Netapp Harvest/NA Box not polling capacity svm

[Zerolimit]

NAbox 2.5.1b (2019-02-22) Visit Web Page

Debian GNU/Linux 8 (jessie) (8.11) Visit Web Page

NetApp SDK v9.5 installed Visit Web Page (NetApp Login Required)
Harvest v1.4.2 installed Visit Web Page (NetApp Login Required)
Grafana v5.4.3 installed Visit Web Page
Graphite v1.2.0 installed Visit Web Page

Elasticsearch 5.6.15 installed. Visit Web Page
Logstash 1:5.6.15-1 installed. Visit Web Page

API Key : Reveal | Create a new key

I upgraded from 2.4.2 and i have tried fresh install of 2.5.1b and im unable to poll netapp capacity metrics for svm.  Has anyone run into this issue.

[vachagan_gratian]

Hi Zerolimit,

Did you look in your OCUM poller logs? Any warnings or errors there?

[yannb]

You probably didn't configure OCUM poller. For capacity information, you need to have OCUM installed and configure it in NAbox.

If you just configure your cluster you will only get performance metrics.

[Zerolimit]

OCUM is configured and running. In the logs i see this  the errors below. 

[2019-03-14 18:46:03] [WARNING] [volume] update failed with reason: Timeout. Could not read API response.
[2019-03-14 18:46:03] [WARNING] [volume] data-list update failed.

[yannb]

Interesting.

Are you sure that NAbox can communicate with OCUM over port 443 ?

If the answer is yes, any time I had a similar issue it was solved by an OCUM reboot...

[Zerolimit]

Once in a while i would have to restart the OCUM services. I thought it was an upgrade issue when we upgrade NAbox from 2.4 to 2.5x but that was not the case.  I also setup an entirely new instance on 2.5.1b and it still has the same issues.  It is communicating over 443 to our OUCM instance it as performance data is being pulled in.  

[yannb]

So you’re saying restarting OCUM fixes the issue ?

[Zerolimit]

in the past that used to work.

[vachagan_gratian]

Hi Zerolimit,

It seems like Harvest is getting late responses from your OCUM system. What you can do is change the default timeout parameter in Harvest and seek help from support for the OCUM API response time.

You can manually edit netapp-worker to change the default timeout setting. This file is usually located in the folder /opt/netapp-harvest/. You can use vim or nano to edit the file.

On line 612 you should see this:

        $out = $s->set_timeout('60');

Change 60 to 180, save and restart Harvest. If it doesn't help, try again by setting it to 300.

[RajeshPanda]

@Zerolimit  following up to see if you got a chance to look into the recommndeation provided , Lte me know if you are still looking for help.

[amozen]

I have exactly the same error. Can someone provide info for the fix. Communication between NAbox and OCUM via port 443 is working. Increasing timeout to 180 seconds didn't help.