Checked but no consensus yet


Advanced search

Message boards : Number crunching : Checked but no consensus yet

Sort
Author Message
Profile scsimodo
Volunteer tester

Joined: Sep 13 06
Posts: 13
ID: 99
Credit: 73,635
RAC: 0
Message 484 - Posted 19 Sep 2006 20:08:42 UTC
Last modified: 19 Sep 2006 20:13:03 UTC

Hi All!

My host shows a lot of WUs where two results were delivered but the state says: "checked but no consensus yet". The third result determines if a result is valid or not. Sometimes my host gets 0 Credits , sometimes one of the others.

Computation seemed to run fine but apparently produces different results compared to others in the quorum. It's not just "sometimes", every result shows this effect.

Profile Andre Kerstens
Forum moderator
Project tester
Volunteer tester
Avatar

Joined: Sep 11 06
Posts: 749
ID: 1
Credit: 15,199
RAC: 0
Message 490 - Posted 19 Sep 2006 21:01:48 UTC - in response to Message ID 484 .

Please check this thread It's the same problem. Make sure you are crunching workunits using app version 5.2 (linux) / 5.3 (win)

Andre

Hi All!

My host shows a lot of WUs where two results were delivered but the state says: "checked but no consensus yet". The third result determines if a result is valid or not. Sometimes my host gets 0 Credits , sometimes one of the others.

Computation seemed to run fine but apparently produces different results compared to others in the quorum. It's not just "sometimes", every result shows this effect.


____________
D@H the greatest project in the world... a while from now!
Profile scsimodo
Volunteer tester

Joined: Sep 13 06
Posts: 13
ID: 99
Credit: 73,635
RAC: 0
Message 492 - Posted 19 Sep 2006 21:09:58 UTC - in response to Message ID 490 .

[quote]Please check this thread It's the same problem. Make sure you are crunching workunits using app version 5.2 (linux) / 5.3 (win)
[quote]

Nope, it's a Mac :) To be specific, a Mac Mini Core Duo 1,66GHz.
Crunched the last WU with version 5.1 and stopped for a while. Will give it another try tomorrow and see what happens.

Thanks for the quick response



Dotsch
Volunteer tester
Avatar

Joined: Sep 13 06
Posts: 49
ID: 75
Credit: 57,728
RAC: 0
Message 496 - Posted 19 Sep 2006 22:49:40 UTC - in response to Message ID 492 .

[quote]Please check this thread It's the same problem. Make sure you are crunching workunits using app version 5.2 (linux) / 5.3 (win)
[quote]

Nope, it's a Mac :) To be specific, a Mac Mini Core Duo 1,66GHz.
Crunched the last WU with version 5.1 and stopped for a while. Will give it another try tomorrow and see what happens.

I have the same problem with two other results from Macs.
Profile Andre Kerstens
Forum moderator
Project tester
Volunteer tester
Avatar

Joined: Sep 11 06
Posts: 749
ID: 1
Credit: 15,199
RAC: 0
Message 497 - Posted 19 Sep 2006 23:26:00 UTC - in response to Message ID 496 .

To be or not to be a Mac, that is the question :-)
No, it is the same problem. Unfortunately, we have problem with our intel compiler on Mac and I haven't been able to release 5.2 for Mac yet. Until that happens, these falsely validated results will continue to appear. Keep an eye on the news for the 5.2 mac release.

Thanks
Andre

[quote]Please check this thread It's the same problem. Make sure you are crunching workunits using app version 5.2 (linux) / 5.3 (win)
[quote]

Nope, it's a Mac :) To be specific, a Mac Mini Core Duo 1,66GHz.
Crunched the last WU with version 5.1 and stopped for a while. Will give it another try tomorrow and see what happens.

I have the same problem with two other results from Macs.


____________
D@H the greatest project in the world... a while from now!
Nicolas
Volunteer tester
Avatar

Joined: Sep 13 06
Posts: 66
ID: 17
Credit: 29,510
RAC: 0
Message 505 - Posted 20 Sep 2006 0:06:09 UTC - in response to Message ID 497 .

To be or not to be a Mac, that is the question :-)
No, it is the same problem. Unfortunately, we have problem with our intel compiler on Mac and I haven't been able to release 5.2 for Mac yet. Until that happens, these falsely validated results will continue to appear. Keep an eye on the news for the 5.2 mac release.

Thanks
Andre


By manually editing the database, you can make BOINC not send workunits if there isn't a version equal to or greater than X for the platform. Table 'app', field 'min_version'. This would help in making BOINC not send any work to Macs until you release 5.2 for them.
Profile Andre Kerstens
Forum moderator
Project tester
Volunteer tester
Avatar

Joined: Sep 11 06
Posts: 749
ID: 1
Credit: 15,199
RAC: 0
Message 506 - Posted 20 Sep 2006 0:08:26 UTC - in response to Message ID 505 .

Thanks for the tip Nicolas. I'll look into this!
Andre


By manually editing the database, you can make BOINC not send workunits if there isn't a version equal to or greater than X for the platform. Table 'app', field 'min_version'. This would help in making BOINC not send any work to Macs until you release 5.2 for them.


____________
D@H the greatest project in the world... a while from now!
Nicolas
Volunteer tester
Avatar

Joined: Sep 13 06
Posts: 66
ID: 17
Credit: 29,510
RAC: 0
Message 509 - Posted 20 Sep 2006 0:12:15 UTC - in response to Message ID 506 .

Thanks for the tip Nicolas. I'll look into this!
Andre

By the way, try to keep application versions consistent among platforms. Example, you have version 5.01. You fix something on all platforms and release an app on all of them as 5.02. Then you find a Windows-specific problem and release 5.03 for Windows. If then you make an all-platforms release, it should be 5.04 for everybody, skipping .03 for Linux and mac.

Otherwise after releasing many fixes to a single platforms, version numbers will be way off :)
Profile Andre Kerstens
Forum moderator
Project tester
Volunteer tester
Avatar

Joined: Sep 11 06
Posts: 749
ID: 1
Credit: 15,199
RAC: 0
Message 514 - Posted 20 Sep 2006 3:55:53 UTC - in response to Message ID 509 .

We just discussed this in our group meeting this afternoon :-) We will start doing this soon, because else things get to confusing even sooner.

Thanks
Andre

Thanks for the tip Nicolas. I'll look into this!
Andre

By the way, try to keep application versions consistent among platforms. Example, you have version 5.01. You fix something on all platforms and release an app on all of them as 5.02. Then you find a Windows-specific problem and release 5.03 for Windows. If then you make an all-platforms release, it should be 5.04 for everybody, skipping .03 for Linux and mac.

Otherwise after releasing many fixes to a single platforms, version numbers will be way off :)


____________
D@H the greatest project in the world... a while from now!
Dotsch
Volunteer tester
Avatar

Joined: Sep 13 06
Posts: 49
ID: 75
Credit: 57,728
RAC: 0
Message 545 - Posted 21 Sep 2006 20:03:42 UTC - in response to Message ID 497 .

No, it is the same problem. Unfortunately, we have problem with our intel compiler on Mac and I haven't been able to release 5.2 for Mac yet. Until that happens, these falsely validated results will continue to appear. Keep an eye on the news for the 5.2 mac release.

Should we disable new work with our Macs and wait for 5.2, or should we continue cruching ?

Profile scsimodo
Volunteer tester

Joined: Sep 13 06
Posts: 13
ID: 99
Credit: 73,635
RAC: 0
Message 546 - Posted 21 Sep 2006 20:17:13 UTC - in response to Message ID 545 .
Last modified: 21 Sep 2006 20:18:03 UTC


Should we disable new work with our Macs and wait for 5.2, or should we continue cruching ?




Hmm. Andre's comment should be clear enough IMHO

To be or not to be a Mac, that is the question :-)
No, it is the same problem. Unfortunately, we have problem with our intel compiler on Mac and I haven't been able to release 5.2 for Mac yet. Until that happens, these falsely validated results will continue to appear. Keep an eye on the news for the 5.2 mac release.


Stop getting new work until 5.2 arrives. I did!

Dotsch
Volunteer tester
Avatar

Joined: Sep 13 06
Posts: 49
ID: 75
Credit: 57,728
RAC: 0
Message 549 - Posted 21 Sep 2006 21:15:52 UTC - in response to Message ID 546 .


Hmm. Andre's comment should be clear enough IMHO

Sorry, I am a non native english speaker, so his comment was not 100% clear for me.
Profile scsimodo
Volunteer tester

Joined: Sep 13 06
Posts: 13
ID: 99
Credit: 73,635
RAC: 0
Message 550 - Posted 21 Sep 2006 21:26:48 UTC - in response to Message ID 549 .
Last modified: 21 Sep 2006 21:31:53 UTC


Sorry, I am a non native english speaker...


Me too, but I thought it was clear! Sorry for that... :shame:
Stop your Mac until 5.2 is released, and then we Maccies rule! :)


[EDIT] Du sprichst deutsch? Stopp Deinen Mac bis die Version 5.2 raus ist. Dann kann es weitergehen. Vorher anscheinend nicht, aber dann geht die Post ab hier :)[/EDIT]


Profile scsimodo
Volunteer tester

Joined: Sep 13 06
Posts: 13
ID: 99
Credit: 73,635
RAC: 0
Message 551 - Posted 21 Sep 2006 21:30:37 UTC - in response to Message ID 550 .
Last modified: 21 Sep 2006 21:31:13 UTC

Sry, double Post

zombie67 [MM]
Volunteer tester
Avatar

Joined: Sep 18 06
Posts: 207
ID: 114
Credit: 2,817,648
RAC: 0
Message 654 - Posted 28 Sep 2006 19:48:23 UTC - in response to Message ID 497 .

To be or not to be a Mac, that is the question :-)
No, it is the same problem. Unfortunately, we have problem with our intel compiler on Mac and I haven't been able to release 5.2 for Mac yet. Until that happens, these falsely validated results will continue to appear. Keep an eye on the news for the 5.2 mac release.


Any news on this?
____________
Dublin, CA
Team SETI.USA
Profile Andre Kerstens
Forum moderator
Project tester
Volunteer tester
Avatar

Joined: Sep 11 06
Posts: 749
ID: 1
Credit: 15,199
RAC: 0
Message 667 - Posted 29 Sep 2006 2:56:38 UTC - in response to Message ID 654 .

Nope. Not yet. The problem only occurs if result are erroneous and that doesn't seem to be the case for Macs. So in principal you should be able to continue running 5.1 on Intel Macs without a problem (as long as your results are good and validated). The PPC mac version hasn't come out yet, because of a problem with the G95 compiler we use (much harder one to solve than the intel problem).

Thanks
Andre

To be or not to be a Mac, that is the question :-)
No, it is the same problem. Unfortunately, we have problem with our intel compiler on Mac and I haven't been able to release 5.2 for Mac yet. Until that happens, these falsely validated results will continue to appear. Keep an eye on the news for the 5.2 mac release.


Any news on this?


____________
D@H the greatest project in the world... a while from now!

Message boards : Number crunching : Checked but no consensus yet

Database Error
: The MySQL server is running with the --read-only option so it cannot execute this statement
array(3) {
  [0]=>
  array(7) {
    ["file"]=>
    string(47) "/boinc/projects/docking/html_v2/inc/db_conn.inc"
    ["line"]=>
    int(97)
    ["function"]=>
    string(8) "do_query"
    ["class"]=>
    string(6) "DbConn"
    ["object"]=>
    object(DbConn)#21 (2) {
      ["db_conn"]=>
      resource(78) of type (mysql link persistent)
      ["db_name"]=>
      string(7) "docking"
    }
    ["type"]=>
    string(2) "->"
    ["args"]=>
    array(1) {
      [0]=>
      &string(50) "update DBNAME.thread set views=views+1 where id=48"
    }
  }
  [1]=>
  array(7) {
    ["file"]=>
    string(48) "/boinc/projects/docking/html_v2/inc/forum_db.inc"
    ["line"]=>
    int(60)
    ["function"]=>
    string(6) "update"
    ["class"]=>
    string(6) "DbConn"
    ["object"]=>
    object(DbConn)#21 (2) {
      ["db_conn"]=>
      resource(78) of type (mysql link persistent)
      ["db_name"]=>
      string(7) "docking"
    }
    ["type"]=>
    string(2) "->"
    ["args"]=>
    array(3) {
      [0]=>
      object(BoincThread)#3 (16) {
        ["id"]=>
        string(2) "48"
        ["forum"]=>
        string(1) "2"
        ["owner"]=>
        string(2) "99"
        ["status"]=>
        string(1) "0"
        ["title"]=>
        string(28) "Checked but no consensus yet"
        ["timestamp"]=>
        string(10) "1159498598"
        ["views"]=>
        string(4) "1498"
        ["replies"]=>
        string(2) "15"
        ["activity"]=>
        string(23) "4.5826916577952995e-130"
        ["sufferers"]=>
        string(1) "0"
        ["score"]=>
        string(1) "0"
        ["votes"]=>
        string(1) "0"
        ["create_time"]=>
        string(10) "1158696522"
        ["hidden"]=>
        string(1) "0"
        ["sticky"]=>
        string(1) "0"
        ["locked"]=>
        string(1) "0"
      }
      [1]=>
      &string(6) "thread"
      [2]=>
      &string(13) "views=views+1"
    }
  }
  [2]=>
  array(7) {
    ["file"]=>
    string(63) "/boinc/projects/docking/html_v2/user/community/forum/thread.php"
    ["line"]=>
    int(184)
    ["function"]=>
    string(6) "update"
    ["class"]=>
    string(11) "BoincThread"
    ["object"]=>
    object(BoincThread)#3 (16) {
      ["id"]=>
      string(2) "48"
      ["forum"]=>
      string(1) "2"
      ["owner"]=>
      string(2) "99"
      ["status"]=>
      string(1) "0"
      ["title"]=>
      string(28) "Checked but no consensus yet"
      ["timestamp"]=>
      string(10) "1159498598"
      ["views"]=>
      string(4) "1498"
      ["replies"]=>
      string(2) "15"
      ["activity"]=>
      string(23) "4.5826916577952995e-130"
      ["sufferers"]=>
      string(1) "0"
      ["score"]=>
      string(1) "0"
      ["votes"]=>
      string(1) "0"
      ["create_time"]=>
      string(10) "1158696522"
      ["hidden"]=>
      string(1) "0"
      ["sticky"]=>
      string(1) "0"
      ["locked"]=>
      string(1) "0"
    }
    ["type"]=>
    string(2) "->"
    ["args"]=>
    array(1) {
      [0]=>
      &string(13) "views=views+1"
    }
  }
}
query: update docking.thread set views=views+1 where id=48