In another thread someone said that is the last thing they are working on. Trying to get the project servers to work right.

Thank you!

Well at least the forums are up.

Can count on the contributors here to give much more information than anyone from Krembil. To say that I'm annoyed with their radio silence for days on end is an understatement.

Problems fine. Come with the territory. Lack of communication shows lack of leadership.

interesting how many active users been before switching to a netw cloud?

Interestingly, there was some communication over the weekend, just not in a place many people were likely to see it :-(

At about 23:20 UTC on 2025-09-21, Dylan posted something in the Teams forum in response to a comment about sending email to team members, and [with slight editing] it would have made an ideal update! (It got tagged in our Project Status thread in the Chat Room but it wasn't quoted in full...)

I'm in favour of better communication, but (given the team size and other responsibilities) it's probably not going to happen if there isn't someone with enough time to take that on. The intern experiment seems to have been abandoned (or become infeasible for some reason) and as I don't believe in the magic money tree I don't see where a spare body will come from...

On the second point... I've seen plenty of evidence that WCG (probably always the same person!) do work over weekends if they become aware of system crashes, and the time of the above-mentioned post coupled with a dramatic improvement in API reliability since the end of 2025-09-21[*1] points to weekend work having been done that recently. That said, there are sometimes things that can't be done without actions by data centre staff, at which point the "nobody working on the weekend" comment is fair and valid!

Cheers - Al.

*1 -- I have API scripts that get invoked several times an hour, 24 hours a day, and have seen one brief issue (less than half an hour) on 2025-09-22 but nothing since. Prior to that, since the post-migration restart a good day was one with less than 12 hours of API problems!