Network Management

Hello,

I have main disk on airwave full over 97% and I want to free some space.

How can I free some space?

Somebody could help me?

 

Thanks.

Try this: https://arubanetworkskb.secure.force.com/pkb/articles/FAQ/Freeing-up-space-when-the-hard-drive-is-full

 

Also, you can reduce your data retention numbers under AMP Setup> General.

---

@felixgr wrote:

Hello,

I have main disk on airwave full over 97% and I want to free some space.

How can I free some space?

Somebody could help me?

 

Thanks.

---

I've had to deal with this, and Colin is correct, reduce your Historical Data Retention numbers.

 

The one that keeps the most data is Client Data Retention Interval and Client Association and VPN Session History.

It also depends how many APs, Clients your AMP is seeing and monitoring daily.

 

Example: I have an AMP with 1.5Tb of HD Space monitoring 2200 RAPs and close to 30,000 users per day and those 2 values above went from the default to 180 to 14, we run reports in order to make sure we save the data.

 

 

Once you make a change to retention numbers, when does it take effect?  Our install is running up to the disk size and we had the large default of 425 days of client data retention, so we greatly reduced that and expected some sort of "purge" to happen, but we remain at the same disk usage.  

As soon as you reduce the number of days, give it a day or so and any RRD file that is older than the retention date will be aged out.

Client Data Retention is a big one and if it was set to 425 days and you reduce to say 180 days, you will see a significant difference.

Just wanted to make sure that this behavior was clear:

 

If the bulk of the data being used is in /var/airwave/rrd, then you'll need to make sure to reduce Client Data Retention Interval, Client Association history, and Inactive Client data.  Just reducing Client Data Retention Interval will do nothing on its own for existing clients.  The clients need to be aged out, and then newly created RRDs adhere to the new sizing.  The system will go through the clean up scripts during nightly maintenance window.  The first large sweep of clean up may inflate the time it takes for nightly maintenance to run.  After that, you can return to regular Client Association and Inactive Client retention values, but maintain the Client Data Retention Interval low to keep the RRDs small.

If the majority of your data in the postgres directory, and cleaning the RRDs does not help as much as you had hoped, we found that doing an upgrade can sometimes cause this to self-clean.  Also, let it sit a bit after the upgrade before rebooting it, even if there are kernel pacthes.