On Tue, 08 Jan 2008 07:46:03 +0900, John Summerfield wrote: > Michael Schwendt wrote: > > On Mon, 07 Jan 2008 07:46:51 -0500, Neal Becker wrote: > > > >> Something a bit strange here (this is F8 with all updates): > >> > >> /sbin/service named status > >> named: unrecognized service > >> > >> sudo /sbin/chkconfig named --list > >> named 0:off 1:off 2:on 3:on 4:on 5:on 6:off > > > > Don't stop there yet. Does /etc/init.d/named exist? Is the > > Oh, Michael Michael > > chkconfig doesn't find it if it doesn't exist. Without testing (or fetching and reading the source as it is a binary not a script) I was unable to tell when and how it examines the files and symlinks and to what extend it strictly needs a chkconfig-specific header in the initscript. And even when testing with a non-existant "blah" service, maybe chkconfig would have given different results only when finding run-level symlinks. I wanted Neal to dig a little deeper and find the problem himself. My small set of questions lead to enough details to locate the problem. > [summer@potoroo ~]$ sudo /sbin/chkconfig named --list > error reading information on service named: No such file or directory > [summer@potoroo ~]$ > > > bind package installed? What do you get for "stat /etc/rc5.d/S*named*"? > > How is that relevant to the immediate problem? I don't see how the > presence of that symlink matters to the command that failed. With a result like $ /sbin/service blah status blah: unrecognized service on F8 and following "grep unrec /sbin/service", my first assumption was not that the initscript was lacking -x permission, but that it was missing. > Besides, chkconfig has already checked those symlinks, that's how it > reported: > named 0:off 1:off 2:on 3:on 4:on 5:on 6:off It checked the symlinks, but didn't report their timestamps and targets. -- fedora-test-list mailing list fedora-test-list@xxxxxxxxxx To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-test-list
