|

»

freebsd-standards

standards/137173: `uname -n` incorrect behavior

View: New views

7 Messages — Rating Filter: Alert me

	standards/137173: `uname -n` incorrect behavior by Garrett Wollman-2 :: Rate this Message: Reply to Author \| View Threaded \| Show Only this Message The following reply was made to PR standards/137173; it has been noted by GNATS. From: Garrett Wollman <wollman@...> To: Andy Kosela <akosela@...> Cc: freebsd-gnats-submit@... Subject: standards/137173: `uname -n` incorrect behavior Date: Mon, 27 Jul 2009 17:19:36 -0400 <<On Mon, 27 Jul 2009 13:26:24 GMT, Andy Kosela <akosela@...> said: > Currently `uname -n` prints the name of the system (FQDN) to standard output. I believe this is incorrect behavior according to IEEE Std 1003.1. > -n > Write the name of this node within an implementation-defined communications network. What makes you think that the behavior of "uname -n" does not match this description? -GAWollman _______________________________________________ freebsd-standards@... mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-standards To unsubscribe, send any mail to "freebsd-standards-unsubscribe@..."
	Re: standards/137173: `uname -n` incorrect behavior by Andy Kosela-5 :: Rate this Message: Reply to Author \| View Threaded \| Show Only this Message The following reply was made to PR standards/137173; it has been noted by GNATS. From: Andy Kosela <akosela@...> To: wollman@... Cc: freebsd-gnats-submit@... Subject: Re: standards/137173: `uname -n` incorrect behavior Date: Tue, 28 Jul 2009 10:17:29 +0200 Garrett Wollman <wollman@...> wrote: > <<On Mon, 27 Jul 2009 13:26:24 GMT, Andy Kosela <akosela@...> said: > > > Currently `uname -n` prints the name of the system (FQDN) to standard output. I believe this is incorrect behavior according to IEEE Std 1003.1. > > > -n > > Write the name of this node within an implementation-defined communications network. > > What makes you think that the behavior of "uname -n" does not match > this description? Hi Garrett, All UNIX systems I got access to prints only hostname without the domain information (same as 'hostname -s'). Is this some historical peculiarity of FreeBSD? I see it uses KERN_HOSTNAME which is indeed FQDN. On top of that common sense tells me that "node within an implementation-defined communications network" is just a node name, and not a full domain name information. What you think? --Andy _______________________________________________ freebsd-standards@... mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-standards To unsubscribe, send any mail to "freebsd-standards-unsubscribe@..."
	Re: standards/137173: `uname -n` incorrect behavior by Garrett Wollman-2 :: Rate this Message: Reply to Author \| View Threaded \| Show Only this Message The following reply was made to PR standards/137173; it has been noted by GNATS. From: Garrett Wollman <wollman@...> To: Andy Kosela <akosela@...> Cc: freebsd-gnats-submit@... Subject: Re: standards/137173: `uname -n` incorrect behavior Date: Tue, 28 Jul 2009 16:29:07 -0400 <<On Tue, 28 Jul 2009 08:40:03 GMT, Andy Kosela <akosela@...> said: > All UNIX systems I got access to prints only hostname without the domain > information (same as 'hostname -s'). Legacy Unix implementations used the UUCP name, which was completely unconnected to any other notion of the host's name. Few people use UUCP any more, and in any case, they are free to set their hostname to something other than an FQDN if they want. > On top of that common sense tells me that "node within an > implementation-defined communications network" is just a node name, and > not a full domain name information. What you think? In what way is an FQDN not a node name? -GAWollman _______________________________________________ freebsd-standards@... mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-standards To unsubscribe, send any mail to "freebsd-standards-unsubscribe@..."
	Re: standards/137173: `uname -n` incorrect behavior by John Hein :: Rate this Message: Reply to Author \| View Threaded \| Show Only this Message The following reply was made to PR standards/137173; it has been noted by GNATS. From: John Hein <jhein@...> To: bug-followup@..., akosela@... Cc: Subject: Re: standards/137173: `uname -n` incorrect behavior Date: Tue, 28 Jul 2009 14:16:17 -0600 It seems at least part of this report is inaccurate. I tested on linux (Fedora 10) and it seems to behave the same as freebsd... [root@foo ~]# hostname foo.example.com [root@foo ~]# uname -n foo.example.com _______________________________________________ freebsd-standards@... mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-standards To unsubscribe, send any mail to "freebsd-standards-unsubscribe@..."
	Re: standards/137173: `uname -n` incorrect behavior by Andy Kosela-5 :: Rate this Message: Reply to Author \| View Threaded \| Show Only this Message The following reply was made to PR standards/137173; it has been noted by GNATS. From: Andy Kosela <akosela@...> To: jhein@..., bug-followup@... Cc: Subject: Re: standards/137173: `uname -n` incorrect behavior Date: Wed, 29 Jul 2009 09:00:52 +0200 John Hein <jhein@...> wrote: > It seems at least part of this report is inaccurate. > I tested on linux (Fedora 10) and it seems to behave > the same as freebsd... > > [root@foo ~]# hostname > foo.example.com > [root@foo ~]# uname -n > foo.example.com I tested it on SLES and Debian only. You are right, RedHat behaves differently. It seems there is no consensus about -n behavior even in the Linux camp. --Andy _______________________________________________ freebsd-standards@... mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-standards To unsubscribe, send any mail to "freebsd-standards-unsubscribe@..."
	Re: standards/137173: `uname -n` incorrect behavior by Andy Kosela-5 :: Rate this Message: Reply to Author \| View Threaded \| Show Only this Message The following reply was made to PR standards/137173; it has been noted by GNATS. From: Andy Kosela <akosela@...> To: wollman@... Cc: freebsd-gnats-submit@... Subject: Re: standards/137173: `uname -n` incorrect behavior Date: Wed, 29 Jul 2009 09:32:53 +0200 Garrett Wollman <wollman@...> wrote: > In what way is an FQDN not a node name? Yes, 'uname -n' comes from UUCP times. I think our discussion boils down to nodename vs hostname, which in legacy UNIX can have different values. For me it seems natural that nodename (coming from old UUCP) should be identical to hostname without the full domain name information. Is out there some standard defining it and explaining how nodename (UUCP) convention should be applied to hostname (ARPA, NFS) convention? --Andy _______________________________________________ freebsd-standards@... mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-standards To unsubscribe, send any mail to "freebsd-standards-unsubscribe@..."
	Re: standards/137173: `uname -n` incorrect behavior by Garrett Wollman-2 :: Rate this Message: Reply to Author \| View Threaded \| Show Only this Message The following reply was made to PR standards/137173; it has been noted by GNATS. From: Garrett Wollman <wollman@...> To: Andy Kosela <akosela@...> Cc: freebsd-gnats-submit@... Subject: Re: standards/137173: `uname -n` incorrect behavior Date: Wed, 29 Jul 2009 11:52:34 -0400 <<On Wed, 29 Jul 2009 09:32:53 +0200, Andy Kosela <akosela@...> said: > information. Is out there some standard defining it and explaining how > nodename (UUCP) convention should be applied to hostname (ARPA, NFS) > convention? No, that's why the POSIX specification leaves it completely implementation-defined. You have to remember that this value originally came from a "struct utsname" which had fixed-length (eight?-byte) fields, and the name was compiled into the kernel. (Hence UUCP names like "ihnp4", "mhuxu", and so on.) -GAWollman _______________________________________________ freebsd-standards@... mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-standards To unsubscribe, send any mail to "freebsd-standards-unsubscribe@..."