OSX test WUs are available
log in

Advanced search

Message boards : News : OSX test WUs are available

Author Message
Profile rebirther
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 2 Jan 13
Posts: 3222
Credit: 22,879,153
RAC: 9
Message 156 - Posted: 12 Dec 2014, 18:34:18 UTC

Pls attach some MACs as soon as possible for testing both apps 32/64bit.

-Reb

Profile Steve Hawker*
Avatar
Send message
Joined: 8 Dec 14
Posts: 26
Credit: 610,307
RAC: 0
Message 157 - Posted: 12 Dec 2014, 18:44:11 UTC - in response to Message 156.

I just get this:

Fri Dec 12 10:41:25 2014 | SRBase | Scheduler request failed: HTTP file not found

I tried resetting and re-attaching, no joy.

[64bit OSX]

Profile rebirther
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 2 Jan 13
Posts: 3222
Credit: 22,879,153
RAC: 9
Message 161 - Posted: 12 Dec 2014, 19:17:59 UTC - in response to Message 157.

I just get this:

Fri Dec 12 10:41:25 2014 | SRBase | Scheduler request failed: HTTP file not found

I tried resetting and re-attaching, no joy.

[64bit OSX]


The second scheduler connection should get an access.

I have updated both OSX now.

Profile Steve Hawker*
Avatar
Send message
Joined: 8 Dec 14
Posts: 26
Credit: 610,307
RAC: 0
Message 162 - Posted: 12 Dec 2014, 19:46:49 UTC - in response to Message 161.

Sorry...

Fri Dec 12 11:46:25 2014 | SRBase | Scheduler request failed: HTTP file not found

Do I need to do reset the project?

Profile rebirther
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 2 Jan 13
Posts: 3222
Credit: 22,879,153
RAC: 9
Message 163 - Posted: 12 Dec 2014, 20:09:30 UTC - in response to Message 162.

Sorry...

Fri Dec 12 11:46:25 2014 | SRBase | Scheduler request failed: HTTP file not found

Do I need to do reset the project?


No.

Profile rebirther
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 2 Jan 13
Posts: 3222
Credit: 22,879,153
RAC: 9
Message 171 - Posted: 13 Dec 2014, 7:44:15 UTC

The residues for OSX, win32/64 and linux32/64 matched in all results, thats a good sign.

zombie67 [MM]
Avatar
Send message
Joined: 4 Dec 14
Posts: 13
Credit: 24,226,718
RAC: 1,764
Message 182 - Posted: 13 Dec 2014, 21:21:59 UTC
Last modified: 13 Dec 2014, 21:23:10 UTC

I am running three OS X machines, all errors so far. I will suspend them for now. Let me know when to try again.

Here is a sample:

http://srbase.myfirewall.org/sr5/result.php?resultid=72452

Stderr output

<core_client_version>7.4.33</core_client_version>
<![CDATA[
<message>
exceeded elapsed time limit 240.08 (825480.00G/3438.33G)
</message>
<stderr_txt>
12:43:51 (39815): wrapper (7.5.26012): starting
12:43:51 (39815): wrapper: running ./llr64mac ( -d -oPgenInputFile=input.prp -oPgenOutputFile=primes.txt -oDiskWriteTime=10 -oOutputIterations=10000 -oResultsFileIterations=99999999)

</stderr_txt>
]]>

____________
Dublin, California
Team: SETI.USA

Profile [AF>Le_Pommier] Aillas
Send message
Joined: 4 Dec 14
Posts: 10
Credit: 9,478,689
RAC: 5,720
Message 183 - Posted: 13 Dec 2014, 21:52:55 UTC

Everything ok for me (64 bit)
____________


Badge Score: 8*3 + 1*4 = 28

zombie67 [MM]
Avatar
Send message
Joined: 4 Dec 14
Posts: 13
Credit: 24,226,718
RAC: 1,764
Message 184 - Posted: 13 Dec 2014, 22:11:46 UTC - in response to Message 183.

Everything ok for me (64 bit)


That is because your tasks are finishing in under 240 seconds. It appears that the tasks self-abort after hitting 240 seconds of run time.

exceeded elapsed time limit 240.08 (825480.00G/3438.33G)

____________
Dublin, California
Team: SETI.USA

Profile rebirther
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 2 Jan 13
Posts: 3222
Credit: 22,879,153
RAC: 9
Message 185 - Posted: 13 Dec 2014, 22:22:00 UTC - in response to Message 184.
Last modified: 13 Dec 2014, 22:26:18 UTC

Everything ok for me (64 bit)


That is because your tasks are finishing in under 240 seconds. It appears that the tasks self-abort after hitting 240 seconds of run time.

exceeded elapsed time limit 240.08 (825480.00G/3438.33G)


Yes its a bit mysterious, the settings in template are:

<rsc_fpops_bound>8.2548e14</rsc_fpops_bound> <rsc_fpops_est>8.2548e10</rsc_fpops_est>


This should be enough time to finish a WU. Perhaps the server software is ignoring the settings somewhow or has something todo with credit new in background? I need more time to investigate whats wrong, the app should be ok.

If you can calculated your values with
<rsc_fpops_bound> / <flops> * <duration_correction_factor>


this could be help me a bit.

zombie67 [MM]
Avatar
Send message
Joined: 4 Dec 14
Posts: 13
Credit: 24,226,718
RAC: 1,764
Message 186 - Posted: 13 Dec 2014, 23:01:13 UTC - in response to Message 185.
Last modified: 13 Dec 2014, 23:03:49 UTC

Sure. Just tell me where to get the values?

3129 12/13/2014 3:00:10 PM Running CPU benchmarks 3130 12/13/2014 3:00:11 PM Suspending computation - CPU benchmarks in progress 3131 12/13/2014 3:00:42 PM Benchmark results: 3132 12/13/2014 3:00:42 PM Number of CPUs: 8 3133 12/13/2014 3:00:42 PM 5278 floating point MIPS (Whetstone) per CPU 3134 12/13/2014 3:00:42 PM 13598 integer MIPS (Dhrystone) per CPU

____________
Dublin, California
Team: SETI.USA

Profile rebirther
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 2 Jan 13
Posts: 3222
Credit: 22,879,153
RAC: 9
Message 187 - Posted: 13 Dec 2014, 23:06:55 UTC - in response to Message 186.

Sure. Just tell me where to get the values?

3129 12/13/2014 3:00:10 PM Running CPU benchmarks 3130 12/13/2014 3:00:11 PM Suspending computation - CPU benchmarks in progress 3131 12/13/2014 3:00:42 PM Benchmark results: 3132 12/13/2014 3:00:42 PM Number of CPUs: 8 3133 12/13/2014 3:00:42 PM 5278 floating point MIPS (Whetstone) per CPU 3134 12/13/2014 3:00:42 PM 13598 integer MIPS (Dhrystone) per CPU


FLOPS = floating points + integer.

Can you try to edit your client_state.xml and change the 8.2548e14 to a much higher level e30?

I think the limit was not enough. Just for testing.

zombie67 [MM]
Avatar
Send message
Joined: 4 Dec 14
Posts: 13
Credit: 24,226,718
RAC: 1,764
Message 188 - Posted: 13 Dec 2014, 23:21:59 UTC - in response to Message 187.


Can you try to edit your client_state.xml and change the <rsc_fpops_bound>8.2548e14</rsc_fpops_bound> to a much higher level e30?

I think the limit was not enough. Just for testing.


I could not find "rsc_fpops_bound" in the client_state.xml for any srbase tasks. In fact, I see no sorbs tasks at all. I see all the tasks for all the other projects that are downloaded to my machine. Andy I see their rsc_fpops_bound values. Very odd!
____________
Dublin, California
Team: SETI.USA

Profile Steve Hawker*
Avatar
Send message
Joined: 8 Dec 14
Posts: 26
Credit: 610,307
RAC: 0
Message 189 - Posted: 13 Dec 2014, 23:23:03 UTC - in response to Message 163.

Sorry...

Fri Dec 12 11:46:25 2014 | SRBase | Scheduler request failed: HTTP file not found

Do I need to do reset the project?


No.


I just tried to get work and still got the HTTP file not found

So I reattached and now I'm getting work. W00t!!

Profile rebirther
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 2 Jan 13
Posts: 3222
Credit: 22,879,153
RAC: 9
Message 190 - Posted: 13 Dec 2014, 23:25:41 UTC - in response to Message 188.


Can you try to edit your client_state.xml and change the 8.2548e14 to a much higher level e30?

I think the limit was not enough. Just for testing.


I could not find "rsc_fpops_bound" in the client_state.xml for any srbase tasks. In fact, I see no sorbs tasks at all. I see all the tasks for all the other projects that are downloaded to my machine. Andy I see their rsc_fpops_bound values. Very odd!


In the next batch I will raise the limits. This could help.

zombie67 [MM]
Avatar
Send message
Joined: 4 Dec 14
Posts: 13
Credit: 24,226,718
RAC: 1,764
Message 191 - Posted: 13 Dec 2014, 23:27:11 UTC

Please let us know when that happens. I will put my machines on NNT until the so I don't just burn up tasks.
____________
Dublin, California
Team: SETI.USA

Profile Steve Hawker*
Avatar
Send message
Joined: 8 Dec 14
Posts: 26
Credit: 610,307
RAC: 0
Message 192 - Posted: 14 Dec 2014, 0:28:03 UTC - in response to Message 190.
Last modified: 14 Dec 2014, 0:30:08 UTC


Can you try to edit your client_state.xml and change the <rsc_fpops_bound>8.2548e14</rsc_fpops_bound> to a much higher level e30?

I think the limit was not enough. Just for testing.


I could not find "rsc_fpops_bound" in the client_state.xml for any srbase tasks. In fact, I see no sorbs tasks at all. I see all the tasks for all the other projects that are downloaded to my machine. Andy I see their rsc_fpops_bound values. Very odd!


In the next batch I will raise the limits. This could help.


I found rsc_fpops_bound and edited it to 8254800000000000000000000.000000 {I added 10 zeroes}

I unsuspended a WU and it finished without any problem. So it did help to edit this parameter

Here's the task: http://srbase.myfirewall.org/sr5/result.php?resultid=81082

You'll probably already know this but the tasks have an estimated duration of 0:00 and go to 100% immediately. I guess some other parameters need adjustment too.


Post to thread

Message boards : News : OSX test WUs are available


Main page · Your account · Message boards


Copyright © 2014-2017 BOINC Confederation / rebirther