Lots of retried queries and slow resolution

Help
1

Please follow the below template, it will help us to help you!

installed pihole via diet-pi distro as a vmware VM. running on a 4-core v3 xeon server with 28gb of ram. vm is outfitted with 1 cpu and 1GB of ram. vm network is bridged. server utilization is ~30 at any given time with ~8gb of available memory. gateway is an eero mesh system. tried with and without homekit router enabled. custom DNS set to pihole IP. pihole has a static IP configured. I'm getting ~20-30k queries every 1 minutes according to pihole charts. Also get ALOT (75%) of retried queries. usually pages of retried queries before a successful forward to upstream DNS. Have tried single and multiple upstream DNS with no difference. Am I jsut hitting the limit of performance on a VM? i can't imagine considering this would work fine on a raspi.

At my wits end... thanks,

Expected Behaviour:

Expecting website and hostname resolution to respond quickly

Actual Behaviour:

Opening new websites fails with DNS_PROBE_FINISHED_BAD_CONFIG for ~1 minute at which point it starts working. Things like streamed videos fail to start but will work after several minutes.

Debug Token:

https://tricorder.pi-hole.network/nskt3tadrl/

2

No.

Look in /var/log/pihole.log and see if the resolution failures are limited to one or a few of your eight upstream DNS servers. Post several representative lines here.

3

Hi!

my /var/log/pihole.log was surprisingly empty. had only these two lines... nothing in pihole.log.1

May 18 00:18:42 dnsmasq[6978]: failed to send packet: Resource temporarily unavailable
May 18 00:22:27 dnsmasq[6978]: failed to send packet: Resource temporarily unavailable

Just a few corrections to my post (did not proof read).

  • i'm getting 20-30k requests every 10 minutes
  • server cpu utilization is only ~10-30%
  • affects both wifi clients and wired clients and even sites that worked earlier will stop working and have to wait again for a new lookup.

I've since turned off custom DNS on my eero gateway but still seeing the slow response via some clients i've hardcoded DNS to the pihole.

I'm still seeing lots and lots of "retried" status queries.

Thanks!

4

Just a quick update....

the veracity of the Retried queries really got me wondering about some old DUP ping behaviour I've lived with for years. My server is has dual NICs which I've teamed in windows server. I never figured out how to fix the DUP pings so I've literally had it like this for almost 10 years with no real erroneous behaviour. I decided to search for a solution and came across a single post suggesting disabling the windows "Routing and Remote Access" service. It sounds important but the post said they didn't see any ill effects of disabling it.

So I disabled it and Pings finally look normal again and the number of retried queries immediately drops from dozens per second to maybe only a dozen or so over 3-4 minutes.

it's late and i haven't been able to reproduce the slow dns resolution since I disabled the windows service. Will monitor more tomorrow

5

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.