New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Radio stream memory leak #4665
Comments
I was going to run clementine in valgrind so I will take a look if I can find anything |
After 2.5h I only got: |
Pleas try http://builds.clementine-player.org/ and report back if problem is still present. Thanks |
I can confirm this issue. Playing for example this stream over the night or the weekend, causes clementine to collect up to 4GB RAM (or even more, but my machine gets laggy at that point and I have to kill clementine every time) at the morning: http://listen.housetime.fm/tunein-mp3-pls I'm experiencing this issue at least since Fedora 19 or 20 and I'm currently using Fedora 21 which comes with clementine 1.2.3-2. I'll try to get a corresponding valgrind output |
@Armadill0 Okay, but what about with the latest build ? |
Just installed clementine-1.2.3-2.993.g1f15df9.fc21.x86_64.rpm. I'll report the result in the next days. |
Thanks :) |
It happened again:
Clementine consumed over 3GB of system memory on my machine when I killed it. |
Anything new about this bug? I am getting heavy leaking while streaming radio with clementine 1.2.3 (arch linux). |
After I saw that my Clementine 1.3.1 (Arch Linux amd64) was using almost 3GB of real memory after a week or so runtime (I only used sleep mode), I ran a leak check and after just a few seconds of streaming two radios Valgrind showed
|
The issue still present. Started with ~60Mb consumed after several hours >400Mb. log summary: found) are not shown. |
Still there. Version 1.3.1 on OS X 10.13.4 |
Hi @jonas747 , are you using Windows version? (from your glasses in avatar) What's behaviour now? Hi @sobkas , still using Clementine on Debian? What's behaviour now? Hi @Armadill0 still on Fedora? What's behaviour now? Hi @schnelle86 and @OlafLostViking , sorry but afaik Clementine is not supported on Arch Linux... unfortunately (I also use that, sometimes). Hi @V-ctor thanks for valgrind package. If several of hours of using was 'only' 400MB then we have enormous improvement, isn't it ;) (vs few GBs in hour, previously...) Hi @risototh I don't have any free Mac OS X to test but if you could you try to use valgrind to help us it will be awesome! :) And again, please remember to test update against up-to-date source. Best using valgrind. This is not very hard to learn how to use that. Best regards. :) |
Hi Volodia,
any thoughts?
risko (master) ~/Downloads/valgrind $ valgrind /Applications/clementine.app/Contents/MacOS/clementine
==32571== Memcheck, a memory error detector
==32571== Copyright (C) 2002-2017, and GNU GPL'd, by Julian Seward et al.
==32571== Using Valgrind-3.16.0.GIT and LibVEX; rerun with -h for copyright info
==32571== Command: /Applications/clementine.app/Contents/MacOS/clementine
==32571==
--32571-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option
--32571-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option (repeated 2 times)
--32571-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option (repeated 4 times)
--32571-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option (repeated 8 times)
--32571-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option (repeated 16 times)
--32571-- WARNING: unhandled amd64-darwin syscall: unix:463
--32571-- You may be able to write your own handler.
--32571-- Read the file README_MISSING_SYSCALL_OR_IOCTL.
--32571-- Nevertheless we consider this a bug. Please report
--32571-- it at http://valgrind.org/support/bug_reports.html.
--32571-- WARNING: unhandled amd64-darwin syscall: unix:463
--32571-- You may be able to write your own handler.
--32571-- Read the file README_MISSING_SYSCALL_OR_IOCTL.
--32571-- Nevertheless we consider this a bug. Please report
--32571-- it at http://valgrind.org/support/bug_reports.html.
==32571== Thread 2:
==32571== Invalid read of size 4
==32571== at 0x107B70E3A: ??? (in /usr/lib/system/libsystem_pthread.dylib)
==32571== by 0x107B70BE8: ??? (in /usr/lib/system/libsystem_pthread.dylib)
==32571== Address 0x18 is not stack'd, malloc'd or (recently) free'd
==32571==
==32571==
==32571== Process terminating with default action of signal 11 (SIGSEGV)
==32571== Access not within mapped region at address 0x18
==32571== at 0x107B70E3A: ??? (in /usr/lib/system/libsystem_pthread.dylib)
==32571== by 0x107B70BE8: ??? (in /usr/lib/system/libsystem_pthread.dylib)
==32571== If you believe this happened as a result of a stack
==32571== overflow in your program's main thread (unlikely but
==32571== possible), you can try to increase the size of the
==32571== main thread stack using the --main-stacksize= flag.
==32571== The main thread stack size used in this run was 8388608.
--32571:0:schedule VG_(sema_down): read returned -4
==32571==
==32571== HEAP SUMMARY:
==32571== in use at exit: 0 bytes in 0 blocks
==32571== total heap usage: 0 allocs, 0 frees, 0 bytes allocated
==32571==
==32571== All heap blocks were freed -- no leaks are possible
==32571==
==32571== For lists of detected and suppressed errors, rerun with: -s
==32571== ERROR SUMMARY: 1 errors from 1 contexts (suppressed: 52 from 4)
Segmentation fault: 11
… Dňa 14. 9. 2019 o 4:25, Volodia ***@***.***> napísal:
Hi @jonas747 <https://github.com/jonas747> , are you using Windows version? (from your glasses in avatar) What's behaviour now?
Hi @sobkas <https://github.com/sobkas> , still using Clementine on Debian? What's behaviour now?
Hi @Armadill0 <https://github.com/Armadill0> still on Fedora? What's behaviour now?
Will be glad if could provide info about leaked memory on your installation if it persist.
Hi @schnelle86 <https://github.com/schnelle86> and @OlafLostViking <https://github.com/OlafLostViking> , sorry but afaik Clementine is not supported on Arch Linux... unfortunately (I also use that, sometimes).
But I think you could test that against lastest and tell us where memory leaks if you know how to do it :) I hope the fix will help for you both, too :)
Hi @V-ctor <https://github.com/V-ctor> thanks for valgrind package. If several of hours of using was 'only' 400MB then we have enormous improvement, isn't it ;) (vs few GBs in hour, previously...)
Sorry, but could you test against stream which @jonas747 <https://github.com/jonas747> in first comment mentioned (I checked it is working, even muted) and see what happens on your set up.
Hi @risototh <https://github.com/risototh> I don't have any free Mac OS X to test but if you could you try to use valgrind to help us it will be awesome! :)
https://blog.mozilla.org/nnethercote/2009/01/18/me-valgrind-and-mac-os-x/ <https://blog.mozilla.org/nnethercote/2009/01/18/me-valgrind-and-mac-os-x/>
http://www.valgrind.org/downloads/current.html <http://www.valgrind.org/downloads/current.html>
And again, please remember to test update against up-to-date source. Best using valgrind. This is not very hard to learn how to use that.
Best regards. :)
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub <#4665?email_source=notifications&email_token=AAGDMS4EDXOHPHPWPQFYLXLQJRDP7A5CNFSM4AZL7BG2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD6WSC4Q#issuecomment-531439986>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAGDMS6WD4MC25YRW3TIVITQJRDP7ANCNFSM4AZL7BGQ>.
--
S pozdravom,
Richard Tóth
mailto: risko@risko.org <mailto:riso@iklub.sk>
ICQ: 223531718
VoIP: +421-41-5131787
O2: +421-908-936652
|
Hmm, I don't know if it is Valgrind error indeed. I bet not (accordingly to part about possible valgrind issue).I will need to spin off some system based on Mach-O in VM... there was time I had one... Need to wait, unfortunatelly... :-( Cheers to Slovenská republika from Poland ;) (only kind thing I could say for now...) PS this is probably because libsqlite3 usage - https://www.mail-archive.com/kde-bugs-dist@kde.org/msg177108.html so is a partially valgrind and partially libsqlite3 exeption handling bug, not know what to do with that valgrind run. |
OK, I will restart the clementine here and there as usually :D
I'm not really at home in this kind of debugging...
I'm mainly a PHP developer... Only here and there a C/C++ HW developer for some MCUs...
… Dňa 27. 9. 2019 o 0:13, Volodia ***@***.***> napísal:
Hi Volodia, any thoughts? risko (master) ~/Downloads/valgrind $ valgrind /Applications/clementine.app/Contents/MacOS/clementine ==32571== Memcheck, a memory error detector ==32571== Copyright (C) 2002-2017, and GNU GPL'd, by Julian Seward et al. ==32571== Using Valgrind-3.16.0.GIT and LibVEX; rerun with -h for copyright info ==32571== Command: /Applications/clementine.app/Contents/MacOS/clementine ==32571== --32571-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option --32571-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option (repeated 2 times) --32571-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option (repeated 4 times) --32571-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option (repeated 8 times) --32571-- UNKNOWN mach_msg unhandled MACH_SEND_TRAILER option (repeated 16 times) --32571-- WARNING: unhandled amd64-darwin syscall: unix:463 --32571-- You may be able to write your own handler. --32571-- Read the file README_MISSING_SYSCALL_OR_IOCTL. --32571-- Nevertheless we consider this a bug. Please report --32571-- it at http://valgrind.org/support/bug_reports.html <http://valgrind.org/support/bug_reports.html>. --32571-- WARNING: unhandled amd64-darwin syscall: unix:463 --32571-- You may be able to write your own handler. --32571-- Read the file README_MISSING_SYSCALL_OR_IOCTL. --32571-- Nevertheless we consider this a bug. Please report --32571-- it at http://valgrind.org/support/bug_reports.html <http://valgrind.org/support/bug_reports.html>. ==32571== Thread 2: ==32571== Invalid read of size 4 ==32571== at 0x107B70E3A: ??? (in /usr/lib/system/libsystem_pthread.dylib) ==32571== by 0x107B70BE8: ??? (in /usr/lib/system/libsystem_pthread.dylib) ==32571== Address 0x18 is not stack'd, malloc'd or (recently) free'd ==32571== ==32571== ==32571== Process terminating with default action of signal 11 (SIGSEGV) ==32571== Access not within mapped region at address 0x18 ==32571== at 0x107B70E3A: ??? (in /usr/lib/system/libsystem_pthread.dylib) ==32571== by 0x107B70BE8: ??? (in /usr/lib/system/libsystem_pthread.dylib) ==32571== If you believe this happened as a result of a stack ==32571== overflow in your program's main thread (unlikely but ==32571== possible), you can try to increase the size of the ==32571== main thread stack using the --main-stacksize= flag. ==32571== The main thread stack size used in this run was 8388608. --32571:0:schedule VG_(sema_down): read returned -4 ==32571== ==32571== HEAP SUMMARY: ==32571== in use at exit: 0 bytes in 0 blocks ==32571== total heap usage: 0 allocs, 0 frees, 0 bytes allocated ==32571== ==32571== All heap blocks were freed -- no leaks are possible ==32571== ==32571== For lists of detected and suppressed errors, rerun with: -s ==32571== ERROR SUMMARY: 1 errors from 1 contexts (suppressed: 52 from 4) Segmentation fault: 11
… <x-msg://4/#>
Dňa 14. 9. 2019 o 4:25, Volodia @.***> napísal: Hi @jonas747 <https://github.com/jonas747> https://github.com/jonas747 <https://github.com/jonas747> , are you using Windows version? (from your glasses in avatar) What's behaviour now? Hi @sobkas <https://github.com/sobkas> https://github.com/sobkas <https://github.com/sobkas> , still using Clementine on Debian? What's behaviour now? Hi @Armadill0 <https://github.com/Armadill0> https://github.com/Armadill0 <https://github.com/Armadill0> still on Fedora? What's behaviour now? Will be glad if could provide info about leaked memory on your installation if it persist. Hi @schnelle86 <https://github.com/schnelle86> https://github.com/schnelle86 <https://github.com/schnelle86> and @OlafLostViking <https://github.com/OlafLostViking> https://github.com/OlafLostViking <https://github.com/OlafLostViking> , sorry but afaik Clementine is not supported on Arch Linux... unfortunately (I also use that, sometimes). But I think you could test that against lastest and tell us where memory leaks if you know how to do it :) I hope the fix will help for you both, too :) Hi @V-ctor <https://github.com/V-ctor> https://github.com/V-ctor <https://github.com/V-ctor> thanks for valgrind package. If several of hours of using was 'only' 400MB then we have enormous improvement, isn't it ;) (vs few GBs in hour, previously...) Sorry, but could you test against stream which @jonas747 <https://github.com/jonas747> https://github.com/jonas747 <https://github.com/jonas747> in first comment mentioned (I checked it is working, even muted) and see what happens on your set up. Hi @risototh <https://github.com/risototh> https://github.com/risototh <https://github.com/risototh> I don't have any free Mac OS X to test but if you could you try to use valgrind to help us it will be awesome! :) https://blog.mozilla.org/nnethercote/2009/01/18/me-valgrind-and-mac-os-x/ <https://blog.mozilla.org/nnethercote/2009/01/18/me-valgrind-and-mac-os-x/> https://blog.mozilla.org/nnethercote/2009/01/18/me-valgrind-and-mac-os-x/ <https://blog.mozilla.org/nnethercote/2009/01/18/me-valgrind-and-mac-os-x/> http://www.valgrind.org/downloads/current.html <http://www.valgrind.org/downloads/current.html> http://www.valgrind.org/downloads/current.html <http://www.valgrind.org/downloads/current.html> And again, please remember to test update against up-to-date source. Best using valgrind. This is not very hard to learn how to use that. Best regards. :) — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#4665 <#4665>?email_source=notifications&email_token=AAGDMS4EDXOHPHPWPQFYLXLQJRDP7A5CNFSM4AZL7BG2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD6WSC4Q#issuecomment-531439986>, or mute the thread https://github.com/notifications/unsubscribe-auth/AAGDMS6WD4MC25YRW3TIVITQJRDP7ANCNFSM4AZL7BGQ <https://github.com/notifications/unsubscribe-auth/AAGDMS6WD4MC25YRW3TIVITQJRDP7ANCNFSM4AZL7BGQ>.
-- S pozdravom, Richard Tóth mailto: ***@***.*** ***@***.***> ***@***.*** ***@***.***> ICQ: 223531718 VoIP: +421-41-5131787 O2: +421-908-936652
Hmm, I don't know if it is Valgrind error indeed. I bet not (accordingly to part about possible valgrind issue).I will need to spin off some system based on Mach-O in VM... there was time I had one... Need to wait, unfortunatelly... :-(
Cheers to Slovenská republika from Poland ;) (only kind thing I could say for now...)
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub <#4665?email_source=notifications&email_token=AAGDMSZBDE34TUNE23FKMC3QLUXXTA5CNFSM4AZL7BG2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD7XDR7Y#issuecomment-535705855>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAGDMS3SOYXMBI3MSCAX7BTQLUXXTANCNFSM4AZL7BGQ>.
--
S pozdravom,
Richard Tóth
mailto: risko@risko.org <mailto:riso@iklub.sk>
ICQ: 223531718
VoIP: +421-41-5131787
O2: +421-908-936652
|
Mmm! Yummy μC! PHP is nice, but it's naming convensions... I have 10.12 to get up but I must have time for it. I will respond when I get something making clementine more juicy! |
@JulianVolodia I'm going to test it the next days. Getting back to you soon. |
@JulianVolodia I couldn't reproduce it on Fedora last week in several tries. Clementine didn't consume more than 400MiB of memory after some hours of usage. |
@Armadill0 thanks! If you could tag version which you have used, from package manager for example? |
@JulianVolodia It's been 1.3.1. |
@Armadill0 thanks. @risototh could you update your MacOS (10.13.4 isn't last update of 10.13...) as I do not recognize bug mentioned above? Could you then retest without any other working software (as browser, desktop custom features, plugins and so on?). I will have long way to set up my "MacOS" station from here but want to triage this ASAP if it is still a bug in Clementine or strange thing with system. |
@JulianVolodia did a new check. As far as I understand it still presents. ==19653== LEAK SUMMARY: |
Listening to a radio stream will just keep consuming more and more memory, after around 10 hours of listening to 1 stream it will be up to around 1GB. Happens on both windows and linux with clementine 1.2.3.
How to reproduce:
Listen to a radio stream (in my case http://lyd.nrk.no/nrk_radio_p3_mp3_h)
The text was updated successfully, but these errors were encountered: