Been getting this for a while:

02/06/2011 01:38:10 | Moo! Wrapper | Server can't open database

No uploads going through, I presume the same will therefore go for downloads

Forum very slow as well, presumably connected with the same problem. Been slowly getting worse, as if its slowly choking, dont know if its network retrys or directly a Db issue

Regards
Zy

Murphys Law ..... cleared now, appears ok. It had been going for about 20mins before the initial post.

Regards
Zy

Hi,

Ah, thanks for letting me know. It's possible there's something funky going on with the DB and I've seen some back end processes restarting randomly. Sometimes there's also huge load on the server.

I'll try to keep an eye on this even though server has recovered automatically by itself. It's about time that we start purging oldest tasks so that might help with DB load.

-w

I did an all nighter last night for various non-BOINC reasons, and I did notice this happening fairly regularly after my two posts above. Seemed like it was roughly every 2 hours or so. I also noticed a fair number of server http errors like below appearing at non regular intervals:

02/06/2011 17:06:09 | Moo! Wrapper | Scheduler request failed: HTTP internal server error

{thats UK Time this end - UTC+1}

Had a few during the day today, although not as many as last night

Regards
Zy

Hi,

Yeah, the DB server was having problems due to memory shortages and tables that have been growing large. I've been re-configuring the DB and optimizing memory usage on the server. (That's why there have been couple maintenance breaks in last two days.)

Hopefully things are (or getting there) better now but I'll keep an eye out on the scheduler logs.

-w

You mentioned some Db troubles you are working on re expanding Tables. Take a look at my aborts for my second machine (ID: 479, Name: Dot-PC) . I had a couple of abort sessions over the last month due to my second machine playing up. There are now 245 aborts in the error list, however vertually all have been cleared and crunched by others, yet they are still listed against my errors, and dont appear to have been cleared out.

If thats repeated across the board as a process (or missed process), there might be a substantial number in the tables that should have been cleared out a while back. I dont know the Db schema, so I dont know if this makes any difference .... but might be worth looking at.

Regards
Zy

If thats repeated across the board as a process (or missed process), there might be a substantial number in the tables that should have been cleared out a while back.

Yeah, we haven't actually purged the DB at all yet and that's certainly part of the problem since DB has grown large. Purging is running but we have been touching results lately due to various problems so they are not yet eligible for a purge. I think active purging will start happening withing two weeks but I'm certainly watching this process (to rule out bugs in the process).

-w