Posts by darkpella

1) Message boards : Questions and problems : HTTP error 0 with World Community Grid and BOINC 6.12.34 (Message 40645)
Posted 12 Oct 2011 by darkpella
Post:
If i downgrade to 6.10.60 then i cannot see any tasks in my system at all. So it doesn't work. And 6.13.1 doesn't fix this issue.


That proves 6.13. works as it should. It's changed your client_state.xml file exactly as it says it will in INCOMPATIBILITY WARNING FOR UPCOMING 6.13 BUILDS thread.

As for the "HTTP error 0 with World Community Grid and BOINC 6.12.34" problem, the solution is posted here on the WCG forums.


Don't know whether reverting to 6.12.33 can be taken as a solution overall, anything starting form 6.12.34 won't simply work with WCG and proxies.

darkpella
2) Message boards : Questions and problems : HTTP error 0 with World Community Grid and BOINC 6.12.34 (Message 40218)
Posted 17 Sep 2011 by darkpella
Post:
Hi,

as reported in this message on WCG forums , my host, with BOINC 6.12.34 installed, can't connect to WCG, every communication attempt resulting in a "HTTP error 0" message.

I tried uninstalling BOINC 6.12.34 and reverting to BOINC 6.10.60 and everything with WCG went back OK.

Then I uninstalled BOINC 6.10.60 and reinstalled 6.12.34 and the issue reappeared.

I also tried with BOINC 6.13.1 (beta) but this didn't fix the issue.

Currently, I reverted once again to 6.10.60, but it looks like the issue is related to BOINC 6.12.34, WCG and (possibly) the way I communicate to the Internet, though I tried with my standard Internet proxy and TOR, none of which gives me problems with other projects.
Also, I don't get any such issue with other BOINC-projects while running BOINC 6.12.34.

I posted this message on WCG forums as well

Bye

darkpella
3) Message boards : Questions and problems : Boinc setting Resource Share not correctly (Message 34489)
Posted 2 Sep 2010 by darkpella
Post:
Hi,

Boinc is setting Resource Share for some projects to 0 (0.00%) through they were set to a different share either through BAM account manager or through the Account Page on the project's web site (for those projects not handled by BAM).

Apart form one project (among those set to 0 RS), CAS@Home, all of the projects set to 0 RS are managed through BAM (primaboinca, VRS, WUProp@home).

I checked on the projects's account pages and the RS there correctly reflects what is set on BAM as RS for those projects.

Also the RS's tho which those projects that are set to 0 in BOINC are set at different values on BAM.

I'm attached to a lot of projects and running BOINC 6.10.58 on a Windows 7 64 bit host w/ 8 GB RAM and a Quadro FX 1800.

Anyone has any idea about what might be wrong and how to fix it?

By the way, i tried setting WCG at a RS of 3000 (previously 1000) both through BAM and through WCG account page but this was not reflected on BOINC (though accepted by both BAM and WCG).

Bye

darkpella
4) Message boards : Questions and problems : BOINC loosing access to proxy (Message 33503)
Posted 23 Jun 2010 by darkpella
Post:
Hi,

BOINC is set on my PC to access the Internet through a BlueCoat proxy.

About once every two days it will no longer be able to connect to the Internet, unless I shut BOINC off (the core client, not just the GUI manager) and start it back again.

This is a major issue with Lattice Project since, w/ their latest GARLI 5.13 WUs, it checkpoints only once in some days, especially towards the completion of the WU, and one host will never be able to complete a GARLI 5.13 WUs if BOINC has to be restarted more often than once per week.

Any idea about what I may do to prevent this?

bye

darkpella
5) Message boards : Questions and problems : HELP: Boinc Manager failing to download .exe files from most of the projects (Message 27773)
Posted 5 Oct 2009 by darkpella
Post:
Hi,

since 2 days BOINC Manager on one of my hosts is failing to download .exe files form most (but not all) of the projects (I'm attached to almost every BOINC-project).

The error message I got is like:
05/10/2009 15.05.08 Cosmology@Home Finished download of camb_2.16_windows_intelx86.exe
05/10/2009 15.05.08 Cosmology@Home [error] Signature verification failed for camb_2.16_windows_intelx86.exe
05/10/2009 15.05.08 Cosmology@Home [error] Checksum or signature error for camb_2.16_windows_intelx86.exe


I tried uninstalling/reinstalling BOINC, deleting the data directory and every reference to BOINC but id didn't work.

What could I try?

Thanks

darkpella
6) Message boards : Questions and problems : BOINC 6.6.20 not reporting WU elapsed running time on remote hosts (Message 24793)
Posted 8 May 2009 by darkpella
Post:
Hi,

I'm running BOINC on two hosts:

  • Host 1 is a domain controller and is running BOINC 5.8.16 as a service.
  • Host 2 is not a domain controller and is running BOINC 6.6.20 as a non-service application.


I use BOINC manager on Host 2 to check the status of BOINC running on host 1 as well.

Since I upgraded Host 2 to BOINC 6.6.20 the "elapsed time" entry of each WU on host 1 is empty.
Elapsed time for WUs on host 2 is shown as well as time "To completion" on both hosts.

Not a real problem, but wanted to let you know.

Bye

darkpella

7) Message boards : Projects : Cels@Home down (Message 23720)
Posted 16 Mar 2009 by darkpella
Post:
Hi folks,

I've been attached to Cels@Hoem project for awhile, but in the last few weeks (since mid February) it has been persistently down.

Does anyone have any news about it?

Bye

darkpella
8) Message boards : BOINC Manager : My Wish List - 2 (Message 12831)
Posted 4 Oct 2007 by darkpella
Post:
Hi guys,

guess soemone already proposed this, but i couldn't succeed in finding any post about it in these "Wish list" threads, so I'll state my wish here..

It would be nice to be able to set a given WU to be kept running until completion form th "TASKS" pane, e.g one selects a TASK and can push a button and get this task to a "run continuously until completion" state, where this task will start running if not running and task switching will be suspended until the WU gets completed, pretty much like the automatically assigned "running, high priority" state that boinc can assign to a given task when it thinks there's a chance it won't complete within deadline.

On multiple CPU systems where more than one task is allowed to run task switching would be suspended foronly one of these "slots".

This would greatly help people taking part in experimental 8alpha or beta) projects along with "stable" ones when it is needed (or desired) to have a WU of one of this projects be completed ASAP (or let run as long as possible in a short time).

Basically only one task should get this state at a given time, hence eneabling it for a given task should reset any other task in "manned forced priority" state to switch back to normal scheduling.

Otherwise multiple tasks might be placed in this "manned priority" state and priorities among them could be assigned (and changed), i.e. the highest priority ine will be run until completion, then then second priority one, then the third ... In this case, on hosts where multiple computing "slots" are available, one should be able to set how many slots can be assigned to "manned priority" tasks.

I guess, anyway, that being able to assign this state to a single task would already be a good deal for all those who would make use of this feature.

Bye

darkpella

Think to have seen this 'wish' before of jobs running uninterrupted to completion. To a certain extend this is already managed by projects and short deadlines. One can further impact this by extending the project switching times. If an alpha runs for 4 hours estimate and the switch is set to 6 hours, it would go in 1 run. The whole time management and debt system would take care if the alpha takes too much.... give it a low weight and it's sure to be put in the corner for a while so other projects can catch up.


Well, I know this can already be done by tweaking BOINC switching parameters, at least to some extent and in a pretty empirical manner, my "wish" would be to be able to do it fast, easy and for good.

Maybe I'm the only one, maybe I'm not, but I do enjoy taking part to most experimental projects with one of my BOINC hosts to help developing, also posting bugs and issues on those projects' boards, hence having a chance to get a WU crunched ASAP when a bugfix is to be tested or an issue is to be checked for repeatbility would be fine for me...

Bye

darkpella
9) Message boards : BOINC Manager : My Wish List - 2 (Message 12827)
Posted 4 Oct 2007 by darkpella
Post:
Hi guys,

guess soemone already proposed this, but i couldn't succeed in finding any post about it in these "Wish list" threads, so I'll state my wish here..

It would be nice to be able to set a given WU to be kept running until completion form th "TASKS" pane, e.g one selects a TASK and can push a button and get this task to a "run continuously until completion" state, where this task will start running if not running and task switching will be suspended until the WU gets completed, pretty much like the automatically assigned "running, high priority" state that boinc can assign to a given task when it thinks there's a chance it won't complete within deadline.

On multiple CPU systems where more than one task is allowed to run task switching would be suspended foronly one of these "slots".

This would greatly help people taking part in experimental 8alpha or beta) projects along with "stable" ones when it is needed (or desired) to have a WU of one of this projects be completed ASAP (or let run as long as possible in a short time).

Basically only one task should get this state at a given time, hence eneabling it for a given task should reset any other task in "manned forced priority" state to switch back to normal scheduling.

Otherwise multiple tasks might be placed in this "manned priority" state and priorities among them could be assigned (and changed), i.e. the highest priority ine will be run until completion, then then second priority one, then the third ... In this case, on hosts where multiple computing "slots" are available, one should be able to set how many slots can be assigned to "manned priority" tasks.

I guess, anyway, that being able to assign this state to a single task would already be a good deal for all those who would make use of this feature.

Bye

darkpella
10) Message boards : BOINC client : Running BOINC as hidden (Message 5982)
Posted 12 Oct 2006 by darkpella
Post:
Hi,
guess there's another post describing this, and that proper directions on how to do this can be found somewhere in core client support as weel, but I really looked for it and couldn't find it... so here's the question:
I'm running Boinc on a WIN2k host which I'm controlling via a remote boinc manager.
I wish to let boinc.exe to start when the host is logged in, but I wish not to run Boinc manager or the boinc.exe console to be shown.
Is there any command line option I could enter to boinc.exe to make this?
Thnx.
darkpella
11) Message boards : BOINC Manager : Can't connect to Predictor@home (Message 5920)
Posted 5 Oct 2006 by darkpella
Post:
Hi,

guess this doesn't belong here, bbut couldn't find a better place for it..

In the last hours BOINC client couldn't connect to P@H site with the following error message:

05/10/2006 19.05.15|Predictor @ Home|Fetching scheduler list
05/10/2006 19.05.38||Project communication failed: attempting access to reference site
05/10/2006 19.05.40||Access to reference site succeeded - project servers may be temporarily down.
05/10/2006 19.05.41|Predictor @ Home|Scheduler list fetch failed: http error
05/10/2006 19.05.41|Predictor @ Home|5 consecutive failures fetching scheduler list - deferring 604800 seconds
05/10/2006 19.05.41|Predictor @ Home|Deferring scheduler requests for 1 weeks, 0 days, 0 hours, 0 minutes and 0 seconds


Neither could I connect to P@H web site through Firefox. Is it the same for everyone else, or do I have a routing problem to P@H? It already happened to me in teh past, but then i couldn't connect to any web site at SCRIPPS, which I can now do (e.g. connecting to www.scripps.edu yields no problem).

Thnx

darkpella
12) Message boards : BOINC Manager : Can't connect to Predictor@home (Message 3766)
Posted 4 Apr 2006 by darkpella
Post:
I've created a thread called "Connecting from Italy" and posted this:


I have a user at the Boinc message board. His username is "darkpella" and here is a link to that thread. He can't attach to PPAH, or even get here with a browser. Can someone help him?

tony


over at ppah, There is a link to this thread.

{edit} this was the first post to their boards in 3 days, so I wouldn't expect an immediate response. LOL


Hope someone will help.
Another big thank to mmxiastro meanwhile.

darkpella
13) Message boards : BOINC Manager : Can't connect to Predictor@home (Message 3762)
Posted 4 Apr 2006 by darkpella
Post:
Thought I had an idea, but I was wrong

Thanks for caring, anyway.

darkpella
14) Message boards : BOINC Manager : Can't connect to Predictor@home (Message 3760)
Posted 4 Apr 2006 by darkpella
Post:
darkpella, your traceroute results look fine to me. They're the same as mine and my BOINC connects to Predictor without problems.

Many sites block ICMP (ping/traceroute) traffic nowadays, so you can't ping or traceroute them, but the site itself is reachable via HTTP/HTTPS (web, used also by BOINC etc).

Do you still have problems attaching to the Predictor@home project?


Hi dhatz,

did the following
1. Clicked on the Project->Attach to project menu on Boinc manager menu bar
2. Clicked Next on the dialog that opened
3. Entered http://predictor.scripps.edu as project URL and pressed Next on the dialog that opened
4. Waited while it tried connecting
5. Got a "Project temporary unavailable" message.
6. Opened Firefox
7. Entered http://www.scripps.edu as URL
8. Got "Connection timed out" error
9. Entered http://predictor.scripps.edu as URL
10. Got "Connection timed out" error.

HEEEEEEEEEEEEEEELLLLLLLLLPPPPPPPPP!!!!!!!

darkpella
15) Message boards : BOINC Manager : Can't connect to Predictor@home (Message 3749)
Posted 3 Apr 2006 by darkpella
Post:
Hi,
I tried attaching to predictor@home, but it looks like I can't connect to any server within the scripps.edu domain. I tried using visualroute from any public servers on the net (Italy, where I live, The US and others) but looks like San Diego Supercomputer Center is blcoking all traffic to www.scripps.edu, regardless where it comes from.
This is happening since months, so I'm wonderingf who's taking part in Predictor@home and how they could do it (and how could I possibly...)


Can you copy here the results of "TRACERT predictor.scripps.edu" command ? (or "traceroute predictor.scripps.edu" under Linux)

Does predictor.scripps.edu have an IP of 137.131.252.96 ?

Because I suspect the problem is due to "DNS poisoning", as I've noticed the same issue in the past.


Hi dhatz,

looks like the IP is right; the tracert reports are below. Please notice that the IP (wahoo.sdsc.edu) it can't go past is the same even tracert-ing from a VisualRoute server and from other online tracert servers.

Thanks for assisting.

darkpella

TRACERT log from my PC:
C:\\Documents and Settings\\Riccardo>tracert predictor.scripps.edu

Rilevazione instradamento verso predictor1.scripps.edu [137.131.252.96]
su un massimo di 30 punti di passaggio:

1 1 ms <10 ms <10 ms 192.168.7.254
2 192 ms 85 ms 57 ms 37.253.124.1
3 21 ms 22 ms 21 ms 10.3.7.194
4 21 ms 21 ms 22 ms 10.254.1.189
5 23 ms 23 ms 22 ms 10.254.0.102
6 27 ms 47 ms 26 ms 10.254.1.69
7 * * * Richiesta scaduta.
8 * * * Richiesta scaduta.
9 * * * Richiesta scaduta.
10 835 ms 661 ms 1281 ms 83-103-100-222.ip.fastwebnet.it [83.103.100.222]

11 1984 ms 1992 ms 2013 ms gi-6-3.rom20.ip.tiscali.net [213.200.68.213]
12 2215 ms 2197 ms 711 ms so-2-0-0.sjc10.ip.tiscali.net [213.200.81.153]
13 779 ms 798 ms 794 ms ge-6-8.car4.SanJose1.Level3.net [4.68.111.165]
14 818 ms 810 ms 808 ms ae-31-51.ebr1.SanJose1.Level3.net [4.68.123.30]

15 849 ms 862 ms 882 ms ae-2.ebr1.LosAngeles1.Level3.net [4.69.132.10]
16 876 ms 857 ms 864 ms ge-0-0-0-55.gar1.LosAngeles1.Level3.net [4.68.10
2.144]
17 906 ms 896 ms 581 ms CENIC.gar1.Level3.net [64.156.191.10]
18 451 ms 470 ms 349 ms inet-sdsc-sdsc2--lax-isp-ge.cenic.net [137.164.2
4.206]
19 397 ms 363 ms 300 ms wahoo.sdsc.edu [198.17.46.12]
20 * * * Richiesta scaduta.
21 * * * Richiesta scaduta.
22 * * * Richiesta scaduta.
23 * * * Richiesta scaduta.
24 * * * Richiesta scaduta.
25 * * * Richiesta scaduta.
26 * * * Richiesta scaduta.
27 * * * Richiesta scaduta.
28 * * * Richiesta scaduta.
29 * * * Richiesta scaduta.
30 * * * Richiesta scaduta.

Rilevazione completata.

TRACERT log from www.subnetonline.com
Input Query: predictor.scripps.edu
Started at: 09:14:28
Stopped at: 09:18:19

1 hosted.by.i-s-e.nl (83.172.128.67) 0.073 ms 0.061 ms 0.040 ms
2 ge-1-3-0-1.cr1.rb2f.nl.proserve.nl (80.84.254.1) 0.397 ms 0.347 ms 0.484 ms
3 gsr12416.ams.he.net (195.69.145.150) 2.984 ms 1.887 ms 1.233 ms
4 pos0-0.gsr12416.lon.he.net (216.66.24.157) 9.488 ms 8.752 ms 8.857 ms
5 pos4-1.gsr12416.nyc.he.net (216.218.200.101) 78.827 ms 78.927 ms 78.915 ms
6 pos7-0.gsr12012.sjc.he.net (216.218.254.153) 155.372 ms 155.319 ms 155.389 ms
7 pos11-3.gsr12416.pao.he.net (216.218.214.245) 160.618 ms 160.575 ms 160.748 ms
8 paix-px1--hurricane-ge.cenic.net (198.32.251.69) 157.358 ms 155.313 ms 155.819 ms
9 dc-oak-dc2--oak-dc1-p2p-2.cenic.net (137.164.22.195) 155.529 ms 155.600 ms 157.134 ms
10 dc-csac-dc1--oak-dc2-ge.cenic.net (137.164.22.111) 163.596 ms 163.749 ms 163.600 ms
11 dc-lax-dc1--sac-dc1-pos.cenic.net (137.164.22.126) 163.229 ms 163.045 ms 163.236 ms
12 hpr-lax-hpr--dc-lax-dc2-ge--2.cenic.net (137.164.22.21) 163.779 ms 163.531 ms 163.594 ms
13 dc-tus-dc1--lax-dc2-pos.cenic.net (137.164.22.43) 156.625 ms 156.260 ms 156.240 ms
14 dc-sdsc-sdsc2--tus-dc1-ge.cenic.net (137.164.24.174) 167.101 ms 166.722 ms 166.990 ms
15 wahoo.sdsc.edu (198.17.46.12) 167.742 ms 167.205 ms 167.381 ms
16 * * *
17 * * *
18 * * *
19 * * *
20 * * *
21 * * *
22 * * *
23 * * *
24 * * *
25 * * *
26 * * *
27 * * *
28 * * *
29 * * *
30 * * *

TRACERT log from centralops.net
Tracing route to predictor.scripps.edu [137.131.252.96]...
hop rtt rtt rtt ip address domain name
1 0 0 0 63.134.215.1
2 0 0 0 216.197.122.81 dr-ded1-ge-1-6.crystaltech.com
3 0 0 0 216.197.122.65 cr-edge1-ge-5-0.crystaltech.com
4 0 0 0 67.131.24.221 phv-edge-01.inet.qwest.net
5 1 1 1 205.171.129.85 tmp-core-01.inet.qwest.net
6 11 11 12 67.14.10.6 bur-core-01.inet.qwest.net
7 12 12 11 205.171.213.106 lap-brdr-01.inet.qwest.net
8 11 11 12 64.200.249.165
9 22 22 22 64.200.200.50 lsanca3lcx1-ceic-14-0.wcg.net
10 14 13 14 137.164.24.206 inet-sdsc-sdsc2--lax-isp-ge.cenic.net
11 14 14 14 198.17.46.12 wahoo.sdsc.edu
12 * * *
13 * * *
14 * * *
15 * * *

Trace aborted

-- end --

TRACERT log from cache.online.ee
traceroute to predictor1.scripps.edu (137.131.252.96): 2-20 hops, 38 byte packets
2 194.106.111.17 kjj-igw5.elion.ee 1.89 ms (ttl=254)
3 194.126.115.195 kjj-bb4-ae-3-0.ee.estpak.ee 1.48 ms (ttl=252!)
4 194.126.97.253 kjj-bb2-ae-0-0.ee.estpak.ee 1.53 ms (ttl=251!)
5 194.126.97.206 ldn-bb1-so-1-0-0-0.uk.estpak.ee 44.4 ms (ttl=250!)
6 195.66.226.21 lon.he.net 80.9 ms (ttl=250)
7 216.218.200.101 pos8-0.gsr12416.nyc.he.net 115 ms (ttl=249)
8 216.218.254.153 pos7-0.gsr12012.sjc.he.net 191 ms (ttl=248)
9 216.218.214.245 pos11-3.gsr12416.pao.he.net 197 ms (ttl=248!)
10 198.32.251.69 paix-px1--hurricane-ge.cenic.net 304 ms (ttl=235!)
11 137.164.22.193 dc-oak-dc2--oakk-dc1-p2p-1.cenic.net 303 ms (ttl=234!)
12 137.164.22.111 dc-csac-dc1--oak-dc2-ge.cenic.net 231 ms (ttl=234!)
13 137.164.22.126 dc-lax-dc1--sac-dc1-pos.cenic.net 207 ms (ttl=235!)
14 137.164.22.21 hpr-lax-hpr--dc-lax-dc2-ge--2.cenic.net 207 ms (ttl=235!)
15 137.164.22.43 dc-tus-dc1--lax-dc2-pos.cenic.net 207 ms (ttl=231!)
16 137.164.24.174 dc-sdsc-sdsc2--tus-dc1-ge.cenic.net 328 ms (ttl=234!)
17 198.17.46.12 wahoo.sdsc.edu 211 ms (ttl=234!)
18 *
19 *
20 *



16) Message boards : BOINC Manager : Can't connect to Predictor@home (Message 3721)
Posted 30 Mar 2006 by darkpella
Post:
boinc@scripps.edu

Thanks Tony
17) Message boards : BOINC Manager : Can't connect to Predictor@home (Message 3718)
Posted 30 Mar 2006 by darkpella
Post:
I just tried www.scripps.edu and got through. The PAH website URL(for attachment) is http://predictor.scripps.edu, and I also can get through with that address.

I don't know what to say. except to try the URL I'm providing.

tony


Tried again, both with boinc manager, Firfox, ping and tracert, but the connection times out.

Could be that scrpps is somehow restrcited to US resident IPs due to federal regulations (I live in Italy)?

darkpella


I'm not currently attached to Predictor, but in the past I had no problems and I'm in South Korea.

Well, could anyoneplease provide me with an e-mail address of predictor's hos administrator, if it can be found on predictor's web page, or with the adiministrator of scripps web page so that i may contact her/him and ask?

Thanks

darkpella
18) Message boards : BOINC Manager : Can't connect to Predictor@home (Message 3715)
Posted 30 Mar 2006 by darkpella
Post:
I just tried www.scripps.edu and got through. The PAH website URL(for attachment) is http://predictor.scripps.edu, and I also can get through with that address.

I don't know what to say. except to try the URL I'm providing.

tony


Tried again, both with boinc manager, Firfox, ping and tracert, but the connection times out.

Could be that scrpps is somehow restrcited to US resident IPs due to federal regulations (I live in Italy)?

darkpella
19) Message boards : BOINC Manager : Can't connect to Predictor@home (Message 3709)
Posted 30 Mar 2006 by darkpella
Post:
Hi,
I tried attaching to predictor@home, but it looks like I can't connect to any server within the scripps.edu domain. I tried using visualroute from any public servers on the net (Italy, where I live, The US and others) but looks like San Diego Supercomputer Center is blcoking all traffic to www.scripps.edu, regardless where it comes from.
This is happening since months, so I'm wonderingf who's taking part in Predictor@home and how they could do it (and how could I possibly...)
.

Bye

darkpella
20) Message boards : BOINC Manager : Not getting any new workunit (Message 3614)
Posted 24 Mar 2006 by darkpella
Post:
Hi,
since some days i'm not getting any new workunit from any project I'm attached to when the nes my PC is working on are finished.

I tried rebooting but without any improvement. I also tried upadting the projects from the project page, the requests to the scheduler succeed but o new workunit is downloaded.

I got boinc 5.2.13 on win2k and I'm attached to the following projects:
BBC Climate Change Experiment
rosetta@home
boincsimap
climateprediction.net
Einstein@home
LHC@home
QMC@home
SETI@home
SZTAKI Desktop Grid
cpdn seasonal

Any idea about what the problem is?

Thanks
darkpella




Copyright © 2024 University of California.
Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.