netapp-harvest: support for ontap 9.2 and 9.3?
2017-10-23 01:28 AM
Is someone working on getting support for ontap 9.2 and 9.3 to netapp-harvest?
it would be nice to get a better suport then just reusing the templates from 9.1
1 REPLY 1
Re: netapp-harvest: support for ontap 9.2 and 9.3?
2017-10-24 06:16 AM
I hesitate before continuing with my reply to your post. I don't want to sound preachy or judgemental. Your post touches on a topic about which I feel rather strongly in regards to "community support" tools and software.
With respect to your desire (and it is one I sort of share), the expectations for Harvest are very clear.
Harvest is a "Community Support" tool - meaning that it pretty much falls to the community or to Chris Madden of NetApp to make such changes. [ Chris - I don't know if updated version support falls to you personally or if you have a/an formal/informagl group that helps you. ]
When using Harvest we all accept that this is the level of support we can expect - questions, fixes, new versions, etc. are disussed here only. Updated releases of the tool will not necessarily track any particular NetApp standard product release, not that they are generally far behind. Because of the work Chris did in creating the tool, if anyone has a particular need we can add that in ourselves until a more formal template is available.
Perhaps the formal template doesn't even cover exactly what any customer needs. Harvest as a framework is seriously well suited to anyone's specific custom need - a tribute to the work that went into creating it. Given that Harvest is distributed fully as source code, we have complete freedom to make it meet our particular needs at anytime.
I have a similar personal example with the author of Grafana in trying to request a new capability in that tool. The request is that the source datastore associated with a dashboard panel template should itself be templated. Right now the definition of the template requires an explicit datasource to populate the template selections. Seems obvious for a particular use case I have. However after a lot of discussion with the author Torkel, he didn't see or agree with the need, so the request didn't get implemented..
I still could do it myself if I really wanted it, and maintain it. The key is I understand the level of support I can expect - maybe something I want gets done in the time I want it done, but I can never depend on it explicitly. Such is the contract we have with authors and maintainers of community support software. To have these tools available, and at such quality, is way more important, in my opinion, than any nits regarding some basic support issues. We know there is momentum and that this will be forthcoming.
And on the day when it is no longer forthcoming, we accept that too as part of the deal we made to use the tool.
Ok. I should probably get off my soap box now. I hope everyone will accept this reply in the spirit in which I intend it.
Senior System Enginer | cStor
NCIE SAN ONTAP, Data Protection