hardme commented on issue #8775:
URL: https://github.com/apache/apisix/issues/8775#issuecomment-1415552404
> > > > in the routing information
> > >
> > >
> > > By saying "in the routing information", what did you mean? Cache DNS
records is a normal behavior for an API Gateway, for the sake of performance.
> >
> >
> > That is, from the routing information matched by apisix in the cache, we
output the matching routing information in the log, and found that when there
is an IP of the upstream node in the routing information, it will not query
DNS, but will directly use the routing information. IP, even if this IP has
expired
>
> APISIX will parse the domain when it matches the route for the current
request, and the upstream nodes' information will be inserted into the object
(which is reachable from the route object). But the resolve is executed each
time APISIX gets the matched route (if there are domains in the upstream node
list).
>
> Whether to use a cached DNS record is decided by the DNS client. So it's
essential to see why the DNS client of APISIX still uses an expired result. If
I remember correctly, the cache TTL is get from the DNS record.
We have made the following configuration, canceled the dns cache, and we set
the TTL returned by the DNS server to be 5s, I am not sure whether our
configuration has an impact on this
```yaml
dns:
validity: ~
```
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
