Hey all,

I installed DB2 ESE V8.2 with DPF on two machines running Linux AS 3.

I had no problem with the installation. And the comunication between the two servers is also ok.

I updated the db2nodes.cfg, also update the /etc/services on the secondary machine, as well as the /etc/hosts on both machines, adding their hostnames for comunication, and created the .rhosts file.

The instance starts up well.

The problem begins when I try to create the database. Either the sample database, or my own, I get, always, the same error:

SQL1229N The current transaction has been rolled back because of a system error. SQLSTATE=40504

I updated the DBM configuration directing the db2diag.log to a directory that´s not shared to make it easier to analyse the log files.

The following text comes from the db2diag.log after the execution of the db2sampl command. I´m creating the database on the /tmp/teste, which is not a shared directory - trying to create a DB on a shared directory returns the same error:

------------------------
-- Log File from the primary server:
2005-02-25-11.10.12.004799-180 I3332G391 LEVEL: Warning
PID : 5172 TID : 3007203840 PROC : db2pdbc
INSTANCE: db2inst1 NODE : 000
FUNCTION: DB2 UDB, base sys utilities, sqleCheckForDynDBMUpdate, probe:50
MESSAGE : Attempted to recover dynamic config updates =
DATA #1 : Hexdump, 4 bytes
0xBFFF6D00 : 0000 0000 ....

2005-02-25-11.10.12.005377-180 I3724G382 LEVEL: Error
PID : 5203 TID : 3007203840 PROC : db2agent (instance)
INSTANCE: db2inst1 NODE : 000
APPHDL : 0-53 APPID: *N0.db2inst1.050225140923
FUNCTION: DB2 UDB, buffer dist serv, sqlkdReceiveReply, probe:10
RETCODE : ZRC=0x81590016=-2124873706=SQLKF_NODE_FAILED "Node Recovery"

2005-02-25-11.10.12.011223-180 I4107G360 LEVEL: Warning
PID : 5203 TID : 3007203840 PROC : db2agent (instance)
INSTANCE: db2inst1 NODE : 000
APPHDL : 0-53 APPID: *N0.db2inst1.050225140923
FUNCTION: DB2 UDB, base sys utilities, sqlePdbRemap, probe:100
MESSAGE : ZRC 81580016 remapped to SQLCODE -1229

2005-02-25-11.10.12.016305-180 I4468G404 LEVEL: Warning
PID : 5204 TID : 3007203840 PROC : db2agntp
INSTANCE: db2inst1 NODE : 000
APPHDL : 0-53 APPID: *N0.db2inst1.050225140923
FUNCTION: DB2 UDB, oper system services, sqloopenp, probe:80
MESSAGE : errno:
DATA #1 : Hexdump, 4 bytes
0xBFFF0CA0 : 1500 0000 ....

2005-02-25-11.10.12.016660-180 I4873G337 LEVEL: Warning
PID : 5204 TID : 3007203840 PROC : db2agntp
INSTANCE: db2inst1 NODE : 000
APPHDL : 0-53 APPID: *N0.db2inst1.050225140923
MESSAGE : Path/Filename
DATA #1 : String, 48 bytes
/tmp/teste/db2inst1/NODE0000/SQL00001/SQLT0000.0

2005-02-25-11.10.12.025995-180 I5211G404 LEVEL: Warning
PID : 5204 TID : 3007203840 PROC : db2agntp
INSTANCE: db2inst1 NODE : 000
APPHDL : 0-53 APPID: *N0.db2inst1.050225140923
FUNCTION: DB2 UDB, oper system services, sqloopenp, probe:80
MESSAGE : errno:
DATA #1 : Hexdump, 4 bytes
0xBFFF0CA0 : 1500 0000 ....

2005-02-25-11.10.12.026192-180 I5616G337 LEVEL: Warning
PID : 5204 TID : 3007203840 PROC : db2agntp
INSTANCE: db2inst1 NODE : 000
APPHDL : 0-53 APPID: *N0.db2inst1.050225140923
MESSAGE : Path/Filename
DATA #1 : String, 48 bytes
/tmp/teste/db2inst1/NODE0000/SQL00001/SQLT0001.0

2005-02-25-11.10.12.026598-180 I5954G404 LEVEL: Warning
PID : 5204 TID : 3007203840 PROC : db2agntp
INSTANCE: db2inst1 NODE : 000
APPHDL : 0-53 APPID: *N0.db2inst1.050225140923
FUNCTION: DB2 UDB, oper system services, sqloopenp, probe:80
MESSAGE : errno:
DATA #1 : Hexdump, 4 bytes
0xBFFF0CA0 : 1500 0000 ....

2005-02-25-11.10.12.026768-180 I6359G337 LEVEL: Warning
PID : 5204 TID : 3007203840 PROC : db2agntp
INSTANCE: db2inst1 NODE : 000
APPHDL : 0-53 APPID: *N0.db2inst1.050225140923
MESSAGE : Path/Filename
DATA #1 : String, 48 bytes
/tmp/teste/db2inst1/NODE0000/SQL00001/SQLT0002.0

2005-02-25-11.11.02.013726-180 I6697G391 LEVEL: Warning
PID : 5172 TID : 3007203840 PROC : db2pdbc
INSTANCE: db2inst1 NODE : 000
FUNCTION: DB2 UDB, base sys utilities, sqleCheckForDynDBMUpdate, probe:50
MESSAGE : Attempted to recover dynamic config updates =
DATA #1 : Hexdump, 4 bytes
0xBFFF6D00 : 0000 0000 ....

2005-02-25-11.11.02.013966-180 I7089G382 LEVEL: Error
PID : 5203 TID : 3007203840 PROC : db2agent (instance)
INSTANCE: db2inst1 NODE : 000
APPHDL : 0-53 APPID: *N0.db2inst1.050225140923
FUNCTION: DB2 UDB, buffer dist serv, sqlkdReceiveReply, probe:10
RETCODE : ZRC=0x81590016=-2124873706=SQLKF_NODE_FAILED "Node Recovery"

-- Log File from the secondary server:
2005-02-25-11.15.09.012523-180 I3743G391 LEVEL: Warning
PID : 20328 TID : 3007212032 PROC : db2pdbc
INSTANCE: db2inst1 NODE : 001
FUNCTION: DB2 UDB, base sys utilities, sqleCheckForDynDBMUpdate, probe:50
MESSAGE : Attempted to recover dynamic config updates =
DATA #1 : Hexdump, 4 bytes
0xBFFF1000 : 0000 0000

------------------------

I´ve tried contacting the IBM support, but they seem as lost as I am. While they´re trying to find what´s wrong, if you have any idea that might help to solve this problem...

Thanks in advance!!

Fernando Ohana

When starting db2 do you see twice a startup message complete for each node ? try to create a db on non shared disk / fs /path. but this path should exist on both nodes /tmp.....

try to create a manual database not dbsample to see the real messages and empty the diag log before start
Before creating the db : try db2_all get dbm cfg to see if both reply
Best Regards, Guy Przytula

Thanks for the quick response, but as I said before, apparently, everything is working fine.

When starting the instance, I see both partitions starting.

When I execute the db2_all "db2 get dbm cfg" I get a response from both servers. And I assume the comunication between the two machines is OK, because once I configured the db2diag.log file to be create on a non-shared directory, each partition will have it´s own file, and they´re both being updated as expected.

Now, about your suggestion to execute the create db manually to a non-shared directory, it returned exactly the same error.

I issued the following statement: db2 create db teste on /tmp/teste

The directory /tmp/teste was created, previously, on each server manually and its not shared.

The error printed on the DB2DIAG.LOG is pretty much the same I´ve posted earlier.

I know I got too few information to give you an image of what´s happening. But that´s just it.. That´s all I got.

TIA
Fernando Ohana

change your diaglevel to 4.
check the kernel parameters.
Also check the fmc dbm parameters.
Check the number of processes that db2inst can have at the OS level.
Also run the command with

db2 -a , you will identify which node is failing.

Try to run the command from a different node.
The new node will become your catalog node.


regards

Mujeeb