cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
885
Views
0
Helpful
4
Replies

PI 2.1 low disk space warning, 67% used

patoberli
VIP Alumni
VIP Alumni

PI 2.1 started to give me a warning that it's low on disk space on partition optvol. This partition is currently used by 67% and has a total size of nearly 200 GB. I did a ncs cleanup, but only said yes to the first option. I also deleted the files out of defaultRepo.

Any ideas why the threshold is set this low, or if there is an other issue?

Thanks

Patrick

4 Replies 4

AFROJ AHMAD
Cisco Employee
Cisco Employee

Hi Patrick,

Prime Infrastructure triggers this alert if the usage of the optvol  partitiongoes above 50%. In this case, the usage is 67% hence a genuine  alert from the code perspective.

===

My suggestion would be to start with the ncs cleanup, then take another look at your how
much data you're keeping and for how long. 

The key areas are under Administration > System Settings > and the pages Alarms & Events, Client and Data Retention. 

check if the values are appropriate for how you use NCS. 

For example, on Alarms and Events, is the box checked to delete old alarms after so many days? 

If not, it probably should be.  If it is, look at the values and consider how likely is it that someone on staff is really going to go back that far to address an alarm.  If not, don't hold the records around--delete
them.

 

Hope it will help

 

Thanks-

Afroz

***Ratings Encourages Contributors ****

 

 

Thanks- Afroz [Do rate the useful post] ****Ratings Encourages Contributors ****

Thanks for your answer. The box checked to delete old alarms after so many days is checked. Currently it's set to 90 days. 

Main information that I require is finding a client based on an IP address. This could be even after a month. 

In any case, I lowered a few parameters and will have an eye on it.

I reduced the Data Retention now to lower values two days ago, but it actually uses even more space now.

It seems as if there is an issue with PI 2.1 and sidk space usage.

Can you check if your server also uses more and more space?

I "fixed" it now by adding another 200 GB drive to the virtual machine and rebooting it. It did all the repartitioning and assigning the diskspace by itself and the usage is now under the water mark. Please ensure that you have the option for automatic tuning on reboot active (see PI manual for more details).