Your pihole is misconfigured if it is out of resources with that few a 
number of queries.

If WU sets their TTL short, all consumers are going to query often because 
nobody's caching nameserver (pihole, local, google, or anybody else) will 
have a valid cache.  That is just how DNS works.  Things are working 
'exactly' as WU wants them to work.  They 'want' you to query often.  Do a 
"dig -q=ANY" and you'll see how WU is currently set up on their end:

   - three actual AWS systems speaking ipv4
   - behind an AWS load balancer for both ipv4 and ipv6 queries
   - and both the rapidfire and non-rapidfire WU hostnames point to the 
   load balancer
   - with 60 second TTL on the actual computers they load balance across

If you check your pihole logs, you'll see why you do so many queries per 
day:

   - each query causes 'two' lookups, one for A records, one for AAAA 
   records
   - so doing the math you get the essentially one call/second average you 
   are seeing

Go to your pihole and look at long term data => top lists and tell us how 
many lookups it reports to WU yesterday.

   - I did 576 such lookups here with rapidfire 'off', all resolved via 
   google DNS
   - For the last month the math works out the same per day so checking 
   yesterday is good enough as a check

As I think Greg suggested, you 'could' run a caching nameserver on your 
weewx system to only look up when the caches expire, but I wouldn't bother.

On Thursday, January 4, 2024 at 6:30:40 AM UTC-8 G Hammer wrote:

> I recently regraded my LAN and in doing so forgot to have my WeeWX machine 
> use external DNS. It was pointed to my pihole DNS servers.
>
> Because of that, I see that the WU update is not very well behaved.
> It seems to call for a new DNS lookup every time it fires.
>
> With rapid fire on, that is 85000+ DNS calls daily!
> rtupdate.wunderground.com 85026
> wx.lan 87934
>
> That number of lookups raised warnings on my pihole machines, they want 
> more resources to handle the high volume of calls.
>
> I reconfigured the WeeWX machine to use external DNS, but the issue is 
> just being pushed elsewhere.
>
> Nothing on my network makes that many DNS lookup calls, not even close.
> Every OS I'm aware of has a DNS cache, WeeWX doesn't seem to use it though.
>
> However, WU has a TTL of only 60 seconds, but that is still around 24X 
> less.
>
> Any thoughts?
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"weewx-development" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to weewx-development+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/weewx-development/997117ac-5eec-4986-be46-06f19173dd43n%40googlegroups.com.

Reply via email to