Change Log

Releases

21.08.0 / 2021-08-31

This major release introduces a Docker workflow that makes it a breeze to standup Grafana, Prometheus, and Harvest with auto-provisioned dashboards. There are seven new dashboards, example Prometheus alerts, and a bunch of fixes detailed below. We haven't forgotten about our 7-mode customers either and have a number of improvements in 7-mode dashboards with more to come.

This release Harvest also sports the most external contributions to date. 🤘 Thanks!

With 284 commits since 21.05, there is a lot to summarize! Make sure you check out the full list of enhancements and improvements in the CHANGELOG.md since 21.05.

IMPORTANT Harvest relies on the autosupport sidecar binary to periodically send usage and support telemetry data to NetApp by default. Usage of the harvest-autosupport binary falls under the NetApp EULA. Automatic sending of this data can be disabled with the autosupport_disabled option. See Autosupport for details.

IMPORTANT RPM and Debian packages will be deprecated in the future, replaced with Docker and native binaries. See #330 for details and tell us what you think. Several of you have already weighed-in. Thanks! If you haven't, please do.

IMPORTANT After upgrade don't forget to re-import all dashboards so you get new dashboard enhancements and fixes. You can re-import via bin/harvest/grafana cli or from the Grafana UI.

Known Issues

We've improved several of the 7-mode dashboards this release, but there are still a number of gaps with 7-mode dashboards when compared to c-mode. We will address these in a point release by splitting the c-mode and 7-mode dashboards. See #423 for details.

On RHEL and Debian, the example Unix collector does not work at the moment due to the harvest user lacking permissions to read the /proc filesystem. See #249 for details.

Enhancements

• Make it easy to install Grafana, Prometheus, and Harvest with Docker Compose and auto-provisioned dashboards. #349

• Lun, Volume Details, Node Details, Network Details, and SVM dashboards added to Harvest. Thanks to @jgasher for contributing five solid dashboards. 🎉 #458 #482

• Disk dashboard added to Harvest with disk type, status, uptime, and aggregate information. Thanks to @faguayot, @bengoldenberg, and @talshechanovitz for helping with this feature #348 #375 #367 #361

• New SVM dashboard with NFS v3, v4, and v4.1 frontend drill-downs. Thanks to @burkl for contributing these. 🎉 #344

• Harvest templates should be extendible without modifying the originals. Thanks to @madhusudhanarya and @albinpopote for reporting. #394 #396 #391

• Sort all variables in Harvest dashboards in ascending order. Thanks to @florianmulatz for raising #350

• Harvest should include example Prometheus alerting rules #414

• Improved documentation on how to send new ZAPIs and modify existing ZAPI templates. Thanks to @albinpopote for reporting. #397

• Improve Harvest ZAPI template selection when monitoring a broader set of ONTAP clusters including 7-mode and 9.10.X #407

• Collectors should log their full ZAPI request/response(s) when their poller includes a log section #382

• Harvest should load config information from the HARVEST_CONF environment variable when set. Thanks to @ybizeul for reporting. #368

• Document how to delete time series data from Prometheus #393

• Harvest ZAPI tool supports printing results in XML and colors. This makes it easier to post-process responses in downstream pipelines #353

• Harvest version should check for a new release and display it when available #323

• Document how client authentication works and how to troubleshoot #325

Fixes

• ZAPI collector should recover after losing connection with ONTAP cluster for several hours. Thanks to @hashi825 for reporting this and helping us track it down #356

• ZAPI templates with the same object name overwrite matrix data (impacted nfs and object_store_client_op templates). Thanks to @hashi825 for reporting this #462

• Lots of fixes for 7-mode dashboards and data collection. Thanks to @madhusudhanarya and @ybizeul for reporting. There's still more work to do for 7-mode, but we understand some of our customers rely on Harvest to help them monitor these legacy systems. #383 #441 #423 #415 #376

• Aggregate dashboard "space used column" should use current fill grade. Thanks to @florianmulatz for reporting. #351

• When building RPMs don't compile Harvest Python test code. Thanks to @madhusudhanarya for reporting. #385

• Harvest should collect include NVMe and fiber channel port counters. Thanks to @jgasher for submitting these. #363

• Harvest should export NFS v4 metrics. It does for v3 and v4.1, but did not for v4 due to a typo in the v4 ZAPI template. Thanks to @jgasher for reporting. #481

• Harvest panics when port_range is used in the Prometheus exporter and address is missing. Thanks to @ybizeul for reporting. #357

• Network dashboard fiber channel ports (FCP) should report read and write throughput #445

• Aggregate dashboard panel titles should match the information being displayed #133

• Harvest should handle ZAPIs that include signed integers. Most ZAPIs use unsigned integers, but a few return signed ones. Thanks for reporting @hashi825 #384