On 12-12-04 13:25, Sam Hartman <hartm...@debian.org> wrote: > Hi. I'm happy to apply the upstream patch to sid. It's a bit more > tricky to get this into wheezy. In evaluating whether that's justified, > can I get any information you have on how often this bug is going to > trigger both in a theoretical sense and a practical sense?
Theoretically this should happen in case of network or server errors causing all the sockets to be in an error state. Perhaps timeouts also play a role, though I'm not sure there. Practically, we roughly see 5 or 6 new looping sshd processes per day appearing on our network of about 250 computers. Logs say those 250 computers have had around 2500 ssh-logins yesterday, so numerically we have either one looping sshd per 1200h or one per 500 ssh-logins. Looping xdm processes also occur in roughly similar frequency. Theory says this frequency should increase with less reliable networks than the plain wired ethernet we have, but I'm not sure how frequently Kerberos would be used in such setups. Ciao, Alexander Wuerstlein. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
