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
Describe the bug
When there is no IPv4 service on the AIL/link, then based on log output of OTBR it looks like it decides to forward the query to "8.8.8.8" by default. This will always fail. The OTBR DNS server then doesn't respond to the DNS query.
In another context, query handling by "127.0.0.1" was observed (see log example in "output" section).
This also seems strange - would the query be sent to the local server itself? Or does this mean the query gets sent via an API to the host processor? It is not clear if the query is then handled at all locally, or not.
To Reproduce
Git commit id: Using the ot-reference-release BR and dongle (as client).
Network topology: Ethernet connection to a switch, no IPv6/IPv4 router on the link
Expected behavior
DNS server, seeing that there's no upstream DNS server reachable, immediate replies with ServFail (2) RCode.
This seems the appropriate response, see e.g. article.
Console/log output
Can be added if needed.
Here is the example where the query goes to "127.0.0.1":
May 10 16:47:13 raspberrypi otbr-agent[2337]: 00:01:31.157 [I] DnssdServer---: Received query from fd00:db9:0:0:6296:678:3037:3f4c
May 10 16:47:13 raspberrypi otbr-agent[2337]: 00:01:31.157 [I] DnssdServer---: Upstream query transaction 0 initialized.
May 10 16:47:13 raspberrypi otbr-agent[2337]: 00:01:31.157 [C] P-Resolver----: No domain name servers found in /etc/resolv.conf, default to 127.0.0.1
May 10 16:47:13 raspberrypi otbr-agent[2337]: 00:01:31.158 [I] P-Resolver----: Forwarded DNS query 0x589730 to 0 server(s).
Additional context
During cert tests, it was found that the reference BR DNS server/resolver sometimes doesn't reply even when it received the request. We'd like to avoid such cases where possible i.e. let the DNS resolver behave more according to specs.
The text was updated successfully, but these errors were encountered:
Describe the bug
When there is no IPv4 service on the AIL/link, then based on log output of OTBR it looks like it decides to forward the query to "8.8.8.8" by default. This will always fail. The OTBR DNS server then doesn't respond to the DNS query.
In another context, query handling by "127.0.0.1" was observed (see log example in "output" section).
This also seems strange - would the query be sent to the local server itself? Or does this mean the query gets sent via an API to the host processor? It is not clear if the query is then handled at all locally, or not.
To Reproduce
ot-reference-release
BR and dongle (as client).Expected behavior
DNS server, seeing that there's no upstream DNS server reachable, immediate replies with ServFail (2) RCode.
This seems the appropriate response, see e.g. article.
Console/log output
Can be added if needed.
Here is the example where the query goes to "127.0.0.1":
Additional context
During cert tests, it was found that the reference BR DNS server/resolver sometimes doesn't reply even when it received the request. We'd like to avoid such cases where possible i.e. let the DNS resolver behave more according to specs.
The text was updated successfully, but these errors were encountered: