I got all the way to the end of receiving and it sat there for quite a while and then said "unable to process task data, backing off". Now I am back in the "connecting", "paused" cycle. There was no update on points on this system.

Done! It's back up and I already have a new work unit.

The problem hasn't exactly been solved... but it isn't a complete shutdown of the system.

The problem is that a certain fraction of the connections fail and lock up a listening port until that conneciton times out. After a while most of the ports are waiting for these timeouts.

I assure you that we are doing everything we can to get everything working properly.

Grid server seems to be up and running again properly now (22:08 here in UK)

David.

I noticed that the WU that the agent hadn't been able to send back earlier had finally gotten thru and a new, small WU downloaded and running. When the new WU ended, it would pause after trying to connect to the server. I basically run thru the process of getting it to where the small "help" window was displayed where I could click on OK to get the agent to retry connecting. After 4-5 tries, it finally did and successfully returned the latest WU and did get yet another one - although there seemed to be a little "hesitation" at the start of downloading the new WU. Have to wonder though if that ends up tieing up ports that are "hung" until the timeout and, while helping me, hurts the overall situation???