| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Support
Forum: BOINC Agent Support
Thread: WCG/Boinc won't call home [ Resolved ] |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 11
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
For the past four days I've had to manually send units for my main machine. It crunches, completes and updates the units, but it will not automatically send or pull new units. I've tried everything I can think of to fix this. Please provide any ideas on how I can find/fix this problem. I'm now at my wit's end. I have tried the following...
------------------------------------------This system is a win xpPro-sp3-32bit...i7-875k, hyperthreaded --It shows the localhost as connected and version number. --Preferences are set at days to connect 0.0000 --Also tried 0.0400 and 0.0200 days --Pc time is correct --re-started the pc --Tried repair/install of ver 6.2.28 --Uninstalled 6.2.28 and tried old ver 5.10.45 --re-installed ver 6.2.28 --Even restored pc to a 10 day earlier restore point. --The router cannot be a factor because my other machines are working correctly and all pc settings are the same. --AVG anti-virus does not indicate any blockage. --Made an exception for WCG in the windows firewall, although it's not required for my other machines. --Win security is standard at mediim-high and privacy is medium. --I made an exception for DEP, although it's set to only run for windows essential programs. --Reset projects under projects tab --Upgraded to ver 6.10.58 Please help. [Edit 1 times, last edit by Former Member at Aug 28, 2010 9:38:20 PM] |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
What is the selection in the Activity menu for your Networking?
----------------------------------------In 6.2.28 the client automatically opens the network connection for 5 minutes when hitting update even when the client connection in the Activity menu is set to Network Suspended.
WCG
Please help to make the Forums an enjoyable experience for All! |
||
|
|
nasher
Veteran Cruncher USA Joined: Dec 2, 2005 Post Count: 1423 Status: Offline Project Badges: 
|
how many work units do you have cached?
----------------------------------------i have found if your time to complete work units is greater than the cache time then BOINC will not communicate untill the 24 hour timer clicks in and it will... is this the problem you are having and are just jumping the gun perchance?  |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
? Translation plz. You're saying if the TTC sum of total work cached is greater than cache time as in e.g. cached 6 days and the cache setting 3 days? Yes, the client is not in a rush to clear the RtR under that condition, but he writes that new work is then pulled hitting update.
----------------------------------------
WCG
----------------------------------------Please help to make the Forums an enjoyable experience for All! [Edit 1 times, last edit by Sekerob at Aug 27, 2010 11:25:59 PM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Sekerob
-All 6 of my machines have always been set for "network activity based on preferances", but just today I changed this particular problem machine to "Network activity always availiable" -As mentioned, I can always send completed work and get new work by manually clicking update in the projects section. nasher -I run all machines on a 5 day cache. I did a clean and through uninstall using Revo-uninstaller. When I re-installed, I cut the cache to one day. This was to prevent turning in so many dead units if I had further problems. The cache remains at 1 day. -I don't exactly know if I'm jumping the gun? I may not understand? For over a month the system was turning in completed units and fetching new work normally. Then suddenly 4 days ago, the units started backing up as "ready to report" but would not turn in. -The hypertherthreaded system runs eight simultaneous units taking 2 hours each to complete. So even at a 1 day (24 hr) cache does that mean 2/24 = 12x cache? ...or 8 units x 2 hrs =16 hrs = 16/24 = 1.5x cache? -I have waited each day until an hour or two before the daily cut-off time to manually send my units. This means that I have allowed 22-24 hours for proper communications, that never ocurred. Thank you both for your help and hopefully more ideas to come. This is my main machine and I really need it running right. I am truly frustrated. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Ray422, this is also happening to me on 2 machines. Both have the new boinc 6.10.58, but one has the x86 version (on WinXP 32b) and the other the x64 version (on Win7 64b).
The former seems to have fixed itself, because it's back to normal after a week or so in the condition you describe. I did a reset of WCG on the second one, to no avail. Tonight it had a bunch of WUs Ready to Report and was crunching the last two. One more hour and it would have been idle (that happened on the other one last week). My theory is that maybe (and this is a shot in the dark) some or all of the following factors have played a role here: 1) The BOINC version (although other machines with 6.10.58 have no problems whatsoever). 2) The combination of projects selected, in my case DDDT2 and C4CW, both with several periods (long and short) of unavailable jobs, both with jobs that have very different running times. 3) Some babysitting, messing with several options in the client and on the website (due to the scarcity of WUs, especially DDDT2). So, you're not alone on this. I hope someone has a clue on what to do. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hi guys.
If it's any help to you i have my 2 rigs set the same both use different Boinc's ones 6.2.14 the other 6.10.17 these are my settings. Computer is connected to the Internet about every (Leave blank or 0 if always connected. BOINC will try to maintain at least this much work.) 0.1 days Maintain enough work for an additional : 1 days Ones a Intel quad the other an AMD x6 both fetch and report o.k. |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Ray422,
----------------------------------------will you please let us know the outcome of "Network Always Available", ultimately and ideally resisting to hit the update button for at least 24 hours. With completed work the client will seek to connect or start shouting something along the line of "need network connection". My settings on one client 6.10.58, W7-32: - Acticity menu > Always Connected - In Options, Connections tab, no default specified, in fact expressly, default cleared so it wont try to get to the disabled modem. ADSL connections are not listed. - In Options, General Tab, the Network Reminder Interval slider set to zero - Modem and NIC disabled in hardware, only WIFI is active. - On client, In Preferences, Network Tab: a) Confirm before connecting to internet NOT ticked b) Disconnect when done ticked [but that's only working when running off a modem] c) Network usage allowance: Everyday from 00:00 to 00:00 which works as 24/7 connected. Suggest to retype and save these scheduled time prefs. And that's all there is needed to never having to hit update and go into set-forget mode.
WCG
----------------------------------------Please help to make the Forums an enjoyable experience for All! [Edit 1 times, last edit by Sekerob at Aug 28, 2010 11:25:31 AM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I checked my settings against your list and they are almost the same. At least the applicable ones. Thank you greatly for your help.
Suddenly and Miraculously after the change to 6.10.58, it has started working. I do not know if the version change/reinstall is why? The network change from "preferenced" to "always" >>might<< have had something to do with it? I may have changed the cache backwards from what I should have? I cut the cache to 0.5 days. My logic was that it would have to send/receive more often. Shortly after the network and cache changes, it pulled 3 new units. About two hours later it turned in completions and also pulled more new units. It is now sending and receiving about every two hours. I will gradually ease the cache back up to 5 days. I do not know which of the two changes ( network or version ) worked, but once again...thanks for your help :) |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Great, and changing cache has all sorts of ''unexpected'' effects. BOINC needs weeks to balance out and can throw fits on things such as changing the number of hours computed in a day.
----------------------------------------Happy Crunching. PS: When you're definitely ok that networking is again working as desired, then please edit your Opening Post and insert [RESOLVED] in the title... then it is ready to sink and good for those seeking a working solution for this issue.
WCG
Please help to make the Forums an enjoyable experience for All! |
||
|
|
|