Server 112 down?
bothered
Manchester UK
I checked this morning and it seems I sent in almost no finished WUs yesterday. I also have no work to do on one PC and the other has almost finished its WU. I looked at the server, 112, and the CPU load is at 202, all the other servers are at below 2. Does this mean it's down? I thought 502 got work from other servers if there was a problem?
0
Comments
111 and 113 are down tho.
In folding's Work directory, see if you have a PARTIAL set of WU files. If so, and no finished WUs waiting to be sent, erase all files in folder. Then go back to Folding's main client directory. Delete these two files:
unitinfo.txt
queue.dat
Between those two files listed just above, your client figures out what server it talked to or tried to talk to last. You do not have to reconfig, just wipe its info on last WU attempted and yuo will get a random server. In my case, Sunday I stuck SuSE on a HD in the Linux box, and the client kept hanging while connecting (as far as Log). BUT, as far as SERVER 143, it was sending WU files. I got a compressed WU file set, and the client sat so long in .143's queue that it timed out-- and never realized the files were THERE to unarchive. Server .143 was up, down, and always network loaded, 182 to 190 network connects most of weekend. It finally took some downtime as far as network connect late Sunday (US time). At one time point Saturday (U.S. Time, late enough that it was early AM Sunday in UK), the collections server was showing it had accepted 16K WUs for other busy servers. Yeah, the servers are busy, our fast boxes are asking for more and turning in OLD ALMOST faster than the dat server and assignment and collections server farm at Stanford can handle of a weekend.... And this year so far, Folding got at least 4 new servers (total, not of any one subtype)....
We are using the latest version 3.0.2 (i think)
I check task manager for activity and check the logfile for frames done. Very frustrating.
Both systems amd and both no flags.
So, I added the -advmethods flag and immediately each one dloaded core_78 1.69 and began folding gromacs.
Now ...all of my P4's use that flag because up until this week it was helping them to get all gromacs but about half now are folding tinkers.
So apparantly the servers which serve the final wu's aren't offering much so I'll stick with -advmethods and fold almost final-betas until the issue is resolved.
Stay tuned folks ...this could get fun!
If you're interested in adding flags and folding is set up as a service then scroll down a few in this thread for reference...http://www.short-media.com/forum/showthread.php?t=21345
The comment about getting mostly gromacs doesn't apply for the time being.
I agree, after reinstalling folding and EM111 on my PC there is -advmethods set, and I got a 300+point gromac that will take 38 hours. The other PC has run out of things to do so is idle. I hope it's fixed soon, I've handed no results in for a few days now.
And here is the link it comes from at the community forum ...http://forum.folding-community.org/viewtopic.php?t=9991&postdays=0&postorder=asc&start=0
Sorry csimon, saw your post after I posted. I guess we'll just have to wait?
How many machines do you have set up folding?
[settings]
username=csimon
team=93
asknet=no
bigpackets=yes
machineid=1
local=16
[http]
active=no
host=localhost
port=8080
usereg=no
[core]
checkpoint=30
ignoredeadlines=yes
[clienttype]
type=2
nonet=yes
I'll show you something that might make you feel a little better about the current situation ...keep in mind that I have over 30 processors folding ...this is how I know when there is trouble usually.
Why don't you post your log file while you're at it maybe there is a clue in there.
You should receive a huge spike as soon as the issue is resolved.
csimon, can I change my user name back to bothered yet?
Sure ...in 2006!