On Sun 20/Oct/2024 13:23:18 +0200 Douglas Foster wrote:
How many Tree Walk pilot implementations do we have?
I have just one:
https://www.tana.it/svn/zdkimfilter/trunk/src/treewalk.c
it queries all domains (up to MAX_SUBDOMAINS) in one shot and then examines any
record found.
How do those implementations handle DNS errors during the Tree Walk?
They are logged.
How do those implementations plan to implement overrides if the Tree Walk does
not return the optimal organizational domain?
On hard error the filter either passes or blocks with 4xx code, depending on
configuration.
How much volume has been evaluated using the Tree Walk algorithm (message
count, unique domain count, evaluated domain pair count)?
N/A (small anyway)
When Tree Walk is necessary, how do results compare to PSL?
* same result, same organization domain
* same result, different organizational domain
* different result, different organizational domain
Finally, when the results are different, how often is the PSL correct and how
often is the Tree Walk correct? Do we have any previously-unknown PSL errors
exposed by the Tree Walk?
In several months, I only found one From: domain where Tree Walk differs from
PSL _and_ alignment differs. Log: "Tree Walk from newsletter.nero.com finds
newsletter.nero.com, while PSL had nero.com: SPF identifier
newsletter6.nero.com becomes not aligned this way". I noticed them once and
had no reply.
Best
Ale
--
_______________________________________________
dmarc mailing list -- [email protected]
To unsubscribe send an email to [email protected]
