ndb - bug#23210

  Fix race-condition between COPY_GCIREQ (GCP) and lcpSetActiveStatusEnd
  Solution is _not_ to copy sysfileData from COPY_GCIREQ from "self"


ndb/src/kernel/blocks/ERROR_codes.txt:
  Add error insert for dealying of copy sysfileData
ndb/src/kernel/blocks/dbdih/DbdihMain.cpp:
  1) Add error insert for delaying of sysfileData
  2) Change to that master is _not_ copying sysfileData from COPY_GCIREQ
     as it might be updating it while COPY_GCIREQ is "in the fly"

2 files changed, 35 insertions, 7 deletions

diff --git a/ndb/src/kernel/blocks/ERROR_codes.txt b/ndb/src/kernel/blocks/ERROR_codes.txt
index d6dfcfe8587..874c128a56d 100644
--- a/ndb/src/kernel/blocks/ERROR_codes.txt
+++ b/ndb/src/kernel/blocks/ERROR_codes.txt
@@ -5,7 +5,7 @@ Next DBACC 3002
 Next DBTUP 4014
 Next DBLQH 5043
 Next DBDICT 6006
-Next DBDIH 7174
+Next DBDIH 7178
 Next DBTC 8038
 Next CMVMI 9000
 Next BACKUP 10022
@@ -66,6 +66,8 @@ Delay GCP_SAVEREQ by 10 secs
 7030: Delay in GCP_PREPARE until node has completed a node failure
 7031: Delay in GCP_PREPARE and die 3s later
 
+7177: Delay copying of sysfileData in execCOPY_GCIREQ
+
 ERROR CODES FOR TESTING NODE FAILURE, LOCAL CHECKPOINT HANDLING:
 -----------------------------------------------------------------
 
diff --git a/ndb/src/kernel/blocks/dbdih/DbdihMain.cpp b/ndb/src/kernel/blocks/dbdih/DbdihMain.cpp
index 7ae7db967b2..fcb9b0fbb60 100644
--- a/ndb/src/kernel/blocks/dbdih/DbdihMain.cpp
+++ b/ndb/src/kernel/blocks/dbdih/DbdihMain.cpp
@@ -626,22 +626,48 @@ void Dbdih::execCOPY_GCIREQ(Signal* signal)
   ndbrequire(c_copyGCISlave.m_copyReason  == CopyGCIReq::IDLE);
   ndbrequire(c_copyGCISlave.m_expectedNextWord == tstart);
   ndbrequire(reason != CopyGCIReq::IDLE);
-  
+  bool isdone = (tstart + CopyGCIReq::DATA_SIZE) >= Sysfile::SYSFILE_SIZE32;
+
+  if (ERROR_INSERTED(7177))
+  {
+    jam();
+
+    if (signal->getLength() == 3)
+    {
+      jam();
+      goto done;
+    }
+  }
+
   arrGuard(tstart + CopyGCIReq::DATA_SIZE, sizeof(sysfileData)/4);
   for(Uint32 i = 0; i<CopyGCIReq::DATA_SIZE; i++)
     cdata[tstart+i] = copyGCI->data[i];
   
-  if ((tstart + CopyGCIReq::DATA_SIZE) >= Sysfile::SYSFILE_SIZE32) {
+  if (ERROR_INSERTED(7177) && isMaster() && isdone)
+  {
+    sendSignalWithDelay(reference(), GSN_COPY_GCIREQ, signal, 1000, 3);
+    return;
+  }
+  
+done:  
+  if (isdone)
+  {
     jam();
     c_copyGCISlave.m_expectedNextWord = 0;
-  } else {
+  } 
+  else 
+  {
     jam();
     c_copyGCISlave.m_expectedNextWord += CopyGCIReq::DATA_SIZE;
     return;
-  }//if
-  
-  memcpy(sysfileData, cdata, sizeof(sysfileData));
+  }
   
+  if (cmasterdihref != reference())
+  {
+    jam();
+    memcpy(sysfileData, cdata, sizeof(sysfileData));
+  }
+
   c_copyGCISlave.m_copyReason = reason;
   c_copyGCISlave.m_senderRef  = signal->senderBlockRef();
   c_copyGCISlave.m_senderData = copyGCI->anyData;