You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
with the change from last that set max-cache-ttl to 0, the lancache DNS service is now quite slow. each lookup can take one or two full seconds, and subsequent checks are not cached at all so they also take a second or two. Further more, even a downstream pihole will not cache the hits because the TTL was set to zero.
of the three changes to the options file:
'forward-only' seems fine, and it accomplishes the goal of that commit.
'max-ncache-ttl' doesn't seem to break anything, although the default bind value is 10800 'max-cache-ttl' of 0 is the problem. Per the docs, the default value is 604800, and zero should not be used because it may cause SERVFAIL errors.
I also found this one while doing some testing.
What I don't get is this is not just for the content domains that are 'poisoned'/'enriched' for caching, it's for all of them. I can imagine the content domains have to have a low TTL to ensure traffic is captured quickly, but for regular domains this really slows down normal internet/browsing tasks. Having this set makes any client OS level DNS caching worthless AFAIK.
If I find some time next week I will create a PR for this :)
This issue has been automatically marked as inactive because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.
Describe the issue you are having
with the change from last that set max-cache-ttl to 0, the lancache DNS service is now quite slow. each lookup can take one or two full seconds, and subsequent checks are not cached at all so they also take a second or two. Further more, even a downstream pihole will not cache the hits because the TTL was set to zero.
The change was made with commit #131
of the three changes to the options file:
'forward-only' seems fine, and it accomplishes the goal of that commit.
'max-ncache-ttl' doesn't seem to break anything, although the default bind value is 10800
'max-cache-ttl' of 0 is the problem. Per the docs, the default value is 604800, and zero should not be used because it may cause SERVFAIL errors.
https://bind9.readthedocs.io/en/latest/reference.html#namedconf-statement-max-ncache-ttl
I've personally modified it to 10,000 and it works well.
my recommendation is to leave in the forward only line and strip out the other two. (barring an enhancement to let the user set the cache values)
How are you running the container(s)
docker compose, pulled latest.
version: '2.1'
services:
dns:
image: lancachenet/lancache-dns:latest
env_file: .env
restart: unless-stopped
and in .env
IP address on the host that the DNS server should bind to
DNS_BIND_IP=192.168.3.204
DNS Resolution for forwarded DNS lookups
UPSTREAM_DNS=149.112.112.11;9.9.9.11
I have nothing.
The text was updated successfully, but these errors were encountered: