EMC Storage Analytics (ESA) 3.3

ESA 3.3 posted today on EMC Support and as usual the development team has gone above and beyond for a quarter’s worth of work by both adding to existing platforms and expanding to new ones.  Here are the highlights:

  • Support for the OpenStack architecture
  • Support for VMware VVols on the vVNX (virtual array) – topology and metrics delivered via the VNXe collector
  • Remedial actions for some alerts for VNX, VNXe, and XtremIO – e.g. the ability to expand a file system on VNX when a running out of space alert is generated
  • New XtremIO Volume and Snapshot Tag metrics can be used in association with Custom Groups to group or filter XtremIO volumes and snapshots by Tag value
  • New alerts for RP4VM
  • ScaleIO and VMAX/VMAX3 views/reports
  • SRDF Topology improvements
  • Unisphere for VMAX metrics delivered via SMI-S

For the VMAX/VMAX3 the new features are shown next.  We have 3 new reports shown here:

reportsClick to enlarge – use browser back button to return to post

and 14 new views:

viewsClick to enlarge – use browser back button to return to post

You’ll notice where necessary there are separate reports/views for VMAX and VMAX3 (e.g. VMAX3 SLO List view).  Here is an example of a VMAX3 report, showing the devices:

report_runClick to enlarge – use browser back button to return to post

The reports and views themselves are great, but what I like most about them is the ability to use them as templates to create customized reports and views to suit a customer’s environment.

In addition to the reports and views development has done quite a bit of work on SRDF topologies.  These are most prominently seen in the Inventory Trees.  Here is an example of a two-site SRDF group (I have both VMAX3 arrays configured):

regular_rdfClick to enlarge – use browser back button to return to post

A more extended view is here, where both on the left and right-hand screens you can navigate the topology:

rdf_extendedClick to enlarge – use browser back button to return to post

Note that we still do not support 3-site (most of these devices should be filtered out).  We also do not support SRDF Metro (assuming you have it via the RPQ), however these devices are not filtered so you will see them show up as both R1 and R2.  So it does work, just not in a way where we call out they are Metro devices.  The support should be added in the next release.  As I have it in my environment , this is what it currently looks like:

metro_rdfClick to enlarge – use browser back button to return to post

The final topic I wanted to address with ESA 3.3 concerns a new set of metrics that are not collected by default.  These metrics are only made available to SMI-S via Unisphere for VMAX and therefore it is necessary to tell SMI-S where a Unisphere installation exists from which to gather these metrics.  In the ESA 3.3 Product Guide you will see the following explanation in the EMC SMI-S Provider for VMAX requirements related to SRP and SRDF Director metrics.  I have highlighted it for ease of viewing:

VMAX_requirementsClick to enlarge – use browser back button to return to post

If you are familiar with TestSmiProvider then this is probably straightforward, but for those who are not, and I’m guessing most are not, I’m going to explain how to do this in detail.  The first thing to know is that this is an optional procedure.  ESA will work perfectly fine without it., but the performance metrics for the SRP and SRDF Director objects will not be collected without the registration.  I think most customers will want these metrics and as long as you have an accessible Unisphere for VMAX installation, then it is not a problem. These are the individual metrics that are not collected without this integration:

metricsClick to enlarge – use browser back button to return to post

The first step is to navigate to the directory where TestSmiProvider is installed.  For Linux (which I am using) this is /opt/emc/ECIM/ECOM/bin and for Windows C:\Program Files\EMC\ECIM\ECOM\bin.  Second, execute the program.  In my example screenshot I took all the defaults as I have not changed the default password for admin.  For registration you will need an admin user, you cannot use a monitor role.

Click to enlarge – use browser back button to return to post

Here at the main menu, you will need to get into the “active” screen where you can register Unisphere.  Type in “active” and now from this sub-menu register Unisphere by typing “reg”.  To register you will need the Unisphere FQDN or IP address and a user with at least monitor role.  Again, I have taken the defaults, using the admin user “smc”.

smi_2Click to enlarge – use browser back button to return to post

That’s it.  You can type “reg” again if you want to see the registration.

smi_3Click to enlarge – use browser back button to return to post

It will take a few hours to start seeing the metrics.  I should note that if the SMI-S environment you are using is on the same host as Unisphere, registration is NOT automatic.  You still need to register it.  The Unisphere vApp, which has this combined SMI-S-Unisphere implementation, is not supported because of the inability to access the root user in the vApp.

A final housekeeping item, ESA 3.3 is not yet certified with vRealize Operations 6.1 which recently released.   We are in the midst of certification, however, which will complete before the next ESA release.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Create a free website or blog at WordPress.com.

Up ↑

%d bloggers like this: