Re: host(1) prints errors to STDOUT

2015-08-13 Thread Craig Skinner
On 2014-10-15 Wed 16:25 PM |, Craig R. Skinner wrote:
 On 2014-10-14 Tue 10:41 AM |, Theo de Raadt wrote:
  Unfortunately host is maintained upstream, in the bind codebase,
  by ISC.
  
  You should file your bug report there, because that is the right way
  to get change into the ecosystem.
  
 
 Submitted, with their GITWEB line number refs.
 
 ISC's bug database is not publicly readable, in order to protect the
 privacy of users who have included identifying information or attached
 logs or crash dumps to their bug reports.
 
 http://www.isc.org/community/report-bug/
 

Update:

On Fri Jul 03 12:29:43 2015, skin...@britvault.co.uk wrote:
 Has this bug been fixed?

Hi again Craig, and sorry for the delay in response.

Here's the thing: a DNS utility either gets a response, or
not.  NXDOMAIN is a response.  Therefore it's not going to
stderr: a successful query was made.

That said, we're aware that people want easy ways to make
distinctions between YXDOMAIN and other successful query
responses.  But at this time it's not a very high priority.

We'll let you know if that changes, and what, if anything,
we decide to do about it.  Thanks


(This thread: http://marc.info/?t=14133048275)



Re: host(1) prints errors to STDOUT

2014-10-15 Thread Craig R. Skinner
On 2014-10-14 Tue 10:41 AM |, Theo de Raadt wrote:
 Unfortunately host is maintained upstream, in the bind codebase,
 by ISC.
 
 You should file your bug report there, because that is the right way
 to get change into the ecosystem.
 

Submitted, with their GITWEB line number refs.

ISC's bug database is not publicly readable, in order to protect the
privacy of users who have included identifying information or attached
logs or crash dumps to their bug reports.

http://www.isc.org/community/report-bug/

-- 
Craig Skinner | http://twitter.com/Craig_Skinner | http://linkd.in/yGqkv7



host(1) prints errors to STDOUT

2014-10-14 Thread Craig R. Skinner
$ host loopy.loo.found.not; print $?
Host loopy.loo.found.not not found: 3(NXDOMAIN)
1

$ host loopy.loo.found.not  /dev/null; print $?
1

$ host loopy.loo.found.not 2/dev/null; print $?
Host loopy.loo.found.not not found: 3(NXDOMAIN)
1


There's a printf at line 429 of /usr/src/usr.sbin/bind/bin/dig/host.c

Line 569's printf may also be going to STDOUT. Maybe others


Successful output to STDOUT:

$ host www.example.org; print $?
www.example.org has address 93.184.216.119
www.example.org has IPv6 address 2606:2800:220:6d:26bf:1447:1097:aa7
0

$ host www.example.org /dev/null; print $?
0



Re: host(1) prints errors to STDOUT

2014-10-14 Thread Theo de Raadt
Unfortunately host is maintained upstream, in the bind codebase,
by ISC.

You should file your bug report there, because that is the right way
to get change into the ecosystem.  We could fix it here, and be
different, then there will be even more problems.

(Unfortunately, I expect them to mumble something like oops
and then backwards compatibility, which basically is the modern way
of of once a bug goes into the ecosystem, we stand behind it and support
it fully.

 $ host loopy.loo.found.not; print $?
 Host loopy.loo.found.not not found: 3(NXDOMAIN)
 1
 
 $ host loopy.loo.found.not  /dev/null; print $?
 1
 
 $ host loopy.loo.found.not 2/dev/null; print $?
 Host loopy.loo.found.not not found: 3(NXDOMAIN)
 1
 
 
 There's a printf at line 429 of /usr/src/usr.sbin/bind/bin/dig/host.c
 
 Line 569's printf may also be going to STDOUT. Maybe others
 
 
 Successful output to STDOUT:
 
 $ host www.example.org; print $?
 www.example.org has address 93.184.216.119
 www.example.org has IPv6 address 2606:2800:220:6d:26bf:1447:1097:aa7
 0
 
 $ host www.example.org /dev/null; print $?
 0