| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Official Messages
Forum: News
Thread: 2023-07-31 Update (MCM1 issue resolved) |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 74
|
|
| Author |
|
|
Grumpy Swede
Master Cruncher Svíþjóð Joined: Apr 10, 2020 Post Count: 2210 Status: Recently Active Project Badges: 
|
And just as the team went home for the weekend, it seems as if my suspicions were right. No new work available, at least not for MCM (which is the only work I want, now that OPN1/OPNG isn't available).
----------------------------------------A couple of MCM resends is coming in, and the usual weekend message "Tasks are committed to other platforms" is back. So, I guess that "MCM1 issue resolved" isn't really correct. [Edit 2 times, last edit by Grumpy Swede at Aug 11, 2023 9:50:06 PM] |
||
|
|
TPCBF
Master Cruncher USA Joined: Jan 2, 2011 Post Count: 1957 Status: Offline Project Badges: 
|
"...On my part, this is the only way to reach someone at WCG, as their support email isn't working, since end of last year now!..." 1) I haven't needed to contact them since last month, but my last reply from WCG Support was July 7th, so email *is* working. Prior to that I have replies from them in May and June of this year as well. And there are several times that people have been advised to post for example those "missing devices" IDs in the forum instead of using the support email too. 2) I have a very high opinion of TigerLily and believe she is truly helping. Dr Jurisica summed it up well in the paragraphs above/below this thread concerning his folks' efforts. Sorry, but I just can't share your enthusiasm. She is even less present than Cyclops was, posts are all over the place instead of trying to keep things on topic and in a logical place, and the responses on the last two incidents about "260 Mac sucking up all the WUs" and "DHCP client failure brought down all the servers", sorry, such stories just don't look even remotely like helping... @Ralf, if you find yourself being moderated, perhaps that might be a reflection point to think about. I find your comments & tone often to be rather caustic to be point of being unbecoming of someone of your intellect. Perhaps if you'd tone it down and not interject such negativity, your posts would have much more impact. Speaking from 45-years of I/S background and working with Teams, I think you have the potential to be an excellent leader, but the sarcasm sinks the boat. Sorry, I have always spoken my mind and I always will. I am not whitewashing anything, and I simply don't appreciate being slapped in the face (virtually) by all the non-info and outright false information being fed for the last 15 months. And the 47 years that I am in IT now has given me a lot of experience in dealing with both, humans and technology. And for both, it is in my experience important to stick to the facts and be honest when it comes to problems. For IT issues, that is the only way you can solve things instead of just making things worse in the long run. With humans, well, more and more seem to fit the description by a Jack Nicholson quote.Everyone on the forum doesn't *have* to publicly know your moderation issues. Well, complain to the person who has after alleged "review" released those posts verbatim...I have a firm belief that we will look back on these early days in coming years and realize just what great effort Jurisica Labs did on a shoestring budget. Well, do you also realize that in a couple of weeks, it will be two (2) years that the move "to Krembil" (which according to Dr.Jurisica isn't involved) was announced, but hat it is only 4 months since it has been mentioned in what dire situation the project is supposed to be? All the while ignoring any offers, directly or indirect, from volunteers to help? Again, a lack of honesty and communication...Ralf  |
||
|
|
TPCBF
Master Cruncher USA Joined: Jan 2, 2011 Post Count: 1957 Status: Offline Project Badges:
|
And just as the team went home for the weekend, it seems as if my suspicions were right. No new work available, at least not for MCM (which is the only work I want, now that OPN1/OPNG isn't available). Well, can't say anything about MCM, as I don't have this currently selected, but it also seems indeed that SCC WUs are running out as well. And it is now already 5:35pm in Toronto and everyone is most likely already gone.A couple of MCM resends is coming in, and the usual weekend message "Tasks are committed to other platforms" is back. But hey, who am I to complain?... Ralf |
||
|
|
alanb1951
Veteran Cruncher Joined: Jan 20, 2006 Post Count: 983 Status: Recently Active Project Badges: 
|
Grumpy Swede,
----------------------------------------And just as the team went home for the weekend, it seems as if my suspicions were right. No new work available, at least not for MCM (which is the only work I want, now that OPN1/OPNG isn't available). Standard BOINC configurations give precedence to retries (or other tasks with lower work-unit numbers), so if there are lots of retries at a given point in time that will act to suppress new work unit release.A couple of MCM resends is coming in, and the usual weekend message "Tasks are committed to other platforms" is back. So, I guess that "MCM1 issue resolved" isn't really correct. Given the number of retries I get for "No Reply" or "Not Started by Deadline" tasks during any given week, and the distribution thereof[1], I rather suspect that what we are seeing is:
I'm not sure how easy it would be to ensure that some new work gets fed along with retries; it might need multiple feeders with different parameters?[2] As long as the total tasks allowed out is large enough the retries should clear at a reasonable pace, so let's see whether we run out over the weekend or not. If we do, then you're right and itr isn't fixed; if we don't then we need to try to find an acceptable way to reduce the number of retries caused by overdue tasks :-) Cheers - Al. [1] Monday or Tuesday to Thursday or Friday most of my MCM1 tasks are _0 or _1 and more than 50% of them validate the same day; the rest of the time most of my tasks are retries, a small proportion of which are due to other tasks failing (download errors, SIGSEGV errors &c) but most of which are for missed deadlines... [2] Perhaps someone who actually runs a BOINC server might tell us; my knowledge is based on source-code dives and reading the documentation... [Edit 1 times, last edit by alanb1951 at Aug 11, 2023 11:29:07 PM] |
||
|
|
TPCBF
Master Cruncher USA Joined: Jan 2, 2011 Post Count: 1957 Status: Offline Project Badges:
|
And, as you have noted, users will see that "other platforms" message. It is not a coincidence that it tends to appear most often every 6 or 7 days; if the cycle I described above started on a Wednesday we could call it "the usual midweek message"... :-) I am not sure that this is this kind of cycle, I can't say that I have ever seen back in the IBM days, and in the "current" form, this just seems to be a more recent event. And as Grumpy mentioned, these kind of behavior have have in recent months almost always resulted in an empty feeder of a weekend.For what it is worth (though my previous post once again just didn't make it through in time), it seems at least for SCC1 WUs, the situation seems to have a bit relaxed, at least two hosts I have been working with in the last hour since getting back into the office have had their number of WUs increased again quite a bit, will have to check when the next small batch is ready to report by how many they were being replenished. Since early this morning it was mostly one or none, with a couple of exceptions... Ralf |
||
|
|
Aperture_Science_Innovators
Advanced Cruncher United States Joined: Jul 6, 2009 Post Count: 139 Status: Offline Project Badges: 
|
And just as the team went home for the weekend, it seems as if my suspicions were right. No new work available, at least not for MCM (which is the only work I want, now that OPN1/OPNG isn't available). A couple of MCM resends is coming in, and the usual weekend message "Tasks are committed to other platforms" is back. So, I guess that "MCM1 issue resolved" isn't really correct. Can't comment on MCM, since I'm chasing a SCC badge, but it seems to be doing just fine: 8/11/2023 8:38:34 PM | World Community Grid | Sending scheduler request: To report completed tasks. 8/11/2023 8:38:34 PM | World Community Grid | Reporting 55 completed tasks 8/11/2023 8:38:34 PM | World Community Grid | Requesting new tasks for CPU 8/11/2023 8:38:37 PM | World Community Grid | Scheduler request completed: got 44 new tasks That's UTC-4 time, so just pulled a fat stack of WUs ten minutes ago. No shortage of tasks for the other devices, at least. EDIT: And, another device, just to show the trend: 8/11/2023 8:54:43 PM | World Community Grid | Sending scheduler request: To fetch work. 8/11/2023 8:54:43 PM | World Community Grid | Reporting 7 completed tasks 8/11/2023 8:54:43 PM | World Community Grid | Requesting new tasks for CPU 8/11/2023 8:54:44 PM | World Community Grid | Started upload of SCC1_0004198_MyoD1-C_14277_0_r1860732764_0 8/11/2023 8:54:45 PM | World Community Grid | Finished upload of SCC1_0004198_MyoD1-C_14277_0_r1860732764_0 8/11/2023 8:54:48 PM | World Community Grid | Scheduler request completed: got 72 new tasks  [Edit 1 times, last edit by Aperture_Science_Innovators at Aug 12, 2023 12:56:17 AM] |
||
|
|
Hans Sveen
Veteran Cruncher Norge Joined: Feb 18, 2008 Post Count: 831 Status: Offline Project Badges:
|
Just got some MCM1 and SCC1 units, fresh ones,, no resends!
|
||
|
|
as1981
Cruncher Joined: Dec 3, 2006 Post Count: 49 Status: Offline Project Badges: 
|
Thank you, as always, for keeping WCG going. A WCG that faces intermittent technical difficulties is far better than no WCG. Thanks for detailing some of the challenges you've faced. I agree. The current situation is definetly better than no WCG. |
||
|
|
gb009761
Master Cruncher Scotland Joined: Apr 6, 2005 Post Count: 2986 Status: Offline Project Badges: 
|
Thank you, as always, for keeping WCG going. A WCG that faces intermittent technical difficulties is far better than no WCG. Thanks for detailing some of the challenges you've faced. I agree. The current situation is definetly better than no WCG. I'll certainly 'second' that motion - we've just got to learn to live with the knowledge that, whereas we were 'spoilt' when it was run/maintained by IBM (considering their resources/financial clout), better to have an 'intermittent WCG' than 'no WCG at all'..  |
||
|
|
Sgt.Joe
Ace Cruncher USA Joined: Jul 4, 2006 Post Count: 7697 Status: Offline Project Badges:
|
Thank you, as always, for keeping WCG going. A WCG that faces intermittent technical difficulties is far better than no WCG. Thanks for detailing some of the challenges you've faced. I agree. The current situation is definetly better than no WCG. I'll certainly 'second' that motion - we've just got to learn to live with the knowledge that, whereas we were 'spoilt' when it was run/maintained by IBM (considering their resources/financial clout), better to have an 'intermittent WCG' than 'no WCG at all'.. +1 Cheers
Sgt. Joe
*Minnesota Crunchers* |
||
|
|
|