Bug #21237 ndbd segfaults when I do an sql on a ndbcluster table

• Submitted: 23 Jul 2006 9:56
• Modified: 24 Aug 2006 15:53
• Reporter: John-Paul Drawneek
• Status: No Feedback
• Category: MySQL Cluster: Cluster (NDB) storage engine
• Severity: S1 (Critical)
• Version: 5.0.22
• OS: Solaris (Solaris 10 u2)
• CPU Architecture: Any

[23 Jul 2006 9:56] John-Paul Drawneek

Description:
Hi, not sure whats up but I cannot do any sql with the clustered db.
I am trying to run egroupware www.egroupware.org on a cluster of 4 sparc solaris 10u2.

ndb_3_error.log

Time: Friday 21 July 2006 - 12:51:41
Status: Temporary error, restart node
Message: Error OS signal received (Internal error, programming error or missing error message, please report a bug)
Error: 6000
Error data: Signal 11 received; Segmentation Fault
Error object: main.cpp
Program: /opt/mysql/mysql/bin/ndbd
Pid: 894
Trace: /MySQLdata/ndb_3_trace.log.6
Version: Version 5.0.22
***EOM***

config.ini

[NDBD DEFAULT]
NoOfReplicas=2
DataMemory=150M
IndexMemory=50M
MaxNoOfOrderedIndexes=3000
MaxNoOfUniqueHashIndexes=1000
MaxNoOfAttributes=3000
TransactionBufferMemory=10M

[TCP DEFAULT]
#portnumber=2202

[NDB_MGMD]
hostname=192.168.0.104
datadir=/var/lib/mysql-cluster

[NDBD]
hostname=192.168.0.21
datadir=/MySQLdata

[NDBD]
hostname=192.168.0.20
datadir=/MySQLdata

[MYSQLD]
hostname=192.168.0.102

How to repeat:
Run sql on the cluster

[23 Jul 2006 10:03] John-Paul Drawneek

ndb_3_trace.log.6

Attachment: ndb_3_trace.log.6.zip (application/zip, text), 47.25 KiB.

[23 Jul 2006 10:17] Valeriy Kravchuk

Thank you for a problem report. I had changed category to a more appropriate one. Please, describe your hardware in more details (how much RAM do you have?). Send the content of MySQL server's error log also.

[23 Jul 2006 10:28] John-Paul Drawneek

The boxes have 512mb at the moment.
ndbds are netra t1 105
mgm and sql are zones on a E250 with 2gb ram

[23 Jul 2006 10:31] John-Paul Drawneek

mgm node logs - i think its log 3, but i been trying lots of times so could be in all of them

Attachment: mgmlog.zip (application/zip, text), 85.46 KiB.

[23 Jul 2006 10:37] John-Paul Drawneek

log file from sql node, hopefully

Attachment: sql.zip (application/zip, text), 791 bytes.

[24 Jul 2006 15:53] Hartmut Holzgraefe

From what i can see in your SQL logs your installation is broken already,
the cluster node failures may just be followup errors, can you verify that
you get the same cluster problems with a working MySQL server, too?

  060720 12:14:58 [ERROR] Fatal error: Can't open and lock privilege tables: Failed to open 'host', error while unpacking from engine

  /usr/local/mysql/bin/mysqld: File 'dir=/var/mysql/data/' not found (Errcode: 2)
 
  060720 12:22:48 [ERROR] Could not use dir=/var/mysql/data/ for logging (error 2). Turning logging off for the whole duration of the MySQL server process. To turn it on again: fix the cause, shutdown the MySQL server and restart it.

[24 Aug 2006 23:00] Bugs System

No feedback was provided for this bug for over a month, so it is
being suspended automatically. If you are able to provide the
information that was originally requested, please do so and change
the status of the bug back to "Open".