DB13 does not work with cluster in different node

[Former Member]

Dear SAP,

We already install BW 3.5 with oracle database on Linux Cluster node.

There are 2 node for Cluster.

host-A ==> Node A will be startup with CI.

host-B ==> Node B will be startup with DB.

We found some problem when we startup SAP with separate node. We can

not do for any operation under DB13. Example : DB Check. We found for

this error.

=============================================

Job started

Step 001 started (program RSDBAJOB, variant &0000000001250, user ID

SARATGON)

No application server on database host - rsh will be used

Execute logical command BRCONNECT On host bwporagrp

Parameters: -u / -c -f check

BR0801I BRCONNECT 7.00 (16)

BR0252E Function fopen() failed for '/oracle/BWP/sapcheck/cebhpirj.chk'

at location main-9

BR0253E errno 13: Permission denied

BR0121E Processing of log file /oracle/BWP/sapcheck/cebhpirj.chk failed

BR0806I End of BRCONNECT processing: cebhpirj.chk2009-08-23 14.33.31

BR0280I BRCONNECT time stamp: 2009-08-23 14.33.31

BR0804I BRCONNECT terminated with errors

External program terminated with exit code 3

BRCONNECT returned error status E

Job finished

=============================================

For more information. In Unix cluster config. We enable only command

SSH to access. So we already set parameter in SAP Profile follow this.

gw/remsh = /usr/bin/ssh

When we test with command under unix. It's work.

===================================================

host-A:bwpadm > ssh host-B ls -al /oracle/BWP/sapcheck/

-

---

WARNING: computer system,

which may be accessed and used only for authorized business by

authorized personnel. Unauthorized access or use of this computer

system may subject violators to criminal, civil and/or administrative action.

-

---

total 12

drwxr-xr-x 3 orabwp dba 8192 2009-08-21 04:25 .

drwxr-xr-x 18 orabwp dba 4096 2009-08-21 12:48 ..

drwxr-xr-x 2 root root 96 2009-08-16 07:36 lost+foun

===================================================

For temporary solution. Now we startup SAP ( CI and DB ) with one node

for golive. Could you analyze for us in case of this config under

Cluster with different node ?

Thanks & regards,

Somkeit

[anindya_bose · Accepted solution]

Som

Please check whether you have TCP/IP connection SAPXPG_DBDEST_<DBHOSTNAME> for both the node.

DB13 access database through this RFC, if you do not have this RFC or the RFC is failing from one node, then you wont be able to do anything from DB13 from that node.

[Former Member]

Dear All,

Now this issue was solved. The problem come from Environment of SAP Execute file was wrong.

After we set it correctly. Now it's fix.

Thank you very much for all information.

regards,

Somkeit

[Former Member]

Dear JPSelter,

Thanks for your information.

Does anybody know for this error ? How can we solved for this ?

Regards,

Somkeit

[Former Member]

I had a question.

When I schedule any job using db13 the program used is rsdbajob. For eg DBA:REORGCK_ALL_____@170000/2000 and DBA:RUNSTAT_TAB_____@230000/2000 both use the same program rsdbajob. How does the scheduler distinguish amongst the two ?

I thought may be using the variant but it also differs and is not constant for a job. i.e when i schedule the same job twice say DBA:REORGCK_ALL_____@170000/2000 the jobs have different varients in the form &0000000000023 and &0000000000024.

So my question is

1) how does the scheduler distinguish between reorg and runstats ?

2) what is the use of the variant ?

thank you.

[Former Member]

Dear JPSelter,

We already use with the lastest Kernel for our system. But we still found for the problem.

What is error message that you found from last time ?

Is it same as our message ?

regards,

Somkeit

[Former Member]

Hello,

unfortunately all logs from that time have been dropped by the system, that was last June. So I cannot see the exact error message when we had the problem. DB13 was running perfectly on node1, but we were not able to start DB13 from node2. I was looking for an email where I described the problem to our system partner but seems as if I was talking to them over phone...

[Former Member]

Dear Bose,

Thanks for your quick reply. We have been see from that note.

Note 446172 - SXPG_COMMAND_EXECUTE (program_start_error) in DB13

Now I think about this note.

Note 188772 - External program does not work

Is it relate to environment of sidadm user ? Cause last time when I try to call from os level.

Example :

ssh hostname-B brtools ==> It doesn't work.

ssh hostname-B /sapmnt/BWP/exe/brtools ==> It work.

So I adjust for environment of sidadm user. Then I can call for command.

ssh hostname-B brtools

It look like SAP try to call program sapxpg from node B. But it can not call. Cause program not found.

Is it correct for analyze ?

Regards,

Somkeit

[Former Member]

We had a similar problem with DB13 on our cluster system (DB13 was working on node1 but not on node2). We upgraded the kernel to a higher version, the problem disappeared after that.

[Former Member]

Dear Bose,

We still fail for load balance setting. Could you analyze from the lastest error ?

Job started

Step 001 started (program RSDBAJOB, variant &0000000001296, user ID KAPILSOO)

No application server on database host - rsh will be used

Execute logical command BRCONNECT On host bwporagrp

Parameters: -u / -c -f check

SXPG_COMMAND_EXECUTE failed for BRCONNECT - Reason: program_start_error: For More Information, See SYS

Job cancelled after system exception ERROR_MESSAGE

Regards,

Somkeit

[anindya_bose]

Som.

> SXPG_COMMAND_EXECUTE

Is the RFC SAPXPG_DBDEST_<> working from both the nodes? Check from SM59.

Also have a look at note https://service.sap.com/sap/support/notes/446172.

This will provide you information about how DB13 works.

[Former Member]

Dear Bose,

Thanks for your reply. I have more question.

1. How about this error message ?

No application server on database host - rsh will be used

We're using with /usr/bin/ssh shell for remote. We don't enable rsh.

How to solved for this error too ?

2. Refer from your information RFC Desc "SAPXPG_xxx_xxx".

Error message

BR0252E Function fopen() failed for '/oracle/BWP/sapcheck/cebhpirj.chk'

at location main-9

BR0253E errno 13: Permission denied

Question : Do we need to adjust DB13 table parameter too ?

I mean to use with ZBRBACKUP, ZBRCONNECT , etc

3. Now we starting SAP CI and DB in one node.

Is it possible to simulate test brconnect from node A to node B same as DB13 ?

We can not shutdown system now.

Regards,

Somkeit

[anindya_bose]

Som

Please check the note below..

https://service.sap.com/sap/support/notes/34219

Best Regards

Anindya