RFR: 8339538: Wrong timeout computations in DnsClient [v5]

[Aleksei Efimov]

On Wed, 11 Sep 2024 16:27:29 GMT, Daniel Fuchs <dfuchs at openjdk.org> wrote:

>> 2 time  is not too high,
>> I have presented, in the comment, a failures with the elapsed time is almost twice the expected time
>> where the elapsed time is 14229 !! which is approx 1.84 * expected timeout
>
> @msheppar with the latest proposed change to set MIN_TIMEOUT to 0, then I believe the `@implNote` you suggested is no longer needed.

To address issue `#3` the UDP part of the `DnsClient` implementation has been changed to track the timeout left for each server for the unfulfilled timeout left for each server. An additional retry attempt was added to compensate for left timeouts.
The exception handling has been also updated to save observed exceptions as suppressed.
Also, the TCP part of the `DnsClient` has been changed to use nano time source for timeout calculations.

-------------

PR Review Comment: https://git.openjdk.org/jdk/pull/20892#discussion_r1767352955