Also a crash report, attached.ĮRROR: apport (pid 5620) Sat May 23 09:38:36 2020: called for pid 5129, signal 11, core limit 0, dump mode 1ĮRROR: apport (pid 5620) Sat May 23 09:38:36 2020: executable: /usr/bin/ gtk-gnutella (command line "/usr/bin/ gtk-gnutella -child")ĮRROR: apport (pid 5620) Sat May 23 09:38:36 2020: gdbus call error: Error: GDBus.Error: org.freedesktop. Tried again and realised that there were messages in apport.log. Then it is not really a problem of cpu-use but that the amount of read and write requests aMule is making is keeping the hard drive really busy.20-05-16 14:04:30.694 (DEBUG): loaded 330525 geographical IPv4 ranges (2900521315 hosts) from "/usr/share/ gtk-gnutella/ geo-ip. Personally, I would never use aMule on a Pi because it has been resource intensive and a bit buggy even on my x86 machine.Īm I right, in that you use an external hard drive with ntfs while the tutorial you referred to used only the internal microSD-card? Maybe that explains the difference in success? (aMule probably does a huge amount of small writes and reads.)Įdit: Maybe top shows IO-Wait as a part of the cpu use, and the reason for the high cpu use is that the process has to wait when the spinning hard drive tries to fulfill the numerous read and write requests from aMule. Maybe adjusting those things could help slightly.
It should also be possible to limit many things like the amount of connections, downloading and uploading speeds etc. I have never used those either.ĪMule got a setting to increase the size of the file-buffer - maybe that could help slightly. For some other networks there are gtk-gnutella and gnunet-gtk. I have never used it, so I don't know if it is more efficient, or if it works at all on on a Pi. I'm too lazy to check Raspbian's repository but my x86 Ubuntu got, for the donkey-network, in addtion to aMule a program called mldonkey-server.
0 kommentar(er)
