| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Support
Forum: Website Support
Thread: Uploads and downloads seem to be blocked |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 3
|
|
| Author |
|
|
robertmiles
Senior Cruncher US Joined: Apr 16, 2008 Post Count: 445 Status: Offline Project Badges: 
|
Uploads and downloads seem to be blocked, so my computers won't be able to do any more tasks until this is fixed.
|
|||
|
||||
|
Col323
Senior Cruncher Joined: Nov 4, 2008 Post Count: 372 Status: Offline Project Badges: 
|
I also woke up to see my machines struggling to transfer work. Some of them have had success after a few rounds of hitting the "Retry Now" button on a hung transfer. But, I'm glad WCG is part of Boinc where I can direct cycles to other projects pretty seamlessly while this sorts out.
----------------------------------------/edit: Here's another thread with a lot of discussion: https://www.worldcommunitygrid.org/forums/wcg/viewthread_thread,39772 [Edit 1 times, last edit by Col323 at Feb 23, 2017 2:29:55 PM] |
||
|
|
Rickjb
Veteran Cruncher Australia Joined: Sep 17, 2006 Post Count: 666 Status: Offline Project Badges: 
|
!!! @WCG SERVER OPERATORS and TECHs
-------------------------------------------------------------------------------------- This is happening again right now ... What was also going on when this happened a few days ago and is also happening now is that a batch or more of very short FAAH Phase 1 (FAHV) that run for only about 2 min (under Linux) are being sent out & results received. Your file transfer servers are not coping. The more usual FAHV WUs run for30-60 min, and have "Rigid" in their name. The short ones do not have "Rigid" in their name. Maybe you could intermix these 2 types of WU in your despatch queue. You might also consider increasing the cap on the no of WUs per CPU core, currently 35, that we can cache. --- Edit: When this happened before, I had machines run out of work and sit idle. When they could get work, the cache was limited to only about 1h due to the short WUs and the WUs-per-core limit. [Edit 1 times, last edit by Rickjb at Feb 28, 2017 5:52:42 AM] |
||
|
|
|