| 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) |
| Member(s) browsing this thread: Paul Schlaffer , RoKKor |
|
Thread Status: Active Total posts in this thread: 615
|
|
| Author |
|
|
alanb1951
Veteran Cruncher Joined: Jan 20, 2006 Post Count: 1327 Status: Offline Project Badges: 
|
Grumpy Swede -- thanks for drawing attention to the Global summary, as I only ever check the application-specific summaries and there's no real evidence of a push for catch-up there
 If you look at the Global numbers alongside the MCM1 numbers for the same days, you'll notice that there are some significant differences on days such as 2025-12-04, 2025-12-01, 2025-11-28, 2025-11-13 and 2025-10-31 (Global higher than MCM1 except on 2025-11-28). And, checking back somewhat further, there are some much smaller differences (all increases) to some of the 2025-08 [pre-migration!] values. An interesting possible insight into how the delayed statistics are being fed into the non-BOINC database? It will be interesting to see what happens to the stats export once they've finished! Cheers - Al. |
||
|
|
Unixchick
Veteran Cruncher Joined: Apr 16, 2020 Post Count: 1307 Status: Offline Project Badges: 
|
Thank you Hans for mentioning the problem with the first post image. I've removed it for now. I'll contact the person who does that img for us, and see if they can fix it.
|
||
|
|
Hans Sveen
Veteran Cruncher Norge Joined: Feb 18, 2008 Post Count: 1009 Status: Offline Project Badges:
|
Now it is back and working!
Thank You, UnixChick |
||
|
|
Unixchick
Veteran Cruncher Joined: Apr 16, 2020 Post Count: 1307 Status: Offline Project Badges:
|
|
||
|
|
alanb1951
Veteran Cruncher Joined: Jan 20, 2006 Post Count: 1327 Status: Offline Project Badges:
|
Would someone else please test the forum search function(s) to confirm my earlier report? -- I still can't get a search to find any posts from before the recent restart (except the first post of this thread, which is visible because Unixchick has edited it since then...)
----------------------------------------I know the search functions aren't super user-friendly, but they can be helpful -- far nicer than sifting through many pages of old threads [that might not have had related titles] to try to find old conversations  Cheers - Al. [Edit 1 times, last edit by alanb1951 at Dec 10, 2025 11:14:03 PM] |
||
|
|
bfmorse
Senior Cruncher US Joined: Jul 26, 2009 Post Count: 448 Status: Offline Project Badges:
|
Did a search for "WCG" for all time and oldest results were from Sep 18, 2025 3:18:43 PM and newer.
----------------------------------------then there was a gap until Dec 8, 2025 9:08:01 PM Edit to add; Did a search for posts by me "bfmorse" and the earliest was Dec 8, 2025 9:48:02 PM by bfmorse [Edit 2 times, last edit by bfmorse at Dec 11, 2025 12:42:57 AM] |
||
|
|
bfmorse
Senior Cruncher US Joined: Jul 26, 2009 Post Count: 448 Status: Offline Project Badges:
|
I remember seeing somewhere that another user was concerned about a laptop overheating.
Of the two laptops that I use for processing WU's one of them needs to have its non-functioning CPU fan replaced. Yes, I have procured the replacement fan but not yet had the time to install it. I was able to locate and purchase "ICE COOREL Aluminum Laptop Cooling Pads" through Amazon.com for the purpose of keeping those laptops cool for about $23.00 USD each. They are powered through one of the computer's USB ports. It works! And to answer your concern: I receive no commission on any products sold by Amazon.com |
||
|
|
Grumpy Swede
Master Cruncher Svíþjóð Joined: Apr 10, 2020 Post Count: 2533 Status: Offline Project Badges: 
|
The entire WCG site is bogged down in cold molasses, and at times even responds with a:
----------------------------------------"503 Service Unavailable No server is available to handle this request." [Edit 1 times, last edit by Grumpy Swede at Dec 11, 2025 2:13:47 AM] |
||
|
|
Grumpy Swede
Master Cruncher Svíþjóð Joined: Apr 10, 2020 Post Count: 2533 Status: Offline Project Badges:
|
@alanb1951
----------------------------------------Yup, the forum search has lost its marbles indeed. Can't find any post before Dec 8, except for Unixchick's post September 18 when she created the latest version of "Project Status (First Post Updated)". But that one is visible only because she edited the post Dec 10, 2025 6:06:05 PM. Time to rebuild the search index I believe. Edit: But the main forum DB is intact, as can be seen if one goes through the "Index" link at the top of the forum pages. I'll alert the WCG team about the problem. Edit 2: And the molasses seems to have been removed from the WCG site now. [Edit 4 times, last edit by Grumpy Swede at Dec 11, 2025 3:09:01 AM] |
||
|
|
alanb1951
Veteran Cruncher Joined: Jan 20, 2006 Post Count: 1327 Status: Offline Project Badges:
|
Regarding "molasses" and help engine indexes...
That sluggish period coincided with my once-a-day bulk completed results statistics collection script run (parallel on two systems) -- the first "molasses" indicator was that a script run that normally fetches off a data block a second was taking 10+ seconds per block without reporting any errors that needed a retry, then came the 503 errors... Interestingly, most of the 503 errors seemed to be clustered around 02:09, 02:14 and 02:19 UTC, with 4 or five minutes of sluggish but warning-free action between times! But at 02:19 one of the scripts got too many retries for the same data block and gave up -- fortunately the other one ran to completion so I could copy the data files and re-run the database part of my failed script!Another script I run (more frequently) to check on tasks waiting for file deletion or database purge had been taking 125 to 150 milliseconds to fetch the [effectively empty] "ready to purge" file, but it started to show signs of sluggishness from about 00:50 UTC (worst case being over 15 seconds for the same empty dataset!), and timings didn't get back to normal until about 02:50 UTC. However, the script never actually failed (503 or otherwise). It would be interesting to know what happened between those times to cause the issues we have seen... On the other topic, thanks to those who checked the forum search function, and special thanks to Grumpy Swede for reporting it to WCG on confirmation of the issue. I hope that there might be a way for the indexing to do a partial recovery if it realizes it has had a problem (the same way that the main databases would do a restore to a known state then replay any outstanding transactions), but that may not actually be feasible. I fear that a full rebuild will take a very long time and hammer whatever server it runs on. Cheers - Al. |
||
|
|
|