This also may be of interest: there is a peropen address created about halfway down the list that never gets destroyed.
Andrew Fleckenstein On Fri, Aug 10, 2012 at 12:49:05AM +0200, Richard Braun wrote: > Hello, > > While investigating an issue in the term server, I think I came across > a weird problem. The issue is that, after logging in and out through > ssh, a pty master will still consider itself open and refuse new > sessions, leading to the exhaustion of all the precreated pairs after > some time. > > The GDB traced I've obtained [1] shows a couple of matching > create/destroy hooks, but the first (create) call, which trivfs control > matches ptyctl (po->cntl == ptyctl), is never destroyed. At the same > time, the last call shows a peropen address never seen before. How is > it possible to see a peropen at release time never seen at creation > time ? > > -- > Richard Braun >
