#Cluster Software Validation: What does it mean?

1 messages · Page 1 of 1 (latest)

twilit crystal
#

A quick question regarding the Software panel under "ONTAP: Cluster". I see a block of yellow "Warnings" in the "Cluster Software Validation" table. For example, one cluster is shown as having warning statuses for several "Pre-update Checks" e.g. for "Cluster/managment switch support". But what does it mean? Some check returned a warning, but what was the check, what was the warning and when did this happen?

I tried to trace this back to the source (endpoint) in ONTAP, but in the Harvest reference document the metric "cluster_software_validation" is described as "Harvest generated" ... so I guess it is not actually coming directly from ONTAP?

I'd guess this might be based on selected info. extracted from the ONTAP Job history?

(Simply clicking on the warning (my first thought) doesn't result in any further info. being shown.)

Thanks in advance!

plain vapor
#

@twilit crystal This is coming from API api/cluster/software?fields=validation_results

twilit crystal
#

I see. I had a look. AFAICT that data is the result of the validation checks performed before an update to a new ONTAP version is installed, it is coming from running "cluster image validate" on the cli. Maybe.

#

I can't tell if it returns any timestamp info. to indicate when the check was carried out, so it's not clear to me that it really fits into the Harvest (Grafana) model ... IMHO this seems to be intended more for a orchestration app, one that is actually perfoming the update ...

plain vapor
#

There is no timestamp information for validation results from ONTAP. We have only included status information and other fields like reason are omitted to keep it time-series friendly.

twilit crystal
#

Exactly, that's what I meant by "doesn't really fit into the model". Something like "cluster image validate" typically only happens (I think) as part of the ONTAP update process. Depending on the end customers own processes, that might only happen a couple of times a year ...

plain vapor
#

@manic vine can add more on this, as it was requested by him.

manic vine
#

I raised this to get more intimate details regarding ONTAP version deployments. These are especially helpful when a rollout fails and leaves a cluster in a bad state as it’ll be reflected here. We have a large 6 node cluster presence and it’s easier to manage these rollouts with this information.

round sun
#

Thanks for the details, @manic vine