Main project server failed about 36h ago due to I/O errors because of bad sectors on it's hard disk. I've migrated the data to the secondary disk and the project services are now recovering. There might be few tasks affected (will be shown as download errors) as some generated data files were not recoverable.

IMO server should be running on a RAID.

what?? The main server without a RAID? Are you kidding me?

Look, it's a bit of a rookie mistake not to use at least RAID-1 or RAID-5 on a server. RAID-6 would be better if possible. As far as recovery goes, try using ddrescue if possible. It works pretty well and has an awesome recovery rate. And, for Pete's sake, use a Raid array. You have too much going on not to.

What kind of server he have? An old pentium 4 era server with only one hard disk ?

Hopefully it's a different issue, but it appears that Moowrap stopped talking to the Distributed.net keyservers to report block ENTIRELY sometime early on July 8.

Well, whatever the issue was, it's either fixed or was short-term.

Main project server failed about... after few days online again