On several dozen occasions, one machine here is spitting out errors similar to the following:

08-Jan-2017 20:16:20 | SETI@home | [error] No app version found for app setiathome_v8 platform windows_intelx86 ver 823 class opencl_ati5_SoG_cat132; discarding 06ap09ab.25042.5389.10.37.248_0
08-Jan-2017 20:16:20 | SETI@home | [error] No app version found for app setiathome_v8 platform windows_intelx86 ver 823 class opencl_ati5_SoG_cat132; discarding 06mr09ab.15349.7434.6.33.254_0
08-Jan-2017 20:16:20 | SETI@home | [error] No app version found for app setiathome_v8 platform windows_intelx86 ver 823 class opencl_ati5_SoG_cat132; discarding 08se09aa.5377.114936.6.33.247_1
08-Jan-2017 20:16:20 | SETI@home | [error] No app version found for app setiathome_v8 platform windows_intelx86 ver 823 class opencl_ati5_SoG_cat132; discarding 05no09ac.30865.20113.16.43.8_0
08-Jan-2017 20:16:20 | SETI@home | [error] No app version found for app setiathome_v8 platform windows_intelx86 ver 823 class opencl_ati5_SoG_cat132; discarding 06mr09ab.15349.7434.6.33.255_0
08-Jan-2017 20:16:20 | SETI@home | [error] No app version found for app setiathome_v8 platform windows_intelx86 ver 823 class opencl_ati5_SoG_cat132; discarding 06ap09ab.9199.12751.8.35.3_0
08-Jan-2017 20:16:20 | SETI@home | [error] No app version found for app setiathome_v8 platform windows_intelx86 ver 823 class opencl_ati5_SoG_cat132; discarding 06ap09ab.25042.5389.10.37.246_0

Ideas?

They have been having some problems with the roll out of v8.23 ATI apps which will be causing these messages, v8.22 apps should still be running though. They are working on it ....

It's dead. It died days before the Centurion succumbed to his abuse. For about a week now all those 8.23 tasks have been instantly becoming Ghosts. Hundreds of thousands of Ghosts...probably. Quite a few anyway. Your one machine is quite Haunted, the same as many others. Let's see, one GPU & one CPU would be 200 tasks while you are showing;
State: All (1848) · In progress (1261)
So..... 1261 - 200 = 1061 Ghosties. Some machines have many thousands.

That means while operating without any BLC tasks, the Arecibo splitters are having to cope with creating Ghosts as well.
Seems to be working so far...

So Centurion falling ill had something to do with this? It's...unusual for this to go this long, despite being noticed. Was rolling back to previous deployments impossible?

But yeah my newly-tuned 8.22 NV SoG with its tiny, old but still fairly efficient GPU is almost jumping for something more substantial than the average Arecibo task now.

Probably just a coincidence. You know how many of those have been around lately...

Your one machine is quite Haunted, the same as many others.

I suppose Poltergeist can now be re-classified as a documentary...