Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

"No SOA found" error when trying to create a (regular) record with API #4705

Closed
hekmon opened this issue Nov 23, 2016 · 6 comments
Closed

"No SOA found" error when trying to create a (regular) record with API #4705

hekmon opened this issue Nov 23, 2016 · 6 comments

Comments

@hekmon
Copy link

hekmon commented Nov 23, 2016

  • Program: Authoritative
  • Issue type: Bug report

Short description

The title is actually not 100% accurate. The bug only occurs if you try to delete the SOA (which fails) and then try to create a record. The bug is also persistent : even if you drop the zone and recreate it, creating a record still won’t work, the SOA error message will pop in {"error": "No SOA found for domain 'domain.net.'"}. Funny thing is : it only last for 60s. If you try again after 60s, it works.

Environment

  • Operating system: Debian Jessie
  • Software version: v4.0.1
  • Software source: deb http://repo.powerdns.com/debian jessie-auth-40 main
  • Backend : MySQL or SQLite

Steps to reproduce

  1. Create a zone
  2. Create a test record (this one should work)
  3. Try to delete the SOA (will fail as expected)
  4. Create a test record (this one will never work)

Repeat the step [4]:

  • before 60s have elapsed => does not work
  • after 60s have elapsed => work as expected

Expected behaviour

The record should be created without the SOA error message.

Actual behaviour

Trying to delete the SOA will prevent further attempt to register or modify (REPLACE rrset) a new record during the next 59 seconds even if the zone is dropped and respawned.

Other information

Deleting the SOA is not something anybody would want : we actually not encountered the bug in production but with our unit tests. It was not easy to isolate and reproduce it but once done, we wrote a small bash script reproducing this issue. It is available here :
https://gist.github.com/Hekmon/6f3be24dff4d33306818b3e421e9aec7

@zeha
Copy link
Collaborator

zeha commented Nov 23, 2016

Thanks for the detailed bug report.

I've quickly tried to reproduce this with current git master, but it appears to work here.
Can you try with the packages available from https://repo.powerdns.com/debian/dists/jessie-auth-master/ ?
(Note that they will have a version number lower than 4.0.)

@zeha
Copy link
Collaborator

zeha commented Nov 23, 2016

Don't bother, I've reran the test a few times and now get the error.

@Habbie Habbie added this to the auth-4.2.0 milestone Nov 9, 2017
@dachinat
Copy link

Any solutions yet?

@Habbie
Copy link
Member

Habbie commented Jan 24, 2020

We've had some caching improvements around the API, can you please check with 4.3.0-alpha1 or master?

@Habbie Habbie modified the milestones: auth-4.3.0, auth-4.4.0 Feb 14, 2020
@Habbie
Copy link
Member

Habbie commented Nov 9, 2020

ping

@hekmon
Copy link
Author

hekmon commented Nov 13, 2020

I have re run the old script in the first post which used to trigger the bug but it seems the issue is not here anymore, closing the issue :)

@hekmon hekmon closed this as completed Nov 13, 2020
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

4 participants