On Donnerstag, 11. Juni 2020 10:05:41 CEST rosea.grammostola wrote: > Ok, that might sound somewhat aggressive, but I've to use 'htop' > sometimes to kill linuxsampler. killall -9 linuxsampler doesn't work always. > > Then the question is, how to I kill the LSCP server, I'm sometimes getting: > > Starting LSCP network server (0.0.0.0:8888)...LSCPServer: Could not bind > server socket, retrying for 180 seconds... > > While Linuxsampler doesn't even seems to run.
Sounds like an old kernel issue to me: Years ago, when LS was killed or crashed, then it used to happen that the Linux kernel still listed its port (8888) as still being in use for a while, even though the LS process (which actually used the port) was gone. However since the kernel thought the port was still in use, it prevented the restarted LS to bind to that port immediately. In that case you had to wait for around 1 or 2 minutes until the kernel finally freed that port and then LS eventually was able to bind to the port in its wait loop that you pasted above. However I have not seen this issue in years, so I thought it was fixed on kernel side long ago. What's the kernel version that you are using, and is there something special regarding your installation? But LS is still able to bind to the port automatically after some waiting time there, right? CU Christian _______________________________________________ Linuxsampler-devel mailing list Linuxsampler-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/linuxsampler-devel
