| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
| World Community Grid Forums
|
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 18
|
|
| Author |
|
|
uplinger
Former World Community Grid Tech Joined: May 23, 2005 Post Count: 3952 Status: Offline Project Badges:
|
One of the file systems is doing a check disk right now. It's still running then I can turn the assimilator back on for some of the projects. After that is done, then I will run the mid day stats again.
Thank you for your patience, -Uplinger |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Keith,
----------------------------------------You could as well skip the rerun of the mid-day stats as my own extract shows nothing at all changed but that time stamp to 12:06UTC. Use the cycles for catching up other processes... such as the validations which I'd guess will be running behind after this longer outage. Running the stats now, if you could, will produce a tally of now 20 hours. Cheers. edit: See it's done. 236 Years it's showing instead of a ''normal'' Saturday noon value of about 165. Looks like 20-25 odd years remain in Limbo assuming it was started after the 19th hours on a "regular'' day :D [Edit 1 times, last edit by Former Member at Jan 22, 2011 8:16:27 PM] |
||
|
|
uplinger
Former World Community Grid Tech Joined: May 23, 2005 Post Count: 3952 Status: Offline Project Badges:
|
Yes, I fear that some of this difference was due to the outage that caused the mid day stats to be off a bit. But since multiple things went down. I was just rerunning them to make sure everything looked ok.
We are waiting on a file system check before we can turn everything back on. -Uplinger |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
seems like everything is working now except for the validater.....
|
||
|
|
uplinger
Former World Community Grid Tech Joined: May 23, 2005 Post Count: 3952 Status: Offline Project Badges:
|
The validator for all projects except c4cw, hcc1, and hfcc should be up and running. They may need some time to catch up to the back log. We are still working on getting everything back to normal.
Thanks for you patience, -Uplinger |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
ty
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
The validator for all projects except c4cw, hcc1, and hfcc should be up and running. They may need some time to catch up to the back log. We are still working on getting everything back to normal. Thanks for you patience, -Uplinger What would be nice to me and maybe others too, if all validators could paused at the dead of midnight till just after the ''historical'' stats lock-in happens at 00:06 UTC. Why you ask? Well, if one adds up the daily history and compares that to the published cumulatives on the Statistics pages, there's always that spill to factor from work validated between 00:00 and 00:06 or whatever time the stats truly commence... last night 3321, when running late, tens/hundreds of thousands. Long as WCG don't publish project level histories, like global, it makes it easy to recompute a missed day, such as Friday due the outage when WCG got online way after the midday stats had passed. With the validators paused for c4cw, hcc1 and hfcc at midnight, those could be abacussed. Now it was a webpage scrape session, still. Mind you, I can see WCG don't want to pause the validators too long as then there's the catch up trouble with now going for a regular 650,000 tasks a day, so on my account, not much longer than the standard time 00:00 > 00:06... say 10-15 minutes. Not watching it, but before there were scheduled validator pauses on a daily basis, so maybe if tight, that time could come off one of the other interruptions. Thinking out loud, could that be done? thanks (and not pushing by adding "in advance" ;>) edit: And it's all for getting me Charts engine running without the weird spikes, the WCG Dashboard especially (See the midnight spill/Morning validation lines... it's presently not adding up to the last tee, being a disliker of "close enough" when it comes to published work :O) [Edit 2 times, last edit by Former Member at Jan 23, 2011 11:55:50 AM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
For the stats, with a near 12 hour outage to include longer time no work fetching possible, the weekend ended with a total of 667 CPU years (287+380). The previous weekend had 672 CPU years (333+339).
1. We seem to be well padded via the Additional Buffer setting in most cases (Yes 1 day is my personal fav setting) 2. The grid is in growth phase, post the holiday month. 3. The WCG team did a superb recovery job... not noticed any backoffs being forced during the restart phase once the upload channels were opened. Forward on we crunch. |
||
|
|
|