Business Objects Data Services CMC console runs real slow

dataservices job design
1

Friends,

Can any one advice casual reason as to why the dataservice CMC console runs very slow.

am accessing eth datservice over web using url:
http://:28080/DataServices/admin.jspwhen I login with admin user and its password the home page open after a long time like 8minutes.

any quick help on this will be appreciated.

-rD

rishi.devanoor :india: (BOB member since 2011-09-21)

2

what is the version of DS ? how many local repositories are registered in Management Console ?

manoj_d (BOB member since 2009-01-02)

3

thanks for reply!

Its BODS 3.2 with SPS1.

I have 8 respositories.
2 repos mapped to 1 job server.

-regards
rD

rishi.devanoor :india: (BOB member since 2011-09-21)

4

Manoj,
Issue got resolved.

I did below:

I deleted all repos, CMC user and Repository entries and then again recreated or updated the Job server entires and re-added users to the CMC console and confugured the job server ./svrcfg.

I bounced the BODS job service - means restarted the job Service and then restarted the tomcat server as well.

it worked :slight_smile:

thanks,
-rD

rishi.devanoor :india: (BOB member since 2011-09-21)

5

I have the same issue on DS 3.2 with 12 repositories. Any other possible solutions out there?

  • E

eepjr24 :us: (BOB member since 2005-09-16)

6

Okie try carrying out below steps:

  1. Login a admin user and navigate to BODS - Business Objects Data Services CCM - Central Management Console - http://:28080/DataServices/launch/logon.do?null
    In CMC under Management > Users : Delete all users and Repositories.

  2. Now (My BODS application server runs on Linux so I use putty to get into application Linux server) navigate to BODS home directory

  3. Navigate to TOMCAT directory and bounce (shutdown and start) web application server /business objects/data services/bin and run

. ./al_env.sh
./svrcfg
now delete all job servers
NOTE: when you delete the BODS sever configuration gui mentions that the BODS configuration file cache will clear all user’s file system repository Cache stored on file system. Say yes to it so that you flush all server cache as well,…

  1. Rest password of all user repositories t Database ;level. DONOT DELETE THE ACTUAL REPSOSITORIES AT DB LEVEL. Just reset the password of all users. YOu should be using the reset password while adding user again in CMC console and also you need same password while adding the repositories to the BODS JOB servers.

  2. Now shutdown the BODS application server.

  3. By now you would have restated the BODS Application server and BODS TOMCAT web Application server by deleting the BODS server entries and CMC mapping to repositories and User ids.

  • By this it means that you have clear the logs as well as the BOE server restarts clears all of the logs as well.

5.Now Start the Tomcat Web Application server and then the BODS Application server. and carry out below:

  • On the BODs application server add repositories. And Create enough Job server entries and add the respective repositories. Ensure you have enabled SNMP. SNMP is responsible for logging the trace and enabling jobs serve processing.
  • On CMC - Login as admin - Add the user and do not forget to generate the password this is very important.And also add the respective repositories.

Now try accessing your CMC console and see if this working faster.

FEW CAUSAL REASON why CMC RUNS SLOW:
-If the admin has not added users ids and generated password files properly then it will occur that the repositories will get wrong user credential and the CMC will get slow.
-Too many logs for CMC to read - Ensure we reboot the BODS and TOMCAT often to clear the logs. I have not heard fo any maintenance job to purge the logs thou.

Please try this and let me know if your CMC runs fast.

-rD

rishi.devanoor :india: (BOB member since 2011-09-21)

7

Thanks Rish. I think we found our culprit, a bit odd though it may be.

One of our remote developers had a local sandbox job server set up and added it to the DEV console to monitor his jobs. Since the console was not always able to reach his job server, it looks like it would time out.

I deleted his user for that box, resynched all the repositories via srvcfg, cleared the DS logs and dump files (we have a sh script to do this that I keep meaning to get around to scheduling), and rebooted Tomcat / JVM. That appears to have solved it.

  • Ernie

eepjr24 :us: (BOB member since 2005-09-16)