Re: Debugging TSIG signed nsupdate problems - Specifically a logging question

2024-05-28 Thread Erik Edwards via bind-users
In the dnssec.log file I only found references to normal key rotation. Adding the section for update_security and running at trace 99 didn't provide _any_  update_security log output, nor did it provide any extra output to the update log. even when running in single combined log format I coul

Re: Debugging TSIG signed nsupdate problems - Specifically a logging question

2024-05-27 Thread Erik Edwards via bind-users
Please allow me to refocus this thread to the original question. I'm asking about the logging facility with respect to the "update" section of code in ISC's bind9 product. Yes, I understand update-policy choices/errors will generate the REFUSED response. _I'm only asking about the logging f

Re: Debugging TSIG signed nsupdate problems

2024-05-27 Thread Mark Andrews
> On 27 May 2024, at 16:06, Erik Edwards via bind-users > wrote: > > Hello Mark & List, > > Thank you for responding, I'm running bind-9.18.26-1.fc40.x86_64 and using > nsupdate 9.16.27-Debian to send the updates, using rndc Version: 9.18.26. > > I'm issuing commands through rndc to set the

Re: Debugging TSIG signed nsupdate problems

2024-05-26 Thread Erik Edwards via bind-users
Hello Mark & List, Thank you for responding, I'm running bind-9.18.26-1.fc40.x86_64 and using nsupdate 9.16.27-Debian to send the updates, using rndc Version: 9.18.26. I'm issuing commands through rndc to set the trace level to 99 -> "rndc trace 99". rndc seems to work correctly in all other

Re: Debugging TSIG signed nsupdate problems

2024-05-26 Thread Mark Andrews
Start from the beginning. Show the actual configuration (named.conf, K* files, etc.). X out the secret keys. Show the actual commands you are running. Show the actual logs being produced. REFUSED can come from lots of things. Named emits log messages for almost all of them without needing to t

Re: Debugging TSIG signed nsupdate problems

2024-05-24 Thread Erik Edwards via bind-users
algorithm hmac-sha256; named-checkconf -p shows the key with the matching name, algo, and secret. When I mis-configure, change, or typo the secret it returns "BAD SECRET" The error I'm seeing is "REFUSED" on a config that worked until the upgrade. It worked on F36-F39, upgrades were seamless.

Re: Debugging TSIG signed nsupdate problems

2024-05-24 Thread John Thurston
It doesn't answer your original question, but I suggest looking at the 'algorithm' of that key. Might it be a hmac-md5 ? If you 'named-conf -px'   does it appear in the list of keys? -- Do things because you should, not just because you can. John Thurston907-465-8591 john.thurs...@alaska.g

Debugging TSIG signed nsupdate problems

2024-05-24 Thread Erik Edwards via bind-users
How can I set debug level log for update events? I've tried "rndc trace 99" which gives *lots* of information expect for UPDATE REFUSED issues even thought the channel is set to dynamic severity. Is there a different way to get named to generate debug level logs for UPDATE events? I'm runni