Re: sortlist usage

2012-06-20 Thread Drunkard Zhang 
2012/6/19 Drunkard Zhang :
> I‘m working for a China ISP, with a very complex network arch. One of
> these complex things is the CDN of ICP returned a bunch of IP
> addresses, which distributed in geographically varies, and some IP
> addresses located way far from us. So I want do some "sortlist" which
> sorts the IP addresses by geographical distance, not considering
> latency and CDN peer's load now.
>
> From ARM 9.9, I get this example:
>
> sortlist {
> // IF the local host
> // THEN first fit on the following nets
> { localhost;
> { localnets;
> 192.168.1/24;
> { 192.168.2/24; 192.168.3/24; }; }; };
> // IF on class C 192.168.1 THEN use .1, or .2 or .3
> { 192.168.1/24;
> { 192.168.1/24;
> { 192.168.2/24; 192.168.3/24; }; }; };
> // IF on class C 192.168.2 THEN use .2, or .1 or .3
> { 192.168.2/24;
> { 192.168.2/24;
> { 192.168.1/24; 192.168.3/24; }; }; };
> // IF on class C 192.168.3 THEN use .3, or .1 or .2
> { 192.168.3/24;
> { 192.168.3/24;
> { 192.168.1/24; 192.168.2/24; }; }; };
> // IF .4 or .5 THEN prefer that net
> { { 192.168.4/24; 192.168.5/24; };
> };
> };
>
> Here's my understanding after some testing:
> 1st level of brace matches client's IP addresses;
> 2nd level of brace matches IP in rrset data;
> 3rd, what's this level used for? Is it just ordering behind 2nd? When
> I added 3rd level of address_match_list, the ordering of rrset just
> failed.
>
> By testing, I found that the 1st level of entry can only be one IP
> range or acl name, sort will fail while 1st level is more than one
> address_match_list. Is this intended or I missed something?
>
> BTW, is 4th or 5th or 6... level of address_match_list supported?
>
> Another thing related: can I use sortlist for determining which
> upstream DNS server to query?
>
> Here's one testing domain name I used:
> 11:11:21 ~ $ dig +short @220.115.251.73 s.kaixin001.com.cn
> s.kaixin001.com.cn.lxdns.com.
> s.kaixin001.z.lxdns.com.
> kx.xdwscache.glb0.lxdns.com.
> 124.14.10.17
> 124.14.10.25
> 124.14.10.18
> 121.251.53.95
> 58.22.135.38
>
> The first 3 IPs located locally, 121.251.53.95 located in Anhui, and
> 58.22.135.38 located in Fuzhou, Fujian. The first 3 IPs returned
> cyclicly, 4th and 5th IP are fixed.

I think there's some trick not mentioned in ARM, like:
* The 1st level of brace can only be ONE element, address_match_list.
bind won't deal with more than one address_match_list;
* You can define more than 2 level of braces, but only configurations
within 2 level actived, others are ignored;
* Misswrite of sortlist config can not be detected, it just failed;

So you can write anything you like with following format:
sortlist {
{ acl-client;
{ acl-client; acl-local-city; acl-local-province; acl-local-contry;
...other-acls;
}; };

configurations out of this frame won't work. Am I right?
___
Please visit https://lists.isc.org/mailman/listinfo/bind-users to unsubscribe 
from this list

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

Interpreting bind logging queries

2012-06-20 Thread Alex 
Hi,

I have a bind-9.8.3 fc16 system and would like to know more about how
logging works. How can I determine whether the results were from the
local cache or it was actually necessary to query a remote server to
return a response?

Given a query log entry such as:

20-Jun-2012 13:23:50.023 queries: client 127.0.0.1#47286: query:
factoryfitpartscorp.info IN A + (127.0.0.1)

How can I determine if this request was made to the server for this
domain or if it had previously been queried and is retrieving it from
a cache?

Can you point me to where I can find information about interpreting query logs?

Thanks,
Alex
___
Please visit https://lists.isc.org/mailman/listinfo/bind-users to unsubscribe 
from this list

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

Compiling and testing on Fedora

2012-06-20 Thread Luther, Dan 
Hi all, 

I've had a major problem with using Fedora Core (10 through 15), when compiling 
and running "make test":

A:System test acl
I:Couldn't start server ns2 (pid=17344)
R:FAIL
S:allow_query:Wed Jun 20 23:21:47 GMT 2012
T:allow_query:1:A
A:System test allow_query
I:Couldn't start server ns2 (pid=17368)
R:FAIL
S:addzone:Wed Jun 20 23:22:01 GMT 2012
T:addzone:1:A
A:System test addzone
I:Couldn't start server ns2 (pid=17393)
R:FAIL
S:autosign:Wed Jun 20 23:22:15 GMT 2012
T:autosign:1:A
A:System test autosign
I:generating keys and preparing zones
I:Couldn't start server ns1 (pid=17734)
R:FAIL
S:builtin:Wed Jun 20 23:22:35 GMT 2012
T:builtin:1:A
A:System test builtin
I:Couldn't start server ns1 (pid=17755)
R:FAIL
S:cacheclean:Wed Jun 20 23:22:49 GMT 2012
T:cacheclean:1:A
A:System test cacheclean
I:Couldn't start server ns1 (pid=17776)
R:FAIL

I'm running the "bin/tests/system/ifconfig.sh up" script, and see the "lo:1" 
through "lo:7" interfaces come up. I don't have this problem on any of my 
Solaris systems, just the Fedora servers. I do have several lo: interfaces 
already defined, and they cannot be removed

Has anyone seen such an issue, and if so, how did you fix it?

Dan Luther
Operations Engineer
Systems Operation Engineering 
Level 3 Communications
One Technology Center, Tulsa OK 74103
p: 918-547-4370
e: dan.lut...@level3.com


___
Please visit https://lists.isc.org/mailman/listinfo/bind-users to unsubscribe 
from this list

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

Re: Compiling and testing on Fedora

2012-06-20 Thread Jeremy C. Reed 
I don't immediately recognize the issue. But hopefully the detailed 
named debugging output is saved. Look for the "*.run" (maybe named.run) 
files.
___
Please visit https://lists.isc.org/mailman/listinfo/bind-users to unsubscribe 
from this list

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

Re: Interpreting bind logging queries

2012-06-20 Thread Mark Andrews 

In message 
, Alex writes:
> Hi,
> 
> I have a bind-9.8.3 fc16 system and would like to know more about how
> logging works. How can I determine whether the results were from the
> local cache or it was actually necessary to query a remote server to
> return a response?
> 
> Given a query log entry such as:
> 
> 20-Jun-2012 13:23:50.023 queries: client 127.0.0.1#47286: query:
> factoryfitpartscorp.info IN A + (127.0.0.1)
> 
> How can I determine if this request was made to the server for this
> domain or if it had previously been queried and is retrieving it from
> a cache?

You can't.  The log entry doesn't record that.
 
> Can you point me to where I can find information about interpreting query log
> s?

The Adminstrators Reference Manual (ARM) full describes these log messages.
 
> Thanks,
> Alex
> ___
> Please visit https://lists.isc.org/mailman/listinfo/bind-users to unsubscribe
>  from this list
> 
> 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: ma...@isc.org
___
Please visit https://lists.isc.org/mailman/listinfo/bind-users to unsubscribe 
from this list

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

RE: Compiling and testing on Fedora

2012-06-20 Thread Shawn Bakhtiar 



Did you turn OFF SELinux? 

prompt>setenforce 0

Then run the test, 

> From: dan.lut...@level3.com
> To: bind-us...@isc.org
> Subject: Compiling and testing on Fedora 
> Date: Wed, 20 Jun 2012 23:33:08 +
> 
> Hi all, 
> 
> I've had a major problem with using Fedora Core (10 through 15), when 
> compiling and running "make test":
> 
> A:System test acl
> I:Couldn't start server ns2 (pid=17344)
> R:FAIL
> S:allow_query:Wed Jun 20 23:21:47 GMT 2012
> T:allow_query:1:A
> A:System test allow_query
> I:Couldn't start server ns2 (pid=17368)
> R:FAIL
> S:addzone:Wed Jun 20 23:22:01 GMT 2012
> T:addzone:1:A
> A:System test addzone
> I:Couldn't start server ns2 (pid=17393)
> R:FAIL
> S:autosign:Wed Jun 20 23:22:15 GMT 2012
> T:autosign:1:A
> A:System test autosign
> I:generating keys and preparing zones
> I:Couldn't start server ns1 (pid=17734)
> R:FAIL
> S:builtin:Wed Jun 20 23:22:35 GMT 2012
> T:builtin:1:A
> A:System test builtin
> I:Couldn't start server ns1 (pid=17755)
> R:FAIL
> S:cacheclean:Wed Jun 20 23:22:49 GMT 2012
> T:cacheclean:1:A
> A:System test cacheclean
> I:Couldn't start server ns1 (pid=17776)
> R:FAIL
> 
> I'm running the "bin/tests/system/ifconfig.sh up" script, and see the "lo:1" 
> through "lo:7" interfaces come up. I don't have this problem on any of my 
> Solaris systems, just the Fedora servers. I do have several lo: interfaces 
> already defined, and they cannot be removed
> 
> Has anyone seen such an issue, and if so, how did you fix it?
> 
> Dan Luther
> Operations Engineer
> Systems Operation Engineering 
> Level 3 Communications
> One Technology Center, Tulsa OK 74103
> p: 918-547-4370
> e: dan.lut...@level3.com
> 
> 
> ___
> Please visit https://lists.isc.org/mailman/listinfo/bind-users to unsubscribe 
> from this list
> 
> bind-users mailing list
> bind-users@lists.isc.org
> https://lists.isc.org/mailman/listinfo/bind-users
  ___
Please visit https://lists.isc.org/mailman/listinfo/bind-users to unsubscribe 
from this list

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