5.1.16 installation

[Johan Jerräng]

Hi!

 

I installed MySQL cluster 5.1.16 a couple of hours ago and the data nodes
just hanged when I started ndbd.

After restarting the hosts I was able to see the following in the
ndb_x_out.log:
2007-03-21 12:32:48 [ndbd] INFO     -- Start initiated (version 5.1.16)

2007-03-21 12:32:48 [ndbd] INFO     -- Ndbd_mem_manager::init(1) min: 20Mb
initial: 20Mb

WOPool::init(61, 9)

RWPool::init(82, 13)

RWPool::init(a2, 18)

RWPool::init(c2, 13)

RWPool::init(122, 18)

RWPool::init(142, 15)

WOPool::init(41, 8)

RWPool::init(e2, 12)

RWPool::init(102, 51)

WOPool::init(21, 6)

 

The log on the management node says the following:

NDB Cluster Management Server. Version 5.1.16 (beta)

Id: 1, Command port: 1186

setEventReportingLevelImpl: failed 2!

setEventReportingLevelImpl(2): failed

setEventReportingLevelImpl: failed 2!

setEventReportingLevelImpl(3): failed

setEventReportingLevelImpl: failed 2!

setEventReportingLevelImpl: failed 2!

setEventReportingLevelImpl: failed 2!

setEventReportingLevelImpl: failed 2!

 

Any ideas??

 

I’m running RHEL4 and here is the config.ini file:

[NDBD DEFAULT]

NoOfReplicas=2

MaxNoOfConcurrentOperations=400000

MaxNoOfAttributes=2000

MaxNoOfOrderedIndexes=2000

MaxNoOfUniqueHashIndexes=1000

DataMemory=512MB

IndexMemory=256MB

LockPagesInMainMemory=1

RedoBuffer=16MB

DiskCheckpointSpeed=32M

TimeBetweenLocalCheckPoints=25

 

[MYSQLD DEFAULT]

BatchByteSize=64K

BatchSize=128

 

[NDB_MGMD DEFAULT]

[TCP DEFAULT]

[SHM DEFAULT]

SigNum=11

 

[NDB_MGMD]

Id=1

HostName=172.19.24.9

DataDir=/storage/mysql-cluster

 

[NDBD]

Id=2

HostName=172.19.24.3

DataDir=/storage/mysql-cluster

BackupDataDir=/storage/mysql-cluster

 

[NDBD]

Id=3

HostName=172.19.24.4

DataDir=/storage/mysql-cluster

BackupDataDir=/storage/mysql-cluster

 

[NDBD]

Id=4

HostName=172.19.24.5

DataDir=/storage/mysql-cluster

BackupDataDir=/storage/mysql-cluster

 

[NDBD]

Id=5

HostName=172.19.24.6

DataDir=/storage/mysql-cluster

BackupDataDir=/storage/mysql-cluster

 

[MYSQLD]

Id=6

HostName=172.19.24.3

 

[MYSQLD]

Id=7

HostName=172.19.24.4

 

[MYSQLD]

Id=8

HostName=172.19.24.5

 

[MYSQLD]

Id=9

HostName=172.19.24.6

 

# For NDB operations from MGM host

[API]

Id=10

HostName=172.19.24.9

 

Best Regards,

 

Johan Jerräng

[Stewart Smith]

On Wed, 2007-03-21 at 14:42 +0100, Johan Jerräng wrote:
>
> setEventReportingLevelImpl: failed 2!

As a bit of a guess... I think this could be due to ndb_mgmd not being
properly connected to data nodes...

what's the output of ndb_mgm -e 'all status' when things have "hung"?
--
Stewart Smith, Software Engineer
MySQL AB, www.mysql.com
Office: [phone] Ext: 6616
VoIP: 6616@...
Mobile: [phone]

Jumpstart your cluster:
http://www.mysql.com/consulting/packaged/cluster.html

[Geeday]

I had a very similar problem to this where the logfile on the mgm node just stopped exactly where yours did (however I can't compare against the other logs you have posted). I had four ndb's and could only get to 2 connect. When i tried to add 4 to the cluster, it would always hang. I kept playing and worked out that one node was messing me up (and often saying it was connected to the cluster, when i don't really think it was!!! I never tested it properly which in hindsight was foolish.). In the end, out of desperation i reinstalled Linux (Fed 6) and Mysql and it worked (reinstalling only Mysql did nothing). I appreciate this is not really any help to you, I'm just sympathising with you and saying how I got round the issue. I am suggesting my problem was down to the OS and not mysql (probably permissions or network/firewall issues). If you have fixed it now, I would be very interested to know what you did!!

Regards

Gordon