clearing local caches

Scott Haneda talklists at newgeo.com
Wed Jul 15 20:07:19 UTC 2009


On Jul 15, 2009, at 12:59 PM, Dave Sparro wrote:
> Scott Haneda wrote:
>> On Jul 15, 2009, at 12:29 PM, Dave Sparro wrote:
>>> Scott Haneda wrote:
>>>> ... However, I would like to just get DNS response times.
>>>>
>>>> Perhaps take the list of hosts and feed them to a iterative  
>>>> script calling dig, and fish out the response time?  This does  
>>>> add the problem of redirects of course would not be followed, so  
>>>> I would have to pre-fetch all my urls and follow them to get my  
>>>> testing list.
>>>
>>> I don't see how you could call the results from any other method  
>>> "DNS response times."
>>>
>>> If you used a web browser to measure from, you'd be introducing  
>>> all sorts of other latencies.   Delays from the web server  
>>> itself.  The webserver may have to talk to a database to output  
>>> the HTML.  The transfer of the actual HTML code isn't  
>>> instantaneous. (ad that's just off the top of my head).
>> Correct.  So I will end up pulling down the file, extracting the  
>> hostnames, following any redirects, and extracting the resulting  
>> hostnames.  This gives me a nice list of hostnames that I can run  
>> through an iterative loop in dig.
>> I just need to make sure that I am not getting a locally cached  
>> result.  I suspect there is no way to force a non caches result  
>> from the remote ended resolver?
>
> If you aim your dig at a specific DNS server you'll be getting the  
> results from that IP address.  There won't be  any local resolver  
> involved.
>
> If you aren't in control of the remote resolver, there's no way to  
> predict the cache status of your query on the remote side.


Thank you, makes sense, I will give it a shot.
-- 
Scott * If you contact me off list replace talklists@ with scott@ *




More information about the bind-users mailing list