NetApp Harvest 1.4.1 released

[vachagan_gratian]

We are happy to announce that Harvest 1.4.1 is released!

You can download the new release from ToolChest. Main updates are:

- Harvest will accept older templates, no need to rename them manually

- Harvest will aggregate FlexGroup volume counters

- We also added a new dashboard in Grafana for FlexGroup counters.

As we continue to develop Harvest, your feedback is much appreciated. If you use Harvest, please share with us:

- What issues do you face when installing Harvest?

- Any bugs you noticed in Harvest? What makes you happy or unhappy when using Harvest?

And last but not least: what new features do you think we should add to the next release of Harvest? We ask for your feedback because we want to work on features that many of you need.

I look forward to your feedback!

Cheers,

- Vachagan

[PabloZorzoli]

Hey @vachagan_gratian,

I was trying the brand new version, and had to slightly patch the new grafana dashboard for flexgroups:

See: https://pastebin.com/avgn0D80

Basically some datasources still have refs to "Graphite" I guess it's your testing environment, and some prefiled values for the template variables.

Other than that no other issues observed, thanks for keeping Harvest alive!

Pablo.

[vachagan_gratian]

Hi Pablo,

You are right! Thanks a lot for pointing this out, I'll make sure to correct this and update the package.

[elic_co]

I ran into an issue with nabox. after upgrading the harvest it seemed to disappear. connecting to the server via ssh showed an error message regarding permissions in a certain path. long story short after chmoding two files (netapp-worker) and another one it seemed to work again but weird

[vachagan_gratian]

Hmm, this shouldn't happen... Did you use the web interface to upgrade Harvest? Because I tested that and everything was ok (both netapp-manager and netapp-worker are executable). And which version of NAbox do you have?

Thanks, V.

[TechopsVendor]

We have 1.4.1 running  against about 80 heads  7-mode and cdot 

We upgradesd multiple  cdot clusters to 9.3P8 and now ethernet graphs are empty while all other graphs seem to work??

[digitalsphere]

@vachagan_gratian is this new harvest tested with release 9.4.P3?

We did an upgrade to our old monitoring system after upgrading to 9.4P3, now it's just this:

• NAbox 2.5b
• NMSDK 9.4
• Harvest 1.4.1

But we are not able to connect to our cluster in that version (while the rest of them are working like a charm) because we are getting this error:

[WARNING] [sysinfo] Update of system-info cache DOT Version failed with reason: in Zapi::invoke, cannot connect to socket

We just tried cloning the template from the last cDOT 9.3.0 but it seems like a dead end here, anything else to give it a try?

Best regards.

Jose.

[dkon]

check https://community.netapp.com/t5/OnCommand-Storage-Management-Software-Discussions/Harvest-for-ONTAP-9-4/td-p/141408

[KTT]

Hi Vachagan,

We've created FlexGroup volumes on ontap 9.4P2 recenty. However we don’t have insight into the space used by the FlexGroups in Harvest 1.4. All I could find is some basic performance data of the FlexGroup and some additional on its constituents. The capacity counters do not pull up any volumes related to FlexgGoups at all. Would like to know if Harvest 1.4.1 does provide the insight into the space used by the FlexGroup volumes? Thanks 

[vachagan_gratian]

Hi,

Sorry, apparently I'd missed your question.

In 1.4.1 we have a new dashboard for FlexGroup volumes (we don't have much insights, though, mainly latency, I/Os and Throughput). Did you try 1.4.1?

if there is any other specific counters that you would like to see in this dashboard, let me know, maybe we'll add them in 1.5.

[lornedornak]

We have 1.4.1 installed and now we can collect performance metrics for flexgroup volumes, but it's missing flexgroup capacity metrics.  When will harvest be able to pull flexgroup capacity metrics from OCUM?

[Zerolimit]

has this been resolved.  Im have tested with 1.4.1 and 1.4.2 and im still unable to poll netapp/capacity/cluster/site/svm metrics

[vachagan_gratian]

Hi lornedornak,

This is a bit of an issue, since the ZAPI interface we use to collect counters from OCUM doesn't give us FlexGroup information. We have now an experimental package that does collect FlexGroup counters alternatively. If you want to try it out, message me.

If everything goes fine, we'll integrate this package with the next release of Harvest.

[rezkmekhael]

can you please send me the  collect FlexGroup capacity package . i like to see "size,used,used_percentage" under capacity like flexvolume in dashbaourd 

Thanks

[Zoltan_Somogyvari]

Hi,

I've faced the same issue, did you manage to solve it?

I mean the flexgroup volume capacity statistic.

Thanks

Zoltan

[Benoit_Plessis]

As for the new fonctionnalities:

Graphite is very I/O heavy, so personnaly i have setup an InfluxDB service with graphite end-point and manually re-created some dashboard over the InfluxDB language.

It would be huge for harvest to natively support multiple TSDB storage (with or without the whole dashboard sets)

[crazy_steven]

Hey @vachagan_gratian:

   I have a question to ask,Why only netapp template on my grafana without monitoring data?What problems might arise?

• Please help me why graphite cannot get stored data...

  ---