Differences

This shows you the differences between two versions of the page.

--- ibm_vios_versus_aix_wpars [2025/04/02 20:13] – [Downsides to VIOS] sgriggs
+++ ibm_vios_versus_aix_wpars [2025/04/02 20:16] (current) – [Downsides to VIOS] sgriggs
@@ Line 35: / Line 35: @@
 VIOS is a big black box. Pretty much none of it is accessible by mere mortal sysadmins. Almost all the time, issues aren't corrected with customer education, they are fixed with patches. Now, IBM is saying they will custom encrypt the VIOS patches so they will only install frame-by-frame on existing entitled customers. So, forget getting one set of patches to apply to your whole farm. In my experience, VIOS fixes are almost all like this: binary blob black boxes responding to APAR bugs.
+Upgrading VIOS is a white knuckled terror. Yes, you might have dual VIO servers in case one fails, but can you clone and rebuild them at will? No? Because you're going to need to do that if one of them cannot be restarted after patching or upgrading it. VIOS patches are big binary blobs that have a pretty janky patch system. Betting the farm is going to stay up while you upgrade is not a fun experience. That's like having to take your operating budget to the casino and bet on Red or Black every time you run low on cash. It's not a great strategy.
 Lastly, VIOS support isn't very well staffed. We hear a lot of complaints (and I've seen it first hand myself) that IBM tends to miss it's SLAs on getting someone who knows VIOS to call you back. Why? Because they offshored all their support to India and very little of this stuff was //engineered// in India. They don't eat their own dog food. They don't have adequate support staff or they fail to manage them correctly, or we wouldn't see folks with a two hour SLA waiting six hours to get a technician.