How to clean prime infrastructure data base

[gaurishsadadekar19931]

Hi,

I had a brief communication with Prime Cisco tac regarding disk cleanup.They deleted some unwanted temp files and the percentage came to 57 from 77.

After a day, again It went to 79.Can you please help wherein I can know which files are increasing?

Also, If possible a method to clean prime database wherein it should not crash since its already in production.If I get a help from your end,I can maintain a weekly checklist.

Regards,

Gaurish Sadadekar.

[Francesco Molino]

Hi

To clean the CPI database, you'll need to use ncs cleanup. However, it is better to use it when ncs is stopped.

You have also to be careful of your data retention Administration > Settings > System Settings > Data Retention and cleanup of alarms/alerts Administration > Settings > System Settings > Alarms and Events > Alarms and Events

You also have some data retention for clients/users at Data Retention under “Clients & User settings

To not face blocking issues, you can decrease threshold for Disk utilization under system event configuration

Hope this helps.

Thanks

PS: Please don't forget to rate and mark as correct answer if this answered your question

[gaurishsadadekar19931]

Hello Partner,

Thanks for the update.

By doing the NCS Cleanup procedure.I agree that the disk utilization decreases.But the main concern is after few days it again went to the same value which was earlier.

Any other procedure to mitigate this issue.

Thanks & Regards,

Gaurish Sadadekar

[Francesco Molino]

Hi Did you tweak all thresholds and logs you want to have?

there is no more procedures. Maybe you're logging to much information or you're facing a bug. If after tweaking, you have same issue, you might have to open a TAC case. 

Thanks. 

PS: Please don't forget to rate and mark as correct answer if this answered your question 

[Nickolus Looper]

From folks' various experiences running 'ncs cleanup', how long does this usually take?

On my system, it's currently been working on the step 'starting database cleanup' for >1.5 hours (current time is 23:06 EDT).

{Wed May 30 21:55:41 EDT 2018} Cleaning database
{Wed May 30 21:55:44 EDT 2018} Stopping database
{Wed May 30 21:55:47 EDT 2018} Starting database
{Wed May 30 21:56:26 EDT 2018} Starting database clean

Thanks

Nick

[tyoung008]

Did it ever get to "Completed database clean" on its own?

[Nickolus Looper]

I need to make my alerts more prominent in these forums, as I'm not noticing the replies - sorry.

I have run this one or two more times since the post, and yes, it always end up taking a lot longer than the documentation shows - like several hours.  This is because I work with very large PI implementations - much larger than the small lab instances typically used to create the documentation.

[iportuga]

Firstly you must identify which is causing filesystem usage growth. Using a couple of scripts we can do that. In case of database, tablespace size analysis must be done in order to detect which database objects are causing the issue.

Please check this TechNotes document: Prime Infrastructure - Troubleshooting Low Disk Space Issues

If needed, contact the Cisco TAC for further assistance.

Regards,
Igor