Posts by Robert Nelson

1) Message boards : SZTAKI Desktop Grid : Permanent Upload Error (Message 8182)
Posted 2153 days ago by Robert Nelson
Please update your server software to disable certificates, this is now a problem for version 7 clients. On some projects not disabling this function causes the work units to fail validation. While that is not the case here, the project is sending out error messages on all uploads. Please see message from David Anderson on the BOINC Alpha forum.

These project need to disable file upload certificate checking
in order to work with 7.0 clients.
Specifically, they need to add
<ignore_upload_certificates/>
<dont_generate_upload_certificates/>
to their project config file.
I posted these instructions to the boinc_projects email list
a long time ago.

Please try to contact the admins of these projects
(email, message boards) etc. to convey this to them.

THanks -- David

Thanks Bob
2) Message boards : Észrevételek, tapasztalatok : Long WUs (Message 2023)
Posted 4289 days ago by Robert Nelson
Similar problem here, the system got adjusted for work units that ran in less than 30 minutes, some cases as few as 10. So now I have received an absolute overload of work units which say will run in less than 30 minutes but take 10 to 12 hours. There is no way that they all can be completed by the deadline.


Simple just abort the ones that won\'t finish and leave BOINC to the rest it will learn again that the WU\'s are long

2 weeks or so and an optimised version will be ready for the 12th dimenson so that should help



Well just did the hatefull and aborted about 100 work units. It is too bad that the venue problem got fixed before the new work units came out. If the bug had been still with us the number of work units sent out would not have been that great when work was requested , giving the software time to adjust estimate for work unit duration before the load hit, thus limiting the flood. With Boinc thinking that the work units would take 10 to 30 minutes depending on machine, they got the equivalent, unfortunetly the base system was 600 to 720 minutes not the 10 or 30. Those 100 work units would have been easily handled before. Did note that the estimated time after the damage had been done was adjusted by the client so they were not showing the 10 or 30 minutes but indeed the 10 to 12 hours. Maybe in the future when a new lot of work units is scheduled to come out, especially if the last batch had short durations, a temporary limitation on work unit distribution until the clients have adjusted to the work length time may be in order.
3) Message boards : Észrevételek, tapasztalatok : Long WUs (Message 1970)
Posted 4290 days ago by Robert Nelson
Similar problem here, the system got adjusted for work units that ran in less than 30 minutes, some cases as few as 10. So now I have received an absolute overload of work units which say will run in less than 30 minutes but take 10 to 12 hours. There is no way that they all can be completed by the deadline.
4) Message boards : Észrevételek, tapasztalatok : VENUE PRoblem (Message 1671)
Posted 4306 days ago by Robert Nelson
Thanks Adam;

It seems to be working, saw correct venue come up on several hosts along with a much larger supply of work, so far no apparent anomalies with the fix.

Bob
5) Message boards : Észrevételek, tapasztalatok : Daily quota problem (Message 1483)
Posted 4340 days ago by Robert Nelson
Try to detach and then attach to project again.

I didn\'t think that this solution would work as the problem most definetly is at the server, but I tried it anyway. As you can see it was unsuccessful, but in fact shows that the database has a problem, as the second time I attached (manually) it changed my venue as well, also stating that I had now fullfilled a quota now of only 2 vs. the previous 1000, when no workunits had been supplied at all for this machine. Like JM7 has pointed out previously, this is very similar to a problem that LHC had after it upgraded, it is too similar to be ignored. Would suggest that Adam touch base with Chrulle at LHC. See following report:

1/6/2006 4:59:10 PM|SZTAKI Desktop Grid|Resetting project
1/6/2006 4:59:10 PM||request_reschedule_cpus: exit_tasks
1/6/2006 4:59:10 PM|SZTAKI Desktop Grid|Detaching from project
1/6/2006 4:59:10 PM||request_reschedule_cpus: project op
1/6/2006 4:59:59 PM||Fetching config info from http://szdg.lpds.sztaki.hu/szdg/get_project_config.php
1/6/2006 5:00:01 PM||Missing account key
1/6/2006 5:00:24 PM|http://szdg.lpds.sztaki.hu/szdg/|Master file download succeeded
1/6/2006 5:00:24 PM|http://szdg.lpds.sztaki.hu/szdg/|Sending scheduler request to http://szdg.lpds.sztaki.hu/szdg/cgi-bin/scheduler
1/6/2006 5:00:24 PM|http://szdg.lpds.sztaki.hu/szdg/|Reason: Requested by user
1/6/2006 5:00:24 PM|http://szdg.lpds.sztaki.hu/szdg/|Requesting 86400 seconds of new work
1/6/2006 5:00:25 PM|http://szdg.lpds.sztaki.hu/szdg/|Scheduler request to http://szdg.lpds.sztaki.hu/szdg/cgi-bin/scheduler succeeded
1/6/2006 5:00:25 PM|SZTAKI Desktop Grid|Message from server: No work sent
1/6/2006 5:00:25 PM|SZTAKI Desktop Grid|Message from server: (there was work but it was committed to other platforms)
1/6/2006 5:00:25 PM|SZTAKI Desktop Grid|New host venue: home
1/6/2006 5:00:25 PM|SZTAKI Desktop Grid|Successfully attached to SZTAKI Desktop Grid
1/6/2006 5:02:34 PM||request_reschedule_cpus: project op
1/6/2006 5:02:36 PM|SZTAKI Desktop Grid|Sending scheduler request to http://szdg.lpds.sztaki.hu/szdg/cgi-bin/scheduler
1/6/2006 5:02:36 PM|SZTAKI Desktop Grid|Reason: Requested by user
1/6/2006 5:02:36 PM|SZTAKI Desktop Grid|Requesting 86400 seconds of new work
1/6/2006 5:02:41 PM|SZTAKI Desktop Grid|Scheduler request to http://szdg.lpds.sztaki.hu/szdg/cgi-bin/scheduler succeeded
1/6/2006 5:02:41 PM|SZTAKI Desktop Grid|Message from server: No work sent
1/6/2006 5:02:41 PM|SZTAKI Desktop Grid|Message from server: (reached daily quota of 2 results)
1/6/2006 5:02:41 PM|SZTAKI Desktop Grid|New host venue: 0

It also could be associated with the no work issue (first refusal message)and mixing up error code but the host venue issue would indicate that that is not the case.
6) Message boards : Észrevételek, tapasztalatok : Daily quota problem (Message 1474)
Posted 4340 days ago by Robert Nelson
I have a couple of machines that reach the daily quota of 500 without downloading a single result.

If the root cause of the problem has been fixed, it may be nessecary to fix the DB so that every host can at least download some work - so that the quota starts to climb again.


I have 3 machines that are at least downloading a couple of units before hitting the limit, but I have one that hasn\'t been able to download for days and everytime it tries the response is like the following:

1/5/2006 7:19:38 PM|SZTAKI Desktop Grid|Sending scheduler request to http://szdg.lpds.sztaki.hu/szdg/cgi-bin/scheduler
1/5/2006 7:19:38 PM|SZTAKI Desktop Grid|Reason: To fetch work
1/5/2006 7:19:38 PM|SZTAKI Desktop Grid|Requesting 86400 seconds of new work
1/5/2006 7:19:43 PM|SZTAKI Desktop Grid|Scheduler request to http://szdg.lpds.sztaki.hu/szdg/cgi-bin/scheduler succeeded
1/5/2006 7:19:43 PM|SZTAKI Desktop Grid|Message from server: No work sent
1/5/2006 7:19:43 PM|SZTAKI Desktop Grid|Message from server: (reached daily quota of 1000 results)
1/5/2006 7:19:43 PM|SZTAKI Desktop Grid|No work from project

Something needs a kick somewhere as the scheduler obviously either can\'t count or has been given the wrong set of data.

7) Message boards : SZTAKI Desktop Grid : Ad@m's BLOG (to live with a grid in your pocket...) (Message 1150)
Posted 4375 days ago by Robert Nelson
Thanks Adam;

That is great news, look forward to that first unit after the switch.
8) Message boards : SZTAKI Desktop Grid : Version 5.x of Server Software (Message 1058)
Posted 4394 days ago by Robert Nelson
Adam;

Thanks for the update. We understand the issues associated with translations, they can be very difficult at times. Don't rush, it is best to get it right the first time, it limits the complaints later. Like waiting for a fine wine, we can wait. A couple of extra weeks will not have much of an effect a year from now. Thats why BOINC allows multiple projects to run, when SZTAKI is at version 5 my systems should notice fairly quickly and we will be crunching again.
9) Message boards : Észrevételek, tapasztalatok : Requesting added compatibility. (Message 947)
Posted 4430 days ago by Robert Nelson
I just downloaded a SZTAKI WU with BOINC v5.1.1 installed, so it seems we will still be able to run the Project with the new version. I haven't finished a WU with it yet though to see what happens when I turn it in ...<I>


Thats interesting because it would appear that Sztaki is only accepting version 4 see this example from one of my version 5 boxes.
10/7/2005 6:00:39 PM|SZTAKI Desktop Grid|Sending scheduler request to http://szdg.lpds.sztaki.hu/szdg/cgi-bin/scheduler
10/7/2005 6:00:39 PM|SZTAKI Desktop Grid|Reason: To fetch work
10/7/2005 6:00:39 PM|SZTAKI Desktop Grid|Requesting 172800 seconds of new work
10/7/2005 6:00:44 PM|SZTAKI Desktop Grid|Scheduler request to http://szdg.lpds.sztaki.hu/szdg/cgi-bin/scheduler succeeded
10/7/2005 6:00:45 PM|SZTAKI Desktop Grid|Message from server: Need major version 4 of the BOINC core client. You have 5.

I know that Adam had some reservations on upgrading. Was wondering if things look a little cleaner for him based on the other projects that have upgraded with apparently, from the perspective of being on the outside looking in, few problems. Rosetta was the most recent to switch over. Thanks for any update.



Home | My Account | Message Boards


Copyright © 2017 SZTAKI Desktop Grid