Another release of EMC Storage Analytics is on the books. For those keeping track these come out quarterly (or as close as we can get them). Our last release (2.3.2) was minor in that it basically covered a couple new models on the VNXe line. This release is major, adding a new platform: VPLEX (note that the release is called 2.4.1 – there is no 2.4). We support both Local and Metro configurations (no Geo yet). On the VPLEX you can traverse from the Metro Cluster itself down to the individual volumes on the underlying arrays. Here is a screenshot of the topology (you’ll see my lab box is a bit beaten up and ESA is letting me know it):
Like the other platforms, there are four platform-specific dashboards delivered for VPLEX, along with the two universal ones. Having developed the VPLEX ones, I can tell you I tried to make them similar to the other arrays with the exception of the VPLEX Overview. Unlike the VNX/VNXe or VMAX, the VPLEX reports health and operational states on many objects so I took advantage of that for the dashboard. Here is a sample of the Overview for a large environment, showing some problems that need looking into:
The other VPLEX dashboards are Performance, Communication, and Topology, each presented below:
If you happen to have VMAX arrays behind your VPLEX (as I do), the ESA code will connect the topologies of VPLEX and VMAX if you also add a VMAX adapter instance to vCOPS. In my environment I have a VPLEX Metro with one cluster backed by a VMAX ending in 068. I have added the VMAX as an adapter instance. When I drill-down into the topology of VPLEX from my extent to my volume (Symm0068_0477), if I have the VMAX instance, it will show me the VMAX device (00477) as a child resource. I can then drill-down into that device and I am into the VMAX topology. Pretty cool stuff.
While VPLEX was the big item in this release, we did get a nice enhancement to VMAX – FAST VP. The FAST VP Policy will now appear as a child of a storage group if there is one. You can drill-down into it and see the tiers and the associated thin pools. There are just a couple fields as metrics for these. I did a short video (2 minutes) walking through it:
I also updated the VMAX Topology to account for the new resource.
One caveat I will point out (as does the Release Notes) is that vCenter Operations 5.8.3 is NOT supported with ESA. 5.8.2 plus the ShellShock patch is the highest supportable version. There is actually a good reason for this – because of a VMware bug in 5.8.3, ESA breaks. We’ll let you know as soon as that one is fixed.
Future releases (disclaimer: easiest just to say no promises) will include some VMAX event correlation – for instance give me an alert on my thin pool resource when I hit 80% full – and of course VMAX3 support. Note there is some capability with alerting if you combine vCenter Log Insight and the VMAX Content Pack with ESA. I’ll try to put together a post on that while we await ESA inherent capabilities.
Here are the links to the docs and software. BTW if you use ESA with VMAX I would be very interested in hearing about what else you would like to see. Just leave me a comment. Thanks.