| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Community
Forum: Chat Room
Thread: Project Status (First Post Updated) |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 260
|
|
| Author |
|
|
Unixchick
Veteran Cruncher Joined: Apr 16, 2020 Post Count: 835 Status: Recently Active Project Badges: 
|
That was a rough weekend. The website went down for a while, and the WUs are down to a trickle. It looks like they are still working on the MCM results issue, my list has shrunk a little bit, but not much.
There also hasn't been an update on ARP in a while. |
||
|
|
Grumpy Swede
Master Cruncher Svíþjóð Joined: Apr 10, 2020 Post Count: 2068 Status: Offline Project Badges: 
|
My MCM1 result list, has shrunk with around 40 results. Lots of them left to purge though.
|
||
|
|
adriverhoef
Master Cruncher The Netherlands Joined: Apr 3, 2009 Post Count: 2069 Status: Offline Project Badges: 
|
Adri - your page link isn't working for me. Unixchick, unaux.com is working again, the webpage with my latest downloaded general task has finally re-appeared.  [Edit 1 times, last edit by adriverhoef at Mar 5, 2024 4:31:55 PM] |
||
|
|
Unixchick
Veteran Cruncher Joined: Apr 16, 2020 Post Count: 835 Status: Recently Active Project Badges:
|
Doesn't look like the WCG can do two things at once. It can either send out all the WUs we need, or it can process the results. It is time to fire up that backup project, and let WCG work through the backlog of results.
|
||
|
|
adriverhoef
Master Cruncher The Netherlands Joined: Apr 3, 2009 Post Count: 2069 Status: Offline Project Badges:
|
While looking at my latest downloaded general task I have noticed that there are gaps that can last typically about half an hour to maybe several hours in which the distribution of MCM1-work is being put on the back-burner temporarily (while old results still are being purged at a killer pace).
Normally, 10,000 workunits (20,000 tasks) are being distributed in about 20 minutes to half an hour, which amounts to 1 or even 1½ million tasks per day; that is, mainly MCM1-workunits. Yesterday, only 600,000 tasks were distributed, so one could say that the distribution of tasks is at 40-60% of its 'normal' capacity. I think the techs are trying to find a good balance between distributing a fair/reasonable amount of tasks and clearing/purging old results from the system at the moment, while still trying to keep the validation process of only a small amount of workunits 'alive'. That, or there is a new problem, which I don't hope, while keeping a good balance is a problem in itself.  Adri |
||
|
|
Just Jake
Cruncher Joined: Nov 15, 2018 Post Count: 22 Status: Offline Project Badges: 
|
Work units dropped to a dribble on 3/1 and nothing after 3/2. Still no work units on 3/5. Reset Project leads to a flash of downloads which never become tasks. Is it me or is it Krembil? Folding@Home is getting a lot of processing recently!
|
||
|
|
Unixchick
Veteran Cruncher Joined: Apr 16, 2020 Post Count: 835 Status: Recently Active Project Badges:
|
Thanks for the information on flow Adri. I'm glad they have a method to balance things while they clean up. This is a temporary situation, so I'm glad we can still get some WUs during this time.
Just Jake - WUs aren't as plentiful as usual, but you should be getting some now and then. If you are getting downloads and not tasks you might want to look at your log (under tools) to see what is happening. |
||
|
|
Just Jake
Cruncher Joined: Nov 15, 2018 Post Count: 22 Status: Offline Project Badges:
|
Just Jake - WUs aren’t as plentiful as usual, but you should be getting some now and then. If you are getting downloads and not tasks you might want to look at your log (under tools) to see what is happening. What’s happening is it’s wasting bandwidth downloading png files and no WUs. Why should I wait around if none of these projects want my computing? No status posted on the main page, so Krembil doesn’t give a s*** either. Plenty of WUs elsewhere. [Edit 1 times, last edit by Just Jake at Mar 5, 2024 10:04:47 PM] |
||
|
|
Unixchick
Veteran Cruncher Joined: Apr 16, 2020 Post Count: 835 Status: Recently Active Project Badges:
|
Just Jake - WCG will probably never have a status page like other projects. It isn't that Krembil doesn't care, it is just a barely funded project. As we donate our cycles we have a choice of where to donate to, so please support the project that speaks to you.
You've been around WCG long enough to know that this project was rescued from IBM throwing it away. WCG is still here, and it may not be all that it was, but it is a miracle that it is still here. The projects here are ones I want to be a part of, but it is frustrating at times. I'm glad that there are many of us that can see past the flaws enough to help out the worthwhile projects here. |
||
|
|
alanb1951
Veteran Cruncher Joined: Jan 20, 2006 Post Count: 858 Status: Offline Project Badges: 
|
Work units dropped to a dribble on 3/1 and nothing after 3/2. Still no work units on 3/5. Reset Project leads to a flash of downloads which never become tasks. Is it me or is it Krembil? Folding@Home is getting a lot of processing recently! I had this happen to one of my systems, starting at almost exactly the time the non-BOINC database stuff seemed to start working again, late on 2023-03-03. I didn't notice for a while (because of when it happened!) but when I did see it I also tried a reset (which made no difference), so I used the WCG device manager to switch the profile away from its normal location (home) to another that was fairly similar in my case (default) and waited until that showed up in the client -- still no work, but when I changed it back to home again I got work. Perhaps there's something about host system profiles that gets cached on a BOINC server and some of that data got damaged when the non-BOINC database server fell off the network and/or when it returned... So if you have a system that's not getting work, try juggling profiles (if that is working at the time!) -- it might just wake the system up again :-) Cheers - Al. P.S. I run small queues and [apart from the incident described above] I have been getting a fairly steady supply of work, subject to the occasional "nothing for an hour or so" times; my queues might shorten but they don't empty and I rarely have so little work that allocated cores are idle... [Edited to fix the typo that amused Adri...] [Edit 2 times, last edit by alanb1951 at Mar 6, 2024 12:01:04 PM] |
||
|
|
|