Skip to content

How to log individual upstream DNS failures #363

@supersebbo

Description

@supersebbo

I am trying (with little success) to debug an issue with intermittent DNS resolution failures on my network. Symptom is that clients report "Temporary failure in name resolution" for periods of 30-60 seconds, then start working again.

To me this smells like there is a single resolver in my pool of 20 that is failing to respond, and this heals itself when stubby moves onto the next resolver in the round-robin pool.

However, when I have tested all the upstream resolvers in the pool, they are all working, so it would seem the issue with the upstream resolver is intermittent.

Is there a level of logging that Stubby can output that will help me track down this unreliable DNS resolver?

The only logging I get from Stubby today is when ALL upstream resolvers have failed (which it logs mercilessly).

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions