Results 1 to 6 of 6
  1. #1
    Join Date
    May 2002
    Posts
    5

    Unanswered: DB2 EEE on RedHat 7.0

    hi,
    I have troubles using DB2 EEE on linux. I installed the EEE on two linux machine. The operating system is Redhat 7.0. I followed the instruction to create nfs home and successfully mounted from one machine to the other. But when I issued "create database test", I always get the following error message:
    05-08-2002 13:32:12 2 0 SQL1063N DB2START processing was successful.
    05-08-2002 13:32:14 1 0 SQL1063N DB2START processing was successful.
    05-08-2002 13:32:15 0 0 SQL1063N DB2START processing was successful.
    SQL1063N DB2START processing was successful.
    [db2inst1@speed db2dump]$ db2 "create database test2"
    SQL1229N The current transaction has been rolled back because of a system
    error. SQLSTATE=40504

    I'm very appreciate for any suggestions to fix this problems.

  2. #2
    Join Date
    Oct 2001
    Posts
    68
    You need to start looking at db2diag.log for the cause of this problem. I don't know why you are playing with "nfs" (Is that a typo.?) Everyone if not all knows that NFS is a serious performance killer in a high-volume environment. Maybe you're just experimenting with fault-tolerance and disaster recovery. Get the db2diag.log. To find out where it is on your system read this post: http://dbforums.com/t319674.html
    for some pointers.

    WP

  3. #3
    Join Date
    May 2002
    Posts
    5
    Hi WP,
    As described in DB2 EEE quick guide, we can create multiple database partitions on a single machine or more than one machines.

    Two days ago, I first create two partition nodes on one machine and issue "create database test" and then partition the test database onto these two nodes. Everything is fine and I successfully partion the test database.
    Then I create one node on the other machine and create another test database to see how this work. But as described in DB2 EEE quick guide in order to make it work, I need to create a directory and do the nfs mount so that each partition nodes see this directory. I followed the quick guide instruction and everything is fine before I issued the create database command. I also looked at the db2diag.log and always get the following messages.

    2002-05-08-14.19.01.010859 Instance:db2inst1 Node:000
    PID:1597(db2fcmdm) Appid:none
    oper_system_services sqloPdQuerySocketErrorStat Probe:15
    errno=111

    Do any one have any idea how to make multiple database partitions on two physical machines work?

    Thanks for any suggestions

  4. #4
    Join Date
    Oct 2001
    Posts
    68
    Ellen, as you've probably determined by now, the db2fcmdm daemon is the FCM (Fast Communications Manager). As shown in your db2diag.log entries, this process is failing.

    On my AIX box in /usr/include/errno.h (which contains operating system error codes), a 111 is:

    #define ESOFT 111 /* I/O completed, but needs relocation */

    I have never seen this one before but will dig into it some more. It must be something to do with the nfs cross-mounting.

    WP

  5. #5
    Join Date
    May 2002
    Posts
    5
    Hi.
    I'm appreciate for those who giving me suggestions, I resolved the problem this afternoon. The problem is that I installed the two db2 instances under different directories on the two machines. I resolved the problems by reinstalled the db2 instance under same directory names on two machines.

    Thanks

    Ellen

  6. #6
    Join Date
    Oct 2001
    Posts
    68
    Well done.

    WP

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •