MySQL Bugs: #30831: All data nodes are listed in nodegroup 0 while starting

Bug #30831	All data nodes are listed in nodegroup 0 while starting
Submitted:	5 Sep 2007 12:27	Modified:	3 Jun 2008 22:10
Reporter:	Kai Voigt	Email Updates:
Status:	Verified	Impact on me:	None
Category:	MySQL Cluster: Cluster (NDB) storage engine	Severity:	S3 (Non-critical)
Version:	mysql-5.1	OS:	Any
Assigned to:		CPU Architecture:	Any
Tags:	5.1.21

Description:
When datanodes are in the starting phase, right after connection, SHOW in ndb_mgm lists all data nodes being in node group 0, no matter what node topology.

How to repeat:
Setup a cluster, start ndb_mgmd, start all data nodes and immediately check SHOW in ndb_mgm for nodegroup numbers.

Suggested fix:
Remove the nodegroup numer from the output until nodes are fully started.

Thank you for the report.

Please indicate accurate version of MySQL you use.

Discussed with Sveta on IRC; I can't reproduce this, either (tried with 5.0.27 and telco-6.2.4-bk).

Please add a note showing the exact steps you're using and the output you're getting.

# cat cluster.cnf 
# Cluster Example Configuration
# 2 Data Nodes
# 1 Management Node
# 1 MySQLd Node
 
# Management Node
[ndb_mgmd]
Id=1
Hostname=127.0.0.1
DataDir=/usr/local/mysql-cluster/
 
# Data Nodes, Defaults
[ndbd default]
NoOfReplicas=2
DataMemory=20M
IndexMemory=10M
DataDir=/usr/local/mysql-cluster/
 
# Data Node #1
[ndbd]
Id=2
Hostname=127.0.0.1
 
# Data Node #2
[ndbd]
Id=3
Hostname=127.0.0.1

[ndbd]
Id=4
Hostname=127.0.0.1

[ndbd]
Id=5
Hostname=127.0.0.1

# MySQLd Node
[mysqld]
Id=11
Hostname=127.0.0.1

# ndb_mgmd
# ndbd --initial
# ndbd --initial
# ndbd --initial
# ndbd --initial
# ndb_mgm
-- NDB Cluster -- Management Client --

ndb_mgm> show
Cluster Configuration
---------------------
[ndbd(NDB)]     4 node(s)
id=2    @127.0.0.1  (Version: 5.1.15, starting, Nodegroup: 0, Master)
id=3    @127.0.0.1  (Version: 5.1.15, starting, Nodegroup: 0)
id=4    @127.0.0.1  (Version: 5.1.15, starting, Nodegroup: 0)
id=5    @127.0.0.1  (Version: 5.1.15, starting, Nodegroup: 0)

[ndb_mgmd(MGM)] 1 node(s)
id=1   (Version: 5.1.15)

[mysqld(API)]   1 node(s)
id=11 (not connected, accepting connect from 127.0.0.1)

Thank you for the feedback.

But version 5.1.15 is old. Please upgrade to current version 5.1.21 and if you can repeat described behaviour reopen the report.

Same issue on 5.1.21, MacOSX and 5.1.20, Linux

All nodes run the same my.cnf

[mysqld]
ndbcluster

[mysql_cluster]
ndb_connectstring=127.0.0.1
 
[ndb_mgmd]
config-file=/usr/local/mysql-cluster/cluster.cnf

verified (i think it is a duplicate though ...)

Have fixed in 5.1.24

It still happens with 5.1.24

ndb_mgm> SHOW
Cluster Configuration
---------------------
[ndbd(NDB)]	4 node(s)
id=2	@127.0.0.1  (mysql-5.1.24 ndb-6.3.14, starting, Nodegroup: 0, Master)
id=3	@127.0.0.1  (mysql-5.1.24 ndb-6.3.14, starting, Nodegroup: 0)
id=12	@127.0.0.1  (mysql-5.1.24 ndb-6.3.14, starting, Nodegroup: 0)
id=13	@127.0.0.1  (mysql-5.1.24 ndb-6.3.14, starting, Nodegroup: 0)

[ndb_mgmd(MGM)]	1 node(s)
id=1   (mysql-5.1.24 ndb-6.3.14)

[mysqld(API)]	2 node(s)
id=4 (not connected, accepting connect from 127.0.0.1)
id=5 (not connected, accepting connect from 127.0.0.1)

Thank you for the feedback.

Re-verified with mysql-5.1-telco-6.4

To repeat.

Start management node.

I one window:

while (true); do $BASEDIR/bin/ndb_mgm --ndb-mgmd-host=127.0.0.1:35118 -e "show" >>bug30831.txt; done

In another:

start data nodes, wait until they started.

Then in first window interrupt loop and examine log file.

Posted by developer:
 
Maybe fixed by Bug #29520284 : WL#12680: NODE CRASHED DURING ADDNODE OPS WITH ERROR "DBDIH (LINE: 26571) 0X0000