On Fri, Oct 03, 2008 at 12:01:43PM -0500, Doug Tucker wrote: > > > If a kill -9 won't kill the process you have either a kernel > > bug or a hardware problem, not a Samba issue. > > Jeremy. > > Thanks, we have 2 cifs servers, and they are both having the issue. I'm > doubting a hardware problem. Not sure where to go with a "kernel bug". > Doing a "restart" on samba does fix the issue, I dont' have to fully > reboot.
When you've done a "restart" are there any of the old smbd processes left around ? If not, then you are incorrect, and a kill -9 does terminate them. If you can pick a process id and kill -9 it and it doesn't die, then you have a kernel bug or hardware issue. That's "the rules" :-). > But, the real issue (I think) is not that I can't kill -9, but that the > processes are refusing to close when the client goes away, or am I off > in my thinking here? Is there some type of timeout value for samba > processes I'm unaware of I should be setting? No, smbd should go away once the client has disconnected (all calls on the socket will return -1 and smbd will terminate itself). Jeremy. -- To unsubscribe from this list go to the following URL and read the instructions: https://lists.samba.org/mailman/listinfo/samba
