Yellow Bricks

by Duncan Epping

Leave a Comment

Changing advanced vSphere FT related settings, is that supported?

This week I received a question around changing the values for vSphere FT related advanced settings. This customer is working on an environment where uptime is key. Of course the application layer is one side, but they also want to have additional availability from an infrastructure perspective. Which means vSphere HA and vSphere FT are key.

They have various VMs they need to enable FT on, these are vSMP VMs (meaning in this case dual CPU). Right now each host is limited to 4 FT VMs and at most 8 vCPUs, this is being controlled by two advanced settings called “das.maxftvmsperhost” and “das.maxFtVCpusPerHost”. The values for these are, obviously, 4 and 8. The question was: can I edit these and still have a supported configuration? Also, why 4 and 8?

I spoke to the product team about this and the answer is: yes, you can safely edit these. These values were set based on typical bandwidth and resource constraints customers have. An FT VM easily consumes between 1-3Gbps of bandwidth, meaning that if you dedicate a 10Gbps link to it you will fit roughly 4 VMs. I say roughly as of course the workload matters: CPU, Memory and IO pattern.

If you have a 40Gbps NIC, and you have plenty of cores and memory you could increase those max numbers for FT VMs per host and FT vCPUs. However, it must be noted that if you run in to problems VMware GSS may request you to revert back to the default just to ensure the issues that occur aren’t due to this change as VMware tests with the default values.

2 Comments

Using a vSphere custom TCP/IP Stack for iSCSI

I noticed a question today on an internal slack channel and it was about the use of custom TCP/IP stacks for iSCSI storage environments. Cormac and I updated a bunch of Core Storage whitepapers recently, and one of them was the iSCSI Best Practices white paper. It appears that this little section about routing and custom TCP/IP stacks is difficult to find, so I figured I would share it here as well. The executive summary is simple: Using custom TCP/IP stacks for iSCSI storage in vSphere is not supported.

What is nice though is that with vSphere 6.5 you can now set a gateway per vmkernel interface, anyway here’s the blurb from the paper:

As mentioned before, for vSphere hosts, the management network is on a VMkernel port and therefore uses the default VMkernel gateway. Only one VMkernel default gateway can be configured on a vSphere host per TCP/IP Stack. You can, however, add static routes from the command line or configure a gateway for each individual VMkernel port.

Setting a gateway on a per VMkernel port granular level has been introduced in vSphere 6.5 and allows for a bit more flexibility. The gateway for a VMkernel port can simply be defined using the vSphere Web Client during the creation of the VMkernel interface. It is also possible to configure it using esxcli. Note: At the time of writing the use of a custom TCP/IP Stack is not supported for iSCSI!

I hope that clarifies things, and makes this support statement easier to find.

6 Comments

Where did ESXi 6.5.0 build 7526125 go?

I had two customers asking today what happened to ESXi 6.5 build 7526125. They downloaded patches and installed them in their test environment. Ready to patch some of their clusters they did a validation and found out that the patch (ESXi650-201801001.zip) has disappeared from the face of the earth. This patch included microcode for Intel processors, and Intel informed VMware that there was potentially an issue with their microcode. As such VMware decided to pull the patch as noted in the KB article. Those who had already downloaded the patches and are manually updating, make sure to delete these. Those who use VUM, make sure to exclude them from your baseline as mentioned in the KB:

Any baseline (including VMware Pre-defined Baseline), that includes one or more of  the bulletins that  correspond to patch VMSA-2018-0004, would experience the above listed error and hence, will not be able to proceed with the remediation process. For such customers, it is recommended to create dynamic or static baseline excluding the bulletins ESXi650-201801401-BG, ESXi650-201801402-BG,  ESXi600-201801401-BG,  ESXi600-201801402-BG ,ESXi550-201801401-BG and continue with the remediation process. For more information on Create and Edit Patch or Extension Baselines see vSphere 6.5 document.

Normally I don’t share these types of things anymore, but as I had two people asking on the same day I figured I would as it seems not everyone had seen that the patches were pulled and replaced. If you haven’t downloaded the patches yet, or haven’t patched your systems but want to, read this advisory first and use the patches mentioned it.