Message boards :
Number crunching :
Please post app 1.17/1.18/1.19 memory leaks/errors
Message board moderation
Previous · 1 · 2 · 3 · 4 · 5 · Next
Author | Message |
---|---|
Send message Joined: 17 Feb 08 Posts: 363 Credit: 258,227,990 RAC: 0 |
wuid=4456887 - This one is just finished with v1.19 (Windows). The "Progress" meter worked, but not correctly. I am guessing it is off by an order of magnitude. That is, Progress only showed about 5% when it had run about 5 minutes (roughly the halfway point timewise). That's correct. It's not linear. about 75% of the time is spend in a more "intensive computation routine" the rest goes pretty quick. Thanks for the info. Glad it's working for you. EDIT OK i just had a look at the app page... http://milkyway.cs.rpi.edu/milkyway/apps.php there's no 64 bit linux app listed there. I'm sure that's just a glitch and it'll be corrected soon. Thanks. /EDIT Join Support science! Joinc Team BOINC United now! |
Send message Joined: 29 Sep 07 Posts: 18 Credit: 4,533,464 RAC: 0 |
I've returned my first 1.19 results. Windows looks like 2 times faster then before for 1.18 and 1.19 based on granted credit. resultid=4641577 1.19 resultid=4628989 1.18 linux 32 looks like 33% faster then before with 1.19 also based on granted credit resultid=4642547 1.19 resultid=4642361 1.19 @Crunch3r: you did an very good job my Progress bar works like Stick wrote. 8 min for 10%, and the last min 90% but no restart of the progress bar with i saw on the old apps EDIT: and no more "Memory Leaks Detected" on the win app Matthias |
Send message Joined: 17 Feb 08 Posts: 363 Credit: 258,227,990 RAC: 0 |
I've returned my first 1.19 results. Thanks! Though i onlyx compiled the apps... Travis and Dave did all the work! So yeah as posted before the progress is not linear... won't go into details much but what you've posted seems to be perfectly ok. The processing is "split" into two separate functions that is why the progress is not linear at all ... The first part is "calculating integrals" wich is slower way slower... after that the app switches to "calculateing likelihood" wich is pretty fast... HTH Join Support science! Joinc Team BOINC United now! |
Send message Joined: 8 Oct 07 Posts: 289 Credit: 3,690,838 RAC: 0 |
Heres my stderr out on 32bit linux box #7406...unless my quotq is reset I can't try anymore for 7 hours and this was version 1,19 <core_client_version>5.10.8</core_client_version> <![CDATA[ <message> process exited with code 127 (0x7f, -129) </message> <stderr_txt> astronomy_1.19_i686-pc-linux-gnu: error while loading shared libraries: libstdc++.so.5: cannot open shared object file: No such file or directory </stderr_txt> |
Send message Joined: 8 Oct 07 Posts: 289 Credit: 3,690,838 RAC: 0 |
double-post |
Send message Joined: 17 Feb 08 Posts: 363 Credit: 258,227,990 RAC: 0 |
Heres my stderr out on 32bit linux box #7406...unless my quotq is reset I can't try anymore for 7 hours and this was version 1,19 Hmmm. that suckz... i'll take a look into it. (basically the 32 bit app that you got isn't for 64 bit host... i'm sure Dave or Travis will add the appropriate 64 bit app as soon as they notice that one here) Join Support science! Joinc Team BOINC United now! |
Send message Joined: 30 Aug 07 Posts: 2046 Credit: 26,480 RAC: 0 |
wuid=4456887 - This one is just finished with v1.19 (Windows). The "Progress" meter worked, but not correctly. I am guessing it is off by an order of magnitude. That is, Progress only showed about 5% when it had run about 5 minutes (roughly the halfway point timewise). yeah crunch3r had it right about this. the first 10% is doing an integral calculation (which is pretty computationally intensive), and the last 90% is comparing the star values to this integral calculation. there's really no good way to calculate the progress in a linear way, since on some architectures the integral takes a lot less time than on others (probably due to optimizations). |
Send message Joined: 30 Aug 07 Posts: 2046 Credit: 26,480 RAC: 0 |
Heres my stderr out on 32bit linux box #7406...unless my quotq is reset I can't try anymore for 7 hours and this was version 1,19 whats the problem here? BOINC is running the 32 bit linux app on a 64 bit machine and it's not working? i was pretty sure we had a 64 bit linux and that it was working... |
Send message Joined: 28 Aug 07 Posts: 146 Credit: 10,702,977 RAC: 1,827 |
Heres my stderr out on 32bit linux box #7406...unless my quotq is reset I can't try anymore for 7 hours and this was version 1,19 Hm, not quite sure, still a newbie to Linux, but this could be due to a missing lib. I think libstdc++.so.5 is part of the libs ia32 resp. ia64. I have them both installed so this could be because I don't have any problem with it... |
Send message Joined: 8 Oct 07 Posts: 289 Credit: 3,690,838 RAC: 0 |
Heres my stderr out on 32bit linux box #7406...unless my quotq is reset I can't try anymore for 7 hours and this was version 1,19 no...host 7406 and 3296 are both 32bit and getting those errors on 1.19 and have reached daily quota due to errors....host 3403 is also 32 bit but running 1.19 ok so I am baffled...what are the missing libraries the stderr is referring to? for 32 or 64 bit? Looks like the server is trying to send a 64bit app to 32 bit machines. |
Send message Joined: 30 Aug 07 Posts: 2046 Credit: 26,480 RAC: 0 |
Heres my stderr out on 32bit linux box #7406...unless my quotq is reset I can't try anymore for 7 hours and this was version 1,19 Hmm, i looked in our applications and it looks like the 64 bit linux was misnamed so it didnt get added. We should have a 1.17 version of x86_64-pc-linux-gnu now. Let me know if you get that and it works (thats 64 bit linux right?). |
Send message Joined: 8 Oct 07 Posts: 289 Credit: 3,690,838 RAC: 0 |
No those are 32bit machines apparently trying to load a 64bit app of 1.19. Can't try anymore unless you reset the quotas for hosts 7406 and 3296.Says I have reached max and defering communications for 7 hrs |
Send message Joined: 30 Aug 07 Posts: 2046 Credit: 26,480 RAC: 0 |
No those are 32bit machines apparently trying to load a 64bit app of 1.19. ahhh ok. we're going to need to recompile the 32 bit linux then. However, it looks like the error you got was more of a linking issue than a problem with the binary... |
Send message Joined: 27 Aug 07 Posts: 647 Credit: 27,592,547 RAC: 0 |
No those are 32bit machines apparently trying to load a 64bit app of 1.19. I'm just crunching Milkyway under Linux 64-bit. Funny enough the "wrong" 1.19 app (32-bit) was running faster than the "correct" 1.17 app (64-bit). Under 1.19 my units took 7:40 minutes to crunch, now under 1.17 they took 10:11 minutes. *scratches head* EDIT: See http://milkyway.cs.rpi.edu/milkyway/results.php?hostid=8261 Lovely greetings, Cori |
Send message Joined: 8 Oct 07 Posts: 289 Credit: 3,690,838 RAC: 0 |
No those are 32bit machines apparently trying to load a 64bit app of 1.19. Ok I am going to try to detach 1 of those hosts and reattach...that should take care of the quota thing and see if it links me to the 32bit app instead of the 64 bit...will let you know. |
Send message Joined: 12 Nov 07 Posts: 2425 Credit: 524,164 RAC: 0 |
1.19, progress meter seems ok. but I'm back to getting the units that freeze again (seems the same prob I had before). 2 stopped at 2 secs, the next at 4:15, out of the 4 that have ran. I had no probs with 1.18, I ran it since it was put out. |
Send message Joined: 28 Aug 07 Posts: 146 Credit: 10,702,977 RAC: 1,827 |
Hmm, i looked in our applications and it looks like the 64 bit linux was misnamed so it didnt get added. Just had the first of the 1.17 WUs for my machine, but strangely they now take a minute longer than the 1.19 32-Bit before... Edit: LOL, Cori posted the same issue. :-) |
Send message Joined: 27 Aug 07 Posts: 647 Credit: 27,592,547 RAC: 0 |
Hmm, i looked in our applications and it looks like the 64 bit linux was misnamed so it didnt get added. Yep. :-) The only difference is that my new units were more than 2 minutes slower. *LOL* Lovely greetings, Cori |
Send message Joined: 12 Oct 07 Posts: 77 Credit: 404,471,187 RAC: 0 |
|
Send message Joined: 17 Feb 08 Posts: 363 Credit: 258,227,990 RAC: 0 |
Funny enough the "wrong" 1.19 app (32-bit) was running faster than the "correct" 1.17 app (64-bit).Same thing here Ok guys/gals... i can offer a quick fix for that one... Travis ? you don't mind ? I'm compiling a proper x64 bit app for linux now ;) Join Support science! Joinc Team BOINC United now! |
©2024 Astroinformatics Group