News for hackers. Stuff that matters.
Re:Memleax, new version 0.4 by hellowub on Tuesday June 28, 2016 @01:15AM (#11991622)
I submitted my project here before when I just wrote it out. Now I have realized all features that I can think (e.g. FreeBSD, libdw, expire memblocks active), fixed all known bugs, tested on CentOS, Ubuntu, Debian and FreeBSD, and have no plan any more on it. Then I post it again here for some new advices.

Besides, if you have a production service with very slow memory leak. The leak is so slow that you do not want to fix it. Then you should try `memleax`. It’s convenient and suitable for this case specially.

Thanks.

Re:Memleax, new version 0.4 by corysama on Tuesday June 28, 2016 @01:53AM (#11991739)
This looks really useful. Thanks for making it!
Re:Memleax, new version 0.4 by hellowub on Tuesday June 28, 2016 @02:02AM (#11991768)
Thanks. Any advice is welcomed.
Re:Memleax, new version 0.4 by lttlrck on Tuesday June 28, 2016 @11:47AM (#11994686)
Thank you!!! This helped me find two memory leaks that I have been trying to locate for 2 weeks - in less than an hour. It wasn't a live system but being able to attach to a running process and specify a timeout allowed me to almost completely filter out init and garbage collection from the output. Valgrind was not helpful, I was at the point of looking at raw core dumps (and now I know what the problem is it would probably have worked...)
Re:Memleax, new version 0.4 by hellowub on Tuesday June 28, 2016 @10:50PM (#11999197)
Glad to hear this.
Re:Memleax, new version 0.4 by cyphar on Tuesday June 28, 2016 @06:04AM (#11992565)
+1 for using the GPL. I'm always happy to see more free software licensed under strong copyleft licenses.