| 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: 14
|
|
| Author |
|
|
Felix Kaeufer
Cruncher Joined: Feb 3, 2012 Post Count: 29 Status: Offline Project Badges:
|
Since Sunday the scheduler isn't able to connect to WCG.
Example: Di 30 Jun 05:53:15 2015 | World Community Grid | Sending scheduler request: To report completed tasks. The mentioned task isn't visible in the task list on WCG anymore. Should I detach and reattach the project? |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Restart your internet modem and computer first before doing a Detach-Add Project cycle.
The "Not requesting tasks: don't need (job cache full)" suggests your computer has more work than it can process in the time it's on, so propose to reduce your minimum work buffer [connect every nn days on the website profile] |
||
|
|
Felix Kaeufer
Cruncher Joined: Feb 3, 2012 Post Count: 29 Status: Offline Project Badges:
|
Thank you, SekeRob. I haven't changed the buffer values for more than a year and their values were "0". I raised the maximum work buffer to 5 days and the communication worked again.
I forgot to add the really interesting lines: Di 30 Jun 05:53:17 2015 | World Community Grid | Scheduler request failed: Unrecognized or bad HTTP Content or Transfer-Encoding However, everything seems to be working again. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Most odd as assigning of work and reporting goes to the same Scheduler process. Anyway, something to keep in mind if this happens with the same situation, increasing buffer to force a work fetch and then successfully connecting hints at a reproducable bug.
----------------------------------------Would though strongly recommend you set the work buffer down again. A task to report and nolonger on the result pages means it took too long to complete... [Edit 1 times, last edit by Former Member at Jun 30, 2015 3:48:08 PM] |
||
|
|
Thyme Lawn
Cruncher Joined: Dec 9, 2008 Post Count: 46 Status: Offline Project Badges:
|
Di 30 Jun 05:53:17 2015 | World Community Grid | Scheduler request failed: Unrecognized or bad HTTP Content or Transfer-Encoding However, everything seems to be working again. I've been getting the same encoding error on my i5 W7 system (running BOINC 6.12.43) since 0916 UTC today. The task has been repeatedly reporting a task and requesting new work since then. My result status page shows that the reported task has been validated and each request generates a new scheduler reply file containing: <message priority="low">Resent lost task MCM1_0014982_4216_0</message> Unfortunately the encoding error reported by BOINC is preventing the reply from being processed, so I enabled HTTP debug which generated the following output (reduced to the relevant lines): 30-Jun-2015 10:16:32 [World Community Grid] Sending scheduler request: To fetch work. I've reset my router and rebooted the system and still have the same problem. None of my other projects are having problems with their scheduler requests.
"The ultimate test of a moral society is the kind of world that it leaves to its children." - Dietrich Bonhoeffer
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I see these every so often usually during the stats runs. They usually resolve themselves.
----------------------------------------1536 World Community Grid 6/30/2015 6:58:49 PM Sending scheduler request: Requested by user. 1537 World Community Grid 6/30/2015 6:58:49 PM Reporting 1 completed tasks 1538 World Community Grid 6/30/2015 6:58:49 PM Not requesting tasks: don't need (job cache full) 1539 6/30/2015 6:58:53 PM Project communication failed: attempting access to reference site 1540 World Community Grid 6/30/2015 6:58:53 PM Scheduler request failed: Unrecognized or bad HTTP Content or Transfer-Encoding 1541 6/30/2015 6:58:56 PM Internet access OK - project servers may be temporarily down. 1542 World Community Grid 6/30/2015 6:59:33 PM update requested by user 1543 World Community Grid 6/30/2015 6:59:33 PM Sending scheduler request: Requested by user. 1544 World Community Grid 6/30/2015 6:59:33 PM Reporting 1 completed tasks 1545 World Community Grid 6/30/2015 6:59:33 PM Not requesting tasks: don't need (job cache full) 1546 6/30/2015 6:59:35 PM Project communication failed: attempting access to reference site 1547 World Community Grid 6/30/2015 6:59:35 PM Scheduler request failed: Unrecognized or bad HTTP Content or Transfer-Encoding 1548 6/30/2015 6:59:37 PM Internet access OK - project servers may be temporarily down. 1549 World Community Grid 6/30/2015 7:07:42 PM update requested by user 1550 World Community Grid 6/30/2015 7:07:47 PM Sending scheduler request: Requested by user. 1551 World Community Grid 6/30/2015 7:07:47 PM Reporting 1 completed tasks 1552 World Community Grid 6/30/2015 7:07:47 PM Not requesting tasks: don't need (job cache full) 1553 6/30/2015 7:07:50 PM Project communication failed: attempting access to reference site 1554 World Community Grid 6/30/2015 7:07:50 PM Scheduler request failed: Unrecognized or bad HTTP Content or Transfer-Encoding 1555 6/30/2015 7:07:52 PM Internet access OK - project servers may be temporarily down. Fixed itself: 1563 World Community Grid 6/30/2015 7:14:52 PM update requested by user 1564 World Community Grid 6/30/2015 7:14:54 PM Sending scheduler request: Requested by user. 1565 World Community Grid 6/30/2015 7:14:54 PM Reporting 1 completed tasks 1566 World Community Grid 6/30/2015 7:14:54 PM Not requesting tasks: don't need (job cache full) 1567 6/30/2015 7:14:55 PM Project communication failed: attempting access to reference site 1568 World Community Grid 6/30/2015 7:14:55 PM Scheduler request failed: Unrecognized or bad HTTP Content or Transfer-Encoding 1569 6/30/2015 7:14:57 PM Internet access OK - project servers may be temporarily down. 1570 World Community Grid 6/30/2015 7:33:51 PM Computation for task ugm1_ugm1_13825_0103_1 finished 1571 World Community Grid 6/30/2015 7:33:51 PM Starting task MCM1_0014949_1085_1 1572 World Community Grid 6/30/2015 7:33:53 PM Started upload of ugm1_ugm1_13825_0103_1_0 1573 World Community Grid 6/30/2015 7:33:59 PM Finished upload of ugm1_ugm1_13825_0103_1_0 1574 World Community Grid 6/30/2015 7:45:15 PM update requested by user 1575 World Community Grid 6/30/2015 7:45:19 PM Sending scheduler request: Requested by user. 1576 World Community Grid 6/30/2015 7:45:19 PM Reporting 2 completed tasks 1577 World Community Grid 6/30/2015 7:45:19 PM Not requesting tasks: don't need (job cache full) 1578 World Community Grid 6/30/2015 7:45:22 PM Scheduler request completed [Edit 2 times, last edit by Doneske at Jul 1, 2015 1:53:46 AM] |
||
|
|
Felix Kaeufer
Cruncher Joined: Feb 3, 2012 Post Count: 29 Status: Offline Project Badges:
|
Would though strongly recommend you set the work buffer down again. Done.A task to report and nolonger on the result pages means it took too long to complete... As far as I can remember it had its deadline in July, so it wasn't too late. My computer is slow but not that slow. I guess the WCG got the "Reporting completed task" signal but the answer didn't make it back to the scheduler. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Yes, that I gathered from the subsequent post by Thyme... So you will have received credit. The way exchanges works is, that the Ready to Report step is sent, the server confirms back the report is received, so the client knows it can clear the report from the list. If that report does not come back, it will try again. Same goes for the result files... sent, confirm back. Never see them, but one way or the other these situations seem to clear themselves.
Interesting "during stats runs"... the nightly [00:06UTC], or midday [12:06 UTC]? |
||
|
|
Thyme Lawn
Cruncher Joined: Dec 9, 2008 Post Count: 46 Status: Offline Project Badges:
|
After 24 hours of failing scheduler requests I resorted to manually merging the new task into client_state.xml and leaving the completed task to be reported again. Lo and behold the next scheduler request succeeded:
----------------------------------------01-Jul-2015 07:54:49 [World Community Grid] Sending scheduler request: To fetch work. I'll post again when I know the outcome of the next work request.
"The ultimate test of a moral society is the kind of world that it leaves to its children." - Dietrich Bonhoeffer
----------------------------------------[Edit 1 times, last edit by Thyme Lawn at Jul 1, 2015 11:10:57 AM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Boy Wonder ;>) [The edit actions not something the regular volunteer is readily able to do... I'd have bluntly hit the Abort for the Tasks view line, but vaguely remember some of these are there till kingdom come, or detach/add]
|
||
|
|
|