Known Issues

No software is without its faults, so I’ve tried to include the ones I see most often in customer environments along with the state/workaround/resolution. When I say “often” I simply mean the support case makes it way to me. These aren’t common issues by any means. They impact a very small number of customers but I hope by including them here it can save you time as you won’t have to open a support case with Dell or VMware.

Currently all issues revolve around VMware SRM.

EnableAutoSSLCertGen.sh in SRDF SRA 9.2 not persistent

When the SRDF SRA 9.2 was first released, the script we included to set the hostname for the SRM appliance (since Docker generates its own hostname which breaks Solutions Enabler) was not persistent across reboots. This meant your SRM environment would break every time you had to restart the VM. I blogged about options to avoid this, but certainly it was an unsatisfactory resolution. We therefore created a hot fix for 9.2 which avoids the problem (10.0 includes it also). At this point, with SRDF SRA 10.0 available, customers should either be at the hot-fix level or the new release, however as I noted there are workarounds in the blog if you can’t upgrade.

Multiple SRM SRAs in the same environment

This issue can occur if you use the SRDF SRA 9.2.x in the same environment with another SRA (ours or another vendors). Because we used a generic tag for our SRA, we can place our hostname file in the wrong docker volume (e.g., the other SRA). This leads to problems with Solutions Enabler communication. I wrote a blog post for this one to explain it.

As this is Dell’s issue, the fix is in the SRDF SRA 10.0, but if you follow the blog post it explains the workaround. There is also a KB I wrote.

Duplicate device IDs across arrays

This is a very rare bug for SRM that requires at least three arrays. It’s easiest to explain this from the KB I wrote.

If two array devices in two different array pairs have the same device ID, e.g., 0003A, VMware is unable to properly identify the protection group for one pair. For example, a customer has three arrays (though they could have four), A, B, and C. They have configured pairs between A and C and B and C. By happenstance, one set of pairs from each has the same device ID on A and C: 

Pairs from A to C: 0003A – 001B8
Pairs from B to C: 0003A – 000F0

These are independent pairs, and the SRDF SRA has no issue recognizing them as shown below:

After creating the Protection Group for each pair set, however, browse each one and look for the pairs. One of the Protection Groups will not recognize the device pairs, or sometimes it shows the wrong pair.

VMware is aware of this bug and will fix it in SRM 8.6 which is the next release. In the meantime, the only workaround is to Storage vMotion the VMs from one of the pairs to a different device ID.

Different SRDF SRA versions at protection and recovery site

Here is a reminder rather than a bug. We don’t support rolling upgrades with the SRDF SRA. Each SRM site in a pair requires the same version of the SRDF SRA. You need to upgrade them together.