2006-05-22 15:58:17 +02:00
@@ -75,7 +75,9 @@
 {
   m_cond = NdbCondition_Create();
   theFacade = facade;
+  lock();
   m_blockNo = theFacade->open(this, execSignal, execNodeStatus);
+  unlock();
   assert(m_blockNo > 0);
 }