climateprediction.net home page
New work Discussion

New work Discussion

Message boards : Number crunching : New work Discussion
Message board moderation

To post messages, you must log in.

Previous · 1 . . . 72 · 73 · 74 · 75 · 76 · 77 · 78 . . . 91 · Next

AuthorMessage
Jean-David Beyer

Send message
Joined: 5 Aug 04
Posts: 1054
Credit: 16,510,175
RAC: 433
Message 65052 - Posted: 4 Feb 2022, 19:53:34 UTC - in response to Message 65051.  

A note from Dave in the getting started area has the answer, these have been amended to be Mac only so we’re out of luck.


Is that not interesting? The only work unit that worked for me in the current batch on my Xeon processor running Linux had failed twice before on machines running on i686-apple-darwin machines
	Computer	Sent	Time reported
or deadline
explain	Status	Run time
(sec)	CPU time
(sec)	Credit	Application
22191699 	1511241 	29 Jan 2022, 20:48:05 UTC 	1 Feb 2022, 13:43:03 UTC 	Completed 	211,754.62 	210,243.20 	0.00 	UK Met Office HadCM3 short v8.36
i686-pc-linux-gnu
22182726 	1477031 	11 Jan 2022, 16:26:17 UTC 	21 Jan 2022, 0:14:22 UTC 	Error while computing 	19,321.89 	12,747.10 	--- 	UK Met Office HadCM3 short v8.36
i686-apple-darwin
22190607 	1525919 	21 Jan 2022, 0:15:59 UTC 	29 Jan 2022, 20:46:50 UTC 	Error while computing 	88,182.34 	80,794.25 	933.12 	UK Met Office HadCM3 short v8.36
i686-apple-darwin

ID: 65052 · Report as offensive
Jim1348

Send message
Joined: 15 Jan 06
Posts: 637
Credit: 26,751,529
RAC: 653
Message 65053 - Posted: 4 Feb 2022, 20:02:31 UTC - in response to Message 65051.  

A note from Dave in the getting started area has the answer, these have been amended to be Mac only so we’re out of luck.

I would never have thought of looking there. It is more the getting stopped area.
Thanks.
ID: 65053 · Report as offensive
Profile Iain Inglis
Volunteer moderator

Send message
Joined: 16 Jan 10
Posts: 1081
Credit: 6,970,380
RAC: 3,938
Message 65054 - Posted: 4 Feb 2022, 20:03:21 UTC

A batch #927 model has downloaded without problem to my Mac.
ID: 65054 · Report as offensive
Profile Dave Jackson
Volunteer moderator

Send message
Joined: 15 May 09
Posts: 4341
Credit: 16,483,830
RAC: 6,109
Message 65059 - Posted: 5 Feb 2022, 9:18:34 UTC

Tasks that have gone to Linux machines are about 70% failures, 20% success and the remainder still in progress. On Macs about 75% success, 20% failures and again the rest in progress. I was a bit surprised when I looked at the graph but I have only looked at a relatively early sudo random selection of tasks and certainly not enough to be close to have come up with a definitive result. FreeBSD interestingly has 10% failures with the rest still in progress.

I wanted the data to favour Linux so I could argue the case but it would seem the decision is justified. It would be interesting to know how many of the Linux failures are missing libraries and so a user problem rather than a Linux one but my script writing skills are not up to trawling through the data for that.
ID: 65059 · Report as offensive
Jean-David Beyer

Send message
Joined: 5 Aug 04
Posts: 1054
Credit: 16,510,175
RAC: 433
Message 65064 - Posted: 5 Feb 2022, 14:27:34 UTC - in response to Message 65059.  

Tasks that have gone to Linux machines are about 70% failures, 20% success and the remainder still in progress. On Macs about 75% success, 20% failures and again the rest in progress. I was a bit surprised when I looked at the graph but I have only looked at a relatively early sudo random selection of tasks and certainly not enough to be close to have come up with a definitive result. FreeBSD interestingly has 10% failures with the rest still in progress.


I have only looked at the ones where I got a work unit. Those of the
UK Met Office HadCM3 short v8.36 i686-pc-linux-gnu
category. IIRC, those from last April, most of them worked. Of the most recent batch, all failed but one.
My system,
Intel(R) Xeon(R) W-2245 CPU @ 3.90GHz [Family 6 Model 85 Stepping 7]
Red Hat Enterprise Linux 8.5 (Ootpa) [4.18.0-348.12.2.el8_5.x86_64|libc 2.28 (GNU libc)]
does have the 32-bit libraries and runs the
UK Met Office HadAM4 at N216 resolution v8.52.i686-pc-linux-gnu
tasks just fine.

I wanted the data to favour Linux so I could argue the case but it would seem the decision is justified. It would be interesting to know how many of the Linux failures are missing libraries and so a user problem rather than a Linux one but my script writing skills are not up to trawling through the data for that.


For the one I ran that succeeded, the other two failed. They both ran on
UK Met Office HadCM3 short v8.36 .i686-apple-darwin
I could not tell why they failed.

Task 22182726
Name hadcm3s_1k9d_200012_168_926_012129726_0
Workunit 12129726

Task 22190607
Name hadcm3s_1k9d_200012_168_926_012129726_1
Workunit 12129726

For the rest, many failed for missing 32-bit libraries, some for segmentation violations, a few for missing files (I think that is what those that said this means.I saw several of these. All mine on my machine failed within a few seconds for segmentation violations. Those on other machines failed for many reasons.
Stderr 	

<core_client_version>7.9.3</core_client_version>
<![CDATA[
<message>
process exited with code 12 (0xc, -244)</message>
<stderr_txt>
unzip:  cannot find or open /mydisks/b/boinc-client/projects/climateprediction.net/hadcm3s_se_8.36_i686-pc-linux-gnu.zip, /mydisks/b/boinc-client/projects/climateprediction.net/hadcm3s_se_8.36_i686-pc-linux-gnu.zip.zip or /mydisks/b/boinc-client/projects/climateprediction.net/hadcm3s_se_8.36_i686-pc-linux-gnu.zip.ZIP.
unzip:  cannot find or open /mydisks/b/boinc-client/projects/climateprediction.net/hadcm3s_um_8.36_i686-pc-linux-gnu.zip, /mydisks/b/boinc-client/projects/climateprediction.net/hadcm3s_um_8.36_i686-pc-linux-gnu.zip.zip or /mydisks/b/boinc-client/projects/climateprediction.net/hadcm3s_um_8.36_i686-pc-linux-gnu.zip.ZIP.
unzip:  cannot find or open hadcm3s_data_8.36_i686-pc-linux-gnu.zip, hadcm3s_data_8.36_i686-pc-linux-gnu.zip.zip or hadcm3s_data_8.36_i686-pc-linux-gnu.zip.ZIP.
unzip:  cannot find or open hadcm3s_1hgp_200012_168_926_012129245.zip, hadcm3s_1hgp_200012_168_926_012129245.zip.zip or hadcm3s_1hgp_200012_168_926_012129245.zip.ZIP.
cpdnmonitor: cannot open input file /mydisks/b/boinc-client/projects/climateprediction.net/hadcm3s_se_8.36_i686-pc-linux-gnu.so after 11 attempts
cpdnmonitor: cannot open input file /mydisks/b/boinc-client/projects/climateprediction.net/hadcm3s_um_8.36_i686-pc-linux-gnu after 11 attempts

</stderr_txt>
]]>


I thought this one very interesting. I saw at least five of these. Notice, I see many Apple Darwin failures.
Application version 	UK Met Office HadCM3 short v8.36
i686-apple-darwin
Stderr 	

<core_client_version>7.14.2</core_client_version>
<![CDATA[
<message>
process exited with code 255 (0xff, -1)</message>
<stderr_txt>
Process creation (../../projects/climateprediction.net/hadcm3s_8.36_i686-apple-darwin) failed: Bad CPU type in executable (errno = -1)

</stderr_txt>
]]>


Anyway, you gt the idea.
ID: 65064 · Report as offensive
Les Bayliss
Volunteer moderator

Send message
Joined: 5 Sep 04
Posts: 7629
Credit: 24,240,330
RAC: 0
Message 65073 - Posted: 5 Feb 2022, 20:31:35 UTC

Keeping in mind that cpdn is just an extra job for the Oxford people, getting these models to run in as quick a means as possible would play a part in the decision.
ID: 65073 · Report as offensive
Profile Alan K

Send message
Joined: 22 Feb 06
Posts: 484
Credit: 29,582,591
RAC: 839
Message 65075 - Posted: 5 Feb 2022, 23:40:51 UTC - in response to Message 65064.  

All 9 of the ones I got failed with segmentation violation. FYI Intel i5, Ubuntu 20.04.
ID: 65075 · Report as offensive
Jean-David Beyer

Send message
Joined: 5 Aug 04
Posts: 1054
Credit: 16,510,175
RAC: 433
Message 65077 - Posted: 6 Feb 2022, 0:47:19 UTC - in response to Message 65075.  

All 9 of the ones I got failed with segmentation violation. FYI Intel i5, Ubuntu 20.04.

14 of the ones I got failed with segmentation violation. The other one completed successfully.

Computer  1511241

CPU type  Genuine Intel   Intel(R) Xeon(R) W-2245 CPU @ 3.90GHz [Family 6 Model 85 Stepping 7]

Operating System 	  Red Hat Enterprise Linux 8.5 (Ootpa) [4.18.0-348.12.2.el8_5.x86_64|libc 2.28 (GNU libc)]

ID: 65077 · Report as offensive
KAMasud

Send message
Joined: 6 Oct 06
Posts: 204
Credit: 7,608,986
RAC: 0
Message 65107 - Posted: 7 Feb 2022, 18:43:20 UTC

All Linux WU's finished and none for Windows. Now what, fishing for WU's?
ID: 65107 · Report as offensive
Les Bayliss
Volunteer moderator

Send message
Joined: 5 Sep 04
Posts: 7629
Credit: 24,240,330
RAC: 0
Message 65108 - Posted: 7 Feb 2022, 19:34:25 UTC - in response to Message 65107.  

Now we wait until a researcher somewhere on the planet wants data for their research and provides new data sets.

Or get a Mac and run those. :)
ID: 65108 · Report as offensive
Profile Dave Jackson
Volunteer moderator

Send message
Joined: 15 May 09
Posts: 4341
Credit: 16,483,830
RAC: 6,109
Message 65109 - Posted: 7 Feb 2022, 20:59:39 UTC - in response to Message 65108.  

Now we wait until a researcher somewhere on the planet wants data for their research and provides new data sets.

Or get a Mac and run those. :)


But not as George has shown post Catalina.
ID: 65109 · Report as offensive
Nigel Garvey

Send message
Joined: 5 May 10
Posts: 69
Credit: 1,169,103
RAC: 2,258
Message 65110 - Posted: 7 Feb 2022, 22:15:20 UTC - in response to Message 65109.  
Last modified: 7 Feb 2022, 22:17:45 UTC

But not as George has shown post Catalina.

Not post Mojave actually. 32-bit support was dropped as from Catalina (macOS 10.15).
NG
ID: 65110 · Report as offensive
KAMasud

Send message
Joined: 6 Oct 06
Posts: 204
Credit: 7,608,986
RAC: 0
Message 65113 - Posted: 8 Feb 2022, 5:14:48 UTC - in response to Message 65108.  

Now we wait until a researcher somewhere on the planet wants data for their research and provides new data sets.

Or get a Mac and run those. :)

____________________

I do not think I am that addicted to now chase after a Mac.
ID: 65113 · Report as offensive
Profile Dave Jackson
Volunteer moderator

Send message
Joined: 15 May 09
Posts: 4341
Credit: 16,483,830
RAC: 6,109
Message 65116 - Posted: 8 Feb 2022, 10:33:18 UTC
Last modified: 8 Feb 2022, 10:33:54 UTC

Two things,
1. The first two successes have come through on the new batch.
2. In future this task type will get 5 attempts like the Linux ones to give a higher chance for each work unit to complete as so many newer machines or those updated to newer versions of the OS are not going to be able to run the 32bit tasks.

Edit: Not quite sure if this is being changed for current batch or not.
ID: 65116 · Report as offensive
Jean-David Beyer

Send message
Joined: 5 Aug 04
Posts: 1054
Credit: 16,510,175
RAC: 433
Message 65117 - Posted: 8 Feb 2022, 13:10:42 UTC - in response to Message 65116.  

In future this task type will get 5 attempts like the Linux ones to give a higher chance for each work unit to complete as so many newer machines or those updated to newer versions of the OS are not going to be able to run the 32bit tasks.


Can the OpenIFS tasks be released sooner to the general populace of users? Or are they so far from ready as not to be worth the bother of doing this?

Right now neither climate prediction.net nor rosetta@home have any work units for normal users.
ID: 65117 · Report as offensive
Profile Dave Jackson
Volunteer moderator

Send message
Joined: 15 May 09
Posts: 4341
Credit: 16,483,830
RAC: 6,109
Message 65118 - Posted: 8 Feb 2022, 13:22:55 UTC

Can the OpenIFS tasks be released sooner to the general populace of users? Or are they so far from ready as not to be worth the bother of doing this?


I have no idea about what needs to happen before OpenIFS makes it to the main site. Of course, even when it does it still misses out the biggest group which is Windows users who don't want to or have the knowledge to jump through the hoops to run Linux tasks on their machines as OpenIFS is Linux and Mac only.
ID: 65118 · Report as offensive
Les Bayliss
Volunteer moderator

Send message
Joined: 5 Sep 04
Posts: 7629
Credit: 24,240,330
RAC: 0
Message 65119 - Posted: 8 Feb 2022, 13:26:38 UTC - in response to Message 65117.  

This new model type is still at the Alpha testing stage.
it could be months before they go live.
ID: 65119 · Report as offensive
Nigel Garvey

Send message
Joined: 5 May 10
Posts: 69
Credit: 1,169,103
RAC: 2,258
Message 65120 - Posted: 8 Feb 2022, 13:46:13 UTC - in response to Message 65117.  

Jean-David Beyer wrote:
… normal users.

:-D
NG
ID: 65120 · Report as offensive
Jean-David Beyer

Send message
Joined: 5 Aug 04
Posts: 1054
Credit: 16,510,175
RAC: 433
Message 65121 - Posted: 8 Feb 2022, 14:53:42 UTC - in response to Message 65119.  

This new model type is still at the Alpha testing stage.
it could be months before they go live.


I wondered because one project, WCG, has a project known as Beta Testing that you can volunteer for and if you do, sometimes you get tasks that are not yet suitable for general use, but helps the projects by discovering problems not discovered by normal development and system tests. Or they may find everything is just fine and can release it to the general users. I even get credit for running those.

Now it may not make sense for volunteers to even run tasks that are only at the Alpha testing stage.
ID: 65121 · Report as offensive
Les Bayliss
Volunteer moderator

Send message
Joined: 5 Sep 04
Posts: 7629
Credit: 24,240,330
RAC: 0
Message 65122 - Posted: 8 Feb 2022, 15:26:02 UTC - in response to Message 65121.  

The beta testers for cpdn are the moderators.
ID: 65122 · Report as offensive
Previous · 1 . . . 72 · 73 · 74 · 75 · 76 · 77 · 78 . . . 91 · Next

Message boards : Number crunching : New work Discussion

©2024 climateprediction.net