Re: question about CNAME

2009-03-12 Thread Stephane Bortzmeyer
On Thu, Mar 12, 2009 at 12:32:35PM +0800,
 tzq tang gelenbert...@gmail.com wrote 
 a message of 132 lines which said:

  I think I should explain the question more clearly,

You need first to learn about email. The superior to sign  is here
to *quote* what you respond to. Do not use it for your own text or it
will be quite difficult to read.

  email.xx  A   12.2.23.4 http://email.xx.xxx.com/

:-) The format of the right-hand side will probably not be accepted by
BIND :-)

  and in another DNS server I add an CNAME record as follows:
  email  CNAME  email.xx.zone.com http://email.xx.xxx.com/.
  but it dosen't work.who can tell me why ?

Invalid syntax. BIND will tell you that at startup.

___
bind-users mailing list
bind-users@lists.isc.org
https://lists.isc.org/mailman/listinfo/bind-users


Re: XFR quota setting?

2009-03-12 Thread Chris Thompson

On Mar 11 2009, Niall O'Reilly wrote:


On Wed, 2009-03-11 at 16:41 -0500, Peter Laws wrote:

Seriously, though, what is the default quota and is it actually
configurable?


Sorry.  No idea what or whether.  RTFM time for us both!  8-)


Which ought to take less than the time to post here :-(

options transfers-out setting, default 10. 


Although the '... quota reached' messages alarmed me at first,
the following message announcing recovery (confirmed by manual
checking of SOA serials) reassured me enough to make me feel 
that no further attention was needed.


If it's occurring a lot, you could have stuck or nearly-stuck
transfers going on. rndc status will tell you how many. You may
need to adjust max-transfer-time-out/max-transfer-idle-out
rather than transfers-out.

--
Chris Thompson
Email: c...@cam.ac.uk
___
bind-users mailing list
bind-users@lists.isc.org
https://lists.isc.org/mailman/listinfo/bind-users


Two outgoing queries for each incoming query

2009-03-12 Thread My Name
Is this possible with 9.6.0-P1 or do I need to change the code (all ideas
where to start are welcome, I haven't looked at the code yet).

I want to setup a forwarder and each incoming query (in fact only A or )
should be sent to two different upstream servers.

Joe
___
bind-users mailing list
bind-users@lists.isc.org
https://lists.isc.org/mailman/listinfo/bind-users

Re: Two outgoing queries for each incoming query

2009-03-12 Thread R Dicaire
On Thu, Mar 12, 2009 at 7:43 AM, My Name mylistuser1...@gmail.com wrote:
 I want to setup a forwarder and each incoming query (in fact only A or )
 should be sent to two different upstream servers.

Why?
___
bind-users mailing list
bind-users@lists.isc.org
https://lists.isc.org/mailman/listinfo/bind-users


RE: XFR quota setting?

2009-03-12 Thread Laws, Peter C.
On Mar 11 2009, Niall O'Reilly wrote:

On Wed, 2009-03-11 at 16:41 -0500, Peter Laws wrote:
 Seriously, though, what is the default quota and is it actually
 configurable?

 Sorry.  No idea what or whether.  RTFM time for us both!  8-)

Which ought to take less than the time to post here :-(

options transfers-out setting, default 10.


Yes, but if you're grepping for xfer, you'll never find that.  

Thanks, though, this is what I wanted to know.  

___
bind-users mailing list
bind-users@lists.isc.org
https://lists.isc.org/mailman/listinfo/bind-users


Re: Two outgoing queries for each incoming query

2009-03-12 Thread Gregory Hicks

 Date: Thu, 12 Mar 2009 13:43:44 +0200
 Subject: Two outgoing queries for each incoming query
 From: My Name mylistuser1...@gmail.com
 To: bind-users@lists.isc.org
 
 Is this possible with 9.6.0-P1 or do I need to change the code (all
 ideas where to start are welcome, I haven't looked at the code yet).

 I want to setup a forwarder and each incoming query (in fact only A
 or ) should be sent to two different upstream servers.

Why?  Bind already does this.  If there are two (or more) servers
serving a zone, it will already query all of them for the initial
query.  However, it uses the answer from the server that has the
fastest response time.

Regards,
Gregory Hicks
-
Gregory Hicks   | Principal Systems Engineer
| Direct:   408.569.7928

People sleep peaceably in their beds at night only because rough men
stand ready to do violence on their behalf -- George Orwell

The price of freedom is eternal vigilance.  -- Thomas Jefferson

The best we can hope for concerning the people at large is that they
be properly armed. --Alexander Hamilton

___
bind-users mailing list
bind-users@lists.isc.org
https://lists.isc.org/mailman/listinfo/bind-users


Peaceful coexistence with Windows domain

2009-03-12 Thread Peter Laws
Our environment includes a couple of AD servers.  They serve DNS to PCs 
using AD (but not all PCs).  They allow DDNS for clients and slave the rest 
of our environment's zones.  For some reason, they *forward* every other 
query to us, but never mind that.  Look it up your own damn ... well, never 
mind.


At any rate, we don't actually delegate their zone to them.  This causes 
problems, as you can imagine.


I'm told that the reason we're doing things this way is that we don't want 
any of those internal addresses to be queried by the unwashed masses 
lurking outside our perimeter.


So my thought was, well, let's delegate the zone to the AD servers.  Since 
they are already ACLed (or whatever MS calls it), no one will be able to 
see their records off-campus but on-campus folks will be able to 
(finally) resolv addresses in that zone regardless of where they point 
(internally) for DNS.


Except that they need an MX record for that zone.

So adding the NS record to delegate the zone to them properly meant that no 
one could see the MX from the outside (since the MS-DNS is ACLed).


If I dump the delegation and make an MX record in the master, mail will be 
OK, but then no one can query records in that zone because it's not 
actually delegated unless they point at MS-DNS.


We thought of slaving that zone on the master, but then we run into 
security, who doesn't want any of that internal information leaked out. 
No problem, since we're slaving the zone, we'll pop an ACL on it.  Problem 
solved!  Hurray.


Except for that MX record.

Once you delegate a zone, you *delegate* the zone.  The MX is invisible.


So my requirements are to 1) allow that MX record to be seen outside, 2) 
allow any host in our environment to be able to query names in any zone 
regardless of which system they point at for DNS, and 3) not have any 
records in that zone be visible outside save for that MX.


I'm assuming that switching our configuration to use views would help, but 
we'd like to avoid that, at least for now.


Any quick fixes?

I checked, and per the MS-People, MS-DNS cannot put ACLs on particular 
records.  Neither can BIND, so no surprise there.


Which rock do I need to look under?

--
Peter Laws / N5UWY
National Weather Center / Network Operations Center
University of Oklahoma Information Technology
pl...@ou.edu
---
Feedback? Contact my director, Craig Cochell, cra...@ou.edu. Thank you!
___
bind-users mailing list
bind-users@lists.isc.org
https://lists.isc.org/mailman/listinfo/bind-users


Re: Two outgoing queries for each incoming query

2009-03-12 Thread Mark Andrews

In message 200903121454.n2cesvel019...@metis.hicks-net.net, Gregory Hicks wri
tes:
 
  Date: Thu, 12 Mar 2009 13:43:44 +0200
  Subject: Two outgoing queries for each incoming query
  From: My Name mylistuser1...@gmail.com
  To: bind-users@lists.isc.org
  
  Is this possible with 9.6.0-P1 or do I need to change the code (all
  ideas where to start are welcome, I haven't looked at the code yet).
 
  I want to setup a forwarder and each incoming query (in fact only A
  or ) should be sent to two different upstream servers.
 
 Why?  Bind already does this.  If there are two (or more) servers
 serving a zone, it will already query all of them for the initial
 query.  However, it uses the answer from the server that has the
 fastest response time.

No.  It will query multiple servers in turn as needed to
satisfy queries.  RTT estimates are most effective with
infrastructure zones as those are the ones queried most
often.  Named tries to minimize the number of queries it
makes.
 
 Regards,
 Gregory Hicks
 -
 Gregory Hicks   | Principal Systems Engineer
 | Direct:   408.569.7928
 
 People sleep peaceably in their beds at night only because rough men
 stand ready to do violence on their behalf -- George Orwell
 
 The price of freedom is eternal vigilance.  -- Thomas Jefferson
 
 The best we can hope for concerning the people at large is that they
 be properly armed. --Alexander Hamilton
 
 ___
 bind-users mailing list
 bind-users@lists.isc.org
 https://lists.isc.org/mailman/listinfo/bind-users
-- 
Mark Andrews, ISC
1 Seymour St., Dundas Valley, NSW 2117, Australia
PHONE: +61 2 9871 4742 INTERNET: mark_andr...@isc.org
___
bind-users mailing list
bind-users@lists.isc.org
https://lists.isc.org/mailman/listinfo/bind-users


Re: Peaceful coexistence with Windows domain

2009-03-12 Thread Kevin Darcy

Peter Laws wrote:
Our environment includes a couple of AD servers. They serve DNS to PCs 
using AD (but not all PCs). They allow DDNS for clients and slave the 
rest of our environment's zones. For some reason, they *forward* every 
other query to us, but never mind that. Look it up your own damn ... 
well, never mind.


At any rate, we don't actually delegate their zone to them. This 
causes problems, as you can imagine.


I'm told that the reason we're doing things this way is that we don't 
want any of those internal addresses to be queried by the unwashed 
masses lurking outside our perimeter.


So my thought was, well, let's delegate the zone to the AD servers. 
Since they are already ACLed (or whatever MS calls it), no one will be 
able to see their records off-campus but on-campus folks will be 
able to (finally) resolv addresses in that zone regardless of where 
they point (internally) for DNS.


Except that they need an MX record for that zone.

So adding the NS record to delegate the zone to them properly meant 
that no one could see the MX from the outside (since the MS-DNS is 
ACLed).


If I dump the delegation and make an MX record in the master, mail 
will be OK, but then no one can query records in that zone because 
it's not actually delegated unless they point at MS-DNS.


We thought of slaving that zone on the master, but then we run into 
security, who doesn't want any of that internal information leaked 
out. No problem, since we're slaving the zone, we'll pop an ACL on it. 
Problem solved! Hurray.


Except for that MX record.

Once you delegate a zone, you *delegate* the zone. The MX is invisible.


So my requirements are to 1) allow that MX record to be seen 
outside, 2) allow any host in our environment to be able to query 
names in any zone regardless of which system they point at for DNS, 
and 3) not have any records in that zone be visible outside save for 
that MX.


I'm assuming that switching our configuration to use views would help, 
but we'd like to avoid that, at least for now.


Any quick fixes?

I checked, and per the MS-People, MS-DNS cannot put ACLs on particular 
records. Neither can BIND, so no surprise there.


Which rock do I need to look under?

You can't really have it both ways, either the zone is visible or it 
isn't. The MX record is at the very top of the zone -- what is often 
called the apex -- so it's not like you can delegate it and put an ACL 
on the delegated zone.


It seems a little inconsistent to me that your Security folks don't mind 
the MX record being exposed but they're paranoid about the zone's other 
internal stuff leaking out.


I don't see any way to to meet the requirements you've been given, as 
you've described them. There's nothing really unusual about having 
separate internal-versus-external versions of your namespace, and the 
delegation structure doesn't need to be the same for both. So why not 
just stick with an undelegated AD zone in the external version? You can 
still delegate it in the internal version, if it makes you feel better.


- Kevin

___
bind-users mailing list
bind-users@lists.isc.org
https://lists.isc.org/mailman/listinfo/bind-users


RE: Peaceful coexistence with Windows domain

2009-03-12 Thread Ben Bridges
 If I dump the delegation and make an MX record in the master, mail will be
 OK, but then no one can query records in that zone because it's not
 actually delegated unless they point at MS-DNS.

Is there a reason why you can't point all of your internal hosts (AD and 
non-AD) at your AD's for resolution?
 



From: bind-users-boun...@lists.isc.org on behalf of Peter Laws
Sent: Thu 3/12/2009 4:51 PM
To: bind-us...@isc.org
Subject: Peaceful coexistence with Windows domain



Our environment includes a couple of AD servers.  They serve DNS to PCs
using AD (but not all PCs).  They allow DDNS for clients and slave the rest
of our environment's zones.  For some reason, they *forward* every other
query to us, but never mind that.  Look it up your own damn ... well, never
mind.

At any rate, we don't actually delegate their zone to them.  This causes
problems, as you can imagine.

I'm told that the reason we're doing things this way is that we don't want
any of those internal addresses to be queried by the unwashed masses
lurking outside our perimeter.

So my thought was, well, let's delegate the zone to the AD servers.  Since
they are already ACLed (or whatever MS calls it), no one will be able to
see their records off-campus but on-campus folks will be able to
(finally) resolv addresses in that zone regardless of where they point
(internally) for DNS.

Except that they need an MX record for that zone.

So adding the NS record to delegate the zone to them properly meant that no
one could see the MX from the outside (since the MS-DNS is ACLed).

If I dump the delegation and make an MX record in the master, mail will be
OK, but then no one can query records in that zone because it's not
actually delegated unless they point at MS-DNS.

We thought of slaving that zone on the master, but then we run into
security, who doesn't want any of that internal information leaked out.
No problem, since we're slaving the zone, we'll pop an ACL on it.  Problem
solved!  Hurray.

Except for that MX record.

Once you delegate a zone, you *delegate* the zone.  The MX is invisible.


So my requirements are to 1) allow that MX record to be seen outside, 2)
allow any host in our environment to be able to query names in any zone
regardless of which system they point at for DNS, and 3) not have any
records in that zone be visible outside save for that MX.

I'm assuming that switching our configuration to use views would help, but
we'd like to avoid that, at least for now.

Any quick fixes?

I checked, and per the MS-People, MS-DNS cannot put ACLs on particular
records.  Neither can BIND, so no surprise there.

Which rock do I need to look under?

--
Peter Laws / N5UWY
National Weather Center / Network Operations Center
University of Oklahoma Information Technology
pl...@ou.edu
---
Feedback? Contact my director, Craig Cochell, cra...@ou.edu. Thank you!
___
bind-users mailing list
bind-users@lists.isc.org
https://lists.isc.org/mailman/listinfo/bind-users


___
bind-users mailing list
bind-users@lists.isc.org
https://lists.isc.org/mailman/listinfo/bind-users

Re: question about CNAME

2009-03-12 Thread tzq tang
thanks for your response.I do this test between two intranet machine and
each of them has a local IP.10.0.0.13,additionally the both domain are in
the same DNS SERVER 10.0.0.13 ,the zone file as follows:
ZONE 1:
$ORIGIN .
$TTL 86400  ; 1 day
test.comIN SOA  test.com. www.test.com. (
19970229 ; serial
10 ; refresh (10 seconds)
10 ; retry (10 seconds)
30 ; expire (30 seconds)
30 ; minimum (30 seconds)
)
NS  localhost.
A   10.0.0.13
$ORIGIN test.com.
1   PTR localhost.
email   CNAME  email.tzqian.com.

ZONE 2:

$ORIGIN .
$TTL 86400  ; 1 day
tzqian.com  IN SOA  tzqian.com. support.tzqian.com. (
1997022706 ; serial
10 ; refresh (10 seconds)
10 ; retry (10 seconds)
30 ; expire (30 seconds)
30 ; minimum (30 seconds)
)
NS  localhost.
A   10.0.0.13
$ORIGIN tzqian.com.
1   PTR localhost.
email   A   10.0.0.1


On Thu, Mar 12, 2009 at 7:55 PM, Scott Haneda talkli...@newgeo.com wrote:

 Can you just post your zone file, copy and paste, and do not change any
 data in it, use real IP's.  Or at least obfuscate them in a way that makes
 sense to the rest of us.  There really is not a lot of sense in obfuscating
 a public data store anyway.

 Here is an example:

 Server #1
 email.serverIN  A  192.168.1.100

 Server #2
 email   IN  CNAME  email.server.yourzone.com.

 I am still not sure I understand your issues, you need to show us both
 zones, and tell us each zones host name.

 On Mar 11, 2009, at 7:28 PM, tzq tang wrote:

  I think I should explain the question more clearly,I am wondering how
 cname reflect the address,so I do a test in my server
 to add an A record as follows (zone.com):
 email.xx  A   12.2.23.4
 and in another DNS server I add an CNAME record as follows:
 email  CNAME  email.xx.zone.com.
 but it dosen't work.who can tell me why ?


 --
 Scott * If you contact me off list replace talklists@ with scott@ *


___
bind-users mailing list
bind-users@lists.isc.org
https://lists.isc.org/mailman/listinfo/bind-users

Re: Trouble publishing dkim via nsupdate

2009-03-12 Thread Mark Andrews

In message 20090313031347.ga19...@csy.ca, Shane W writes:
 Hey all,
 
 I am trying to publish a dkim record in a signed dynamic
 zone using nsupdate.  My query looks like the below but
 nsupdate is having none of it, giving formerr.  Can anyone
 see an obvious error with this query:  Pasting the entry
 directly into the zone (freeze/thaw) does work but then the
 record doesn't get signed.
 
 nsupdate:
 zone csy.ca
 update delete continuum._domainkey.csy.ca any
 update add continuum._domainkey.csy.ca 86400 txt k=rsa\; t=y\; p=MIGfMA0GCSq
 GSIb3DQEBAQUAA4GNADCBiQKBgQDGDqQOjvR2NkesUp+rMl164OdruvyT/hcvwWpPJVZZpYJ7C0rU
 FoZeGdIsi0Riv8wbMd0YspPEfXEslt+neNBTp+nGtkbzpV23PnVwxaqaCpUOZtc7LN2BTKLnpQATL
 30JJE6LwafHPmM5I9S6y1pBQBV9KLdBuxG4+xlIwQf6HwIDAQAB
 send

Remove the any from the delete command.
update delete continuum._domainkey.csy.ca


 output with -d
 Reply from SOA query:
 ;; -HEADER- opcode: QUERY, status: NOERROR, id:  11757
 ;; flags: qr aa rd ra ; QUESTION: 1, ANSWER: 1, AUTHORITY: 4, ADDITIONAL: 4
 ;; QUESTION SECTION:
 ;csy.ca.ANY SOA
 
 ;; ANSWER SECTION:
 csy.ca. 86400   IN  SOA continuum.ns.csy.ca. hostmast
 er.csy.ca. 207 14400 900 2419200 3600
 
 ;; AUTHORITY SECTION:
 csy.ca. 86400   IN  NS  dme6.ns.csy.ca.
 csy.ca. 86400   IN  NS  dme7.ns.csy.ca.
 csy.ca. 86400   IN  NS  continuum.ns.csy.ca.
 csy.ca. 86400   IN  NS  dme5.ns.csy.ca.
 
 ;; ADDITIONAL SECTION:
 dme5.ns.csy.ca. 86400   IN  A   63.219.151.12
 dme6.ns.csy.ca. 86400   IN  A   64.246.42.203
 dme7.ns.csy.ca. 86400   IN  A   205.234.170.139
 continuum.ns.csy.ca.3600IN  A   70.71.3.27
 
 Found zone name: csy.ca
 The master is: continuum.ns.csy.ca
 Sending update to 70.71.3.27#53
 Outgoing update query:
 ;; -HEADER- opcode: UPDATE, status: NOERROR, id:   2080
 ;; flags: ; ZONE: 1, PREREQ: 0, UPDATE: 2, ADDITIONAL: 0
 ;; ZONE SECTION:
 ;csy.ca.ANY SOA
 
 ;; UPDATE SECTION:
 continuum._domainkey.csy.ca. 0  ANY ANY
 continuum._domainkey.csy.ca. 86400 ANY  TXT k=rsa\; t=y\; p=MIGfMA0GCSqG
 SIb3DQEBAQUAA4GNADCBiQKBgQDGDqQOjvR2NkesUp+rMl164OdruvyT/hcvwWpPJVZZpYJ7C0rUF
 oZeGdIsi0Riv8wbMd0YspPEfXEslt+neNBTp+nGtkbzpV23PnVwxaqaCpUOZtc7LN2BTKLnpQATL3
 0JJE6LwafHPmM5I9S6y1pBQBV9KLdBuxG4+xlIwQf6HwIDAQAB
 
 
 Reply from update query:
 ;; -HEADER- opcode: UPDATE, status: FORMERR, id:   2080
 ;; flags: qr ; ZONE: 0, PREREQ: 0, UPDATE: 0, ADDITIONAL: 0
 
 Thanks,
 Shane
 ___
 bind-users mailing list
 bind-users@lists.isc.org
 https://lists.isc.org/mailman/listinfo/bind-users
-- 
Mark Andrews, ISC
1 Seymour St., Dundas Valley, NSW 2117, Australia
PHONE: +61 2 9871 4742 INTERNET: mark_andr...@isc.org
___
bind-users mailing list
bind-users@lists.isc.org
https://lists.isc.org/mailman/listinfo/bind-users