View Full Version : Berkley is doing great!
I cant believe how good they are doing... its just amazing :thup: .
Tons of W/U's are ready and the validation is exteremly fast since that added that 4th validator.
Hope it can stay like this for longer than a few days.
zulfy26
02-27-05, 05:15 PM
when are they gonna update the XML?? its been over 4 days now :-(
heezer7
02-27-05, 06:09 PM
Yeah really. I keep checking on my app because I am not getting notices of it updating teh web site. Kind of annonying....
How is S@W still getting stats updates?
February 27, 2005
The project is down for a short bit while we clean up the upload/download volume.
Well AT LEAST IT WAS SCHEDULED!!
Adam
Steven4563
02-28-05, 12:50 AM
hmmm ive uploaded over 40wu's and all ive seen is about 400 credits i hope that changes or ill be abit pi**ed :mad:
How is S@W still getting stats updates?
S@W dosen't rely on the XML data for it's updates. Instead, it updates user by user by going through the database for the latest stats (which are updated farily close to realtime regardless of whatever is happing with the XML).
@Steven - My credits don't seem to be lagging much (I've only got a handful pending), so I'm not sure what's up. Are you looking at credit from a site that uses the XML updates, or Berkeley's own account data? If it's a site using the XML, that's probably the reason, since that puts me about 4K behind reality.
JigPu
heezer7
02-28-05, 02:36 AM
Yeah, JigPu is right, but them doing this creates a massive load on the berkeley servers every hour. They have to parse through the web site, through each users' page to get their current information. I have been working on our new stats page and will NOT do it this way. I believe that it is the "wrong" thing to do. Berkeley releases the XML files only once a day because their servers are under such a heavy load already. They want the stats site to use this and I respect that. They will release data faster once they have the hardware to support it. Massive numbers of web site hits each hour is NOT helping with thier problems at all.
Jon
FloridaBear
02-28-05, 09:31 AM
Yeah, JigPu is right, but them doing this creates a massive load on the berkeley servers every hour. They have to parse through the web site, through each users' page to get their current information. I have been working on our new stats page and will NOT do it this way. I believe that it is the "wrong" thing to do. Berkeley releases the XML files only once a day because their servers are under such a heavy load already. They want the stats site to use this and I respect that. They will release data faster once they have the hardware to support it. Massive numbers of web site hits each hour is NOT helping with thier problems at all.
Jon
I agree. By the way, you've shot past me in the rankings! Congrats on the top 50...
Steven4563
02-28-05, 11:53 AM
im looking on Boinc seti@work website
where can u see how many are pending ??
Basically all my pendings have cleared and given credit - if anyone is still having trouble out there I would say you won't be for long - they seem to be cranking pretty fast now. Also keep in mind that sometimes you'll get a batch of WUs that are all waiting on the third cruncher out there to upload - I've had times where almost 100 WUs were waiting on others, very annoying - but also is the way it is supposed to work.
@Heezer: Grats on your T50 mate!
Hmm lol seems like Brekeley is down again. From web site:
"Around 18:00 UTC we had another unexpected lab-wide power outage. Systems were able to shut down more gracefully than last time, but we are leaving many services off as we survey the damage. The cause of these outages is still unknown."
My fault - I forgot to say "knock on wood" :) They got everything gracefully shutdown though - so there shouldn't be any major downtime once the power comes back.
Edit: re "there shouldnt be any major downtime" -> Knock On Wood :p
heezer7
02-28-05, 05:53 PM
Thanks for the congrats. I put seti on the dual opteron server here at school. I am its only admin and I looked at the logs and saw no one has logged into it all year besides me. Only a few few number of people have access to it. Now if i could only get my dad to install it on the 2 dual opteron 246 servers we just built i would be in heaven... :-)
Thanks for the congrats. I put seti on the dual opteron server here at school. I am its only admin and I looked at the logs and saw no one has logged into it all year besides me. Only a few few number of people have access to it. Now if i could only get my dad to install it on the 2 dual opteron 246 servers we just built i would be in heaven... :-)
Excellent - and indeed it is so frustrating to see powerful crunchers not crunching - have you tried begging? :)
heezer7
02-28-05, 07:16 PM
yeah, it sucks. One is a DB server and the other is just the application server. He does put a pretty good strain on them. I was like, I can have it run just at night... no luck. I will fight him a bit more over the summer once I am home. Worse part is I picked out every piece and built the damn things for him. Oh well. I am still doing pretty good.
Cúchulainn
02-28-05, 09:13 PM
The cause of these outages is still unknown."
The E.T.I.s are apparently already here and don't want to be found :p
SunRedRX7
02-28-05, 09:23 PM
Wouldn't be the first time aliens were blamed for a power outage.
UFOs & POWER OUTAGES - BLACKOUT RELATED TO UFO SIGHTINGS? (http://www.mt.net/~watcher/ufoniagarafalls.html)
Living near NFalls, driving past one of the plant's relay stations, my Dad would always point out thats where the UFO stopped to fuel up and took out the grid.
lol,
these guys have absolutly no luck...
maybe they can find the cause of the outages. At least all the servers are ok :)
SunRedRX7
03-01-05, 07:59 AM
Looks like there gonna take care of a bunch of things while its down.
February 28, 2005 - 22:30 UTC
So we had another unexpected lab-wide power outage again this morning. This time around we had the BOINC database on battery backup so we were able to shut it down safely. After the power returned we brought the database back up briefly to check it out - and it's in perfect health. You can all thank Court for bringing in his personal UPS (and leaving his own systems unprotected) to put on the BOINC database server until we were able to obtain a new one.
But we shut the BOINC database right back down, and will leave most of the BOINC back-end services off for the time being until we have all our important systems on smart UPS (the systems will shut themselves off once they realize they are on battery power). This has always been the future plan (and please note that our previous configuration allowed for zero or minimal loss in the event of a power failure), but now that frequent random outages are part of the scenario, it would make life easier not to have to do damage control every time.
We are actually going to take this time off to do additional maintenance. For example, the disk array holding the upload/download directories is 98% full - Jeff discovered a bug in the file_deleter code that left a lot of old workunits around. So we need to get rid of those stale files before anything else.
Gotta look on the sunny side, everbody should have got their caches filled huh? As long I can keep crunching I can live without my stats for oh a day or two??
Boy, talk about running a project on a shoestring budget...
Do they have these servers running on utility power? Are they even doing backups?
SunRedRX7
03-01-05, 08:36 AM
A lot of the budget problems and lack of resources have to do with them running Classic and BOINC. Once they feel BOINC is stable enough to take over the full brunt of classic users, they'll be able to move hardware over to BOINC, and clear out closet space and electrical resources for BOINC.
There pretty close to being ready for BOINC, they were saying the servers were barely being pushed with the current load, which they said was about 25% of what there looking to do.
If this electrical stuff didn't come up they'd be doing pretty good right now I'd bet.
Also, remember that one of the main purposes for BOINC was that in times like this, your CPU cycles need not go to waste!
Also, remember that one of the main purposes for BOINC was that in times like this, your CPU cycles need not go to waste!
Indeed - Caches people, Caches! That is what they are there for :) If you are running out of work units during outages, jack your caches to 10 days. I have yet to run out of WUs for any of these outages and my cache is only set at 5 or 6 days.
Seti Classic had all kinds of problems like this too (especially at the beginning), and is indeed one of the major reasons software like SetiQueue was developed :)
vBulletin® v3.8.7, Copyright ©2000-2012, vBulletin Solutions, Inc.