on 05-16-2012 12:20 AM
Our management console is very slow. From a search of the forum there are people suggesting changing the tomcat scripts (for v3) to allocate more memory, but I couldn't work out how to do this for version 4.
It seems that this is a fairly common issue. What is normal ? What can I change to improve performance ? Today I got a timeout trying to look at the log for a large job that has been running all night. It took 3 tries to get it to show me the job.
Any advice is appreciated.
Thanks.
DS management Console only perform slow when you try to see the repository job information?
If that's the case, please try the following to see if it can work.
1) Decreasing the log retention period in DS management console.
2) Compact repository through DS designer.
5) Restart web server.
Regards
Helen
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Hi,
1. Decreasing the log retention period.
1) Select Start > Programs > BusinessObjects XI 3.0 > Data Services Management Console > Administrator
2) Expand Management and select Log Retention Period
3) In Log retention period, enter the number of days to retain logs
4) In Job Server Log retention period, enter the number of days to retain logs .
2. Find out the highest number in the OBJECT_KEY column in the AL_LANG table of the PROD repo.
1) Connect to the repository db.
2) Select max(OBJECT_KEY) from OPETLPRC.AL_LANG. If the return number is very big, try compact the repository by following stpes.
1.Log into the repository you desire to compress in Designer.
2.Choose Project > Compact Repository.
3.Choose Local Repository or Central Repository depending on what is required.
4.Click OK.
5.In the box that states "This action may take a long time. Press Ok to continue." press OK.
Note: I think for this steps you can ask your developer for the batch jobs to do through BODS designer.
Regards
Helen
Hi,
In 4.0, we do the configuration through CMC
Log into the Central Management Console (CMC) as a user with administrative rights to the Data Services application.
Go to the “Applications” management area of the CMC.
The “Applications” dialog box appears.
Right-click the Data Services application and select Settings.
The “Settings” dialog box appears.
In the Job Server Log Retention Period box, enter the number of days that you want to retain the following:
Historical batch job error, trace, and monitor logs
Current service provider trace and error logs
Current and historical Access Server logs
The software deletes all log files beyond this period. For example:
If you enter 1, then the software displays the logs for today only. After 12:00 AM, these logs clear and the software begins saving logs for the next day.
If you enter 0, then no logs are maintained.
If you enter -1, then no logs are deleted.
If you enter 1095, then the software deletes logs older than approximately three years.
Click Save.
Changes you make to the log retention period occur as a background clean-up process so they do not interrupt more important message processing. Therefore, logs might not be deleted immediately when you select Save. Changes can take up to an hour to take effect.
Hi,
If it is perform well with the status now, you do not need to do the compact now.
By creating new repository.
Export the repository to a file.
The file type can be either XML or ATL. The latest version of each object is exported.
Choose Start > Programs > SAP BusinessObjects Data Services 4.0 > Data Services Repository Manager .
From the Repository Manager, add the database connection information for the repository.
Click Create.
The software warns that a valid repository already exists.
Click Yes to overwrite the old repository.
The Repository Manager creates a new repository, removing all of the old objects.
Import the previously exported repository.
.
Thanks
Helen
UPDATE: The only problem I have not is viewing the monitor tab on big jobs. The underlying text file for this is over 20mb. Can I up memory parameters or somthing so this works ? The underlying files are not easy to read; the monitor processes them to show the latest state.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
I would not suggest playing around with memory arguments unless you know problem is because of lower memory configuration or insuffucient memory for tomcat
what is the error you see in the management console ? Out Of Memory or timeout ? if it's timeout then stop tomcat, go to %LINK_DIR%\conf folder, open admin.xml and modify the <js-socket-timeout>30000</js-socket-timeout> paramter, first try with a value of 120000, start tomcat and check if the monitor tab shows the log
which page of Management Console is slow ? I think you are using SQL server as DS repo, if yes, are you using Winodws Authentication for DS repo database logon ? how many repos are configured in CMC that are using windows auth ? I am currently fixing an performance issue related to that
to see if it's the same problem for you, then change all the DS repo connection to sql auth ( for testing purpose only) restart tomcat and see if you still see performance issue
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
log retention task no longer runs in DS Webserver (Management Console) from DS 4.0 onwards, large history retention may slow down the list of batch job history page
if you are not using windows authnetication then we need to first identify what is actually causing the perfromance issue
where do you see performance issue, does logon takes long time ?
does the first page when you click on Administrator takes long time ?
if you select a repo and the list of batch job takes long time to display ?
can you quantify it ? some sec, minutes, hours ?
how many local repositories are configured in CMC for DS ? do you see the status of all these repositories in CMC active ?
the first page has 2 frames, the left frames shows the list of repositories and other MC Admin options, this frame may take long time if there are inactive repo connection since while displaying this page connection will be made for each repo so number of repo also affect this
the rgiht frame shows the status of Repos, Access Server, Adapter Job Server etc
if the right frame is taking long time then check if there are any invalid jobservers. access server references in your env deleting those may help.
Is the trace log file big ? the file is retreived by sending a request to job Server there is nothing much that happens in Management Console for that, if the job server is taking long time to retreive the file it may cause this issue, can you check the JobServer log direct and also sub directories under job server if they have lot of files ? do you see very old files in those folders ?
these logs should delete by the LogDeleter task that run in EIMAdaptiveProcessingServer, instead of delete the files manually, it would be better you can find out why the logDeleter is not cleaning up these files
Check the EIMAdaptiveProcessingServer's log file in default BOE Logging directory if you see errors related to log deleter.
did you move your job server from one machine to other or change the JS port ?
to do a complete cleanup maually other than deleting the files manually using your own script you will also have to make sure that the history in repo is cleaned up else you will see the same isuse n batch job history page
you can check the tables that are required to cleaned up in the following post
http://www.forumtopics.com/busobj/viewtopic.php?t=145031
also, check how many rows are there in AL_HISTORY table ?
User | Count |
---|---|
80 | |
9 | |
9 | |
7 | |
7 | |
7 | |
6 | |
5 | |
5 | |
4 |
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.