On 4/18/08, Bill Nottingham <notting@xxxxxxxxxx> wrote: > David Mack (davidwmack@xxxxxxxxx) said: > > This is still broken. The idea that ypbind should be able to function > > even if it can only use loopback is absurd. It has to contact a > > server. This not only breaks ypbind, it breaks autofs (the maps come > > from NIS) and it will break everything else that depends on NIS data. > > If you use the current rawhide NetworkManager & haldaemon, setting > NETWORKWAIT in /etc/sysconfig/network will give you the behavior > you desire. (You may need to run chkconfig --resetpriorities on > haldaemon and NetworkManager to move them earlier.) No, it doesn't. I tried that after seeing the comments to that effect in the bug report and before I sent out that email. I ran resetpriorities on both haldaemon and NetworkManager as well as adding NETWORKWAIT=1 in /etc/sysconfig/network. No change in behavior on reboot, which is not really surprising, since ypbind runs at rc3.d/S27ypbind and NetworkManager and haldaemon both run at rc3.d/S98. ypbind is long dead when these things get invoked. Dave -- fedora-test-list mailing list fedora-test-list@xxxxxxxxxx To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-test-list
