[Swan-dev] TCP: retransmit skipped because TCP is handling retransmits

[Andrew Cagney]

On Thu, 19 Aug 2021 at 19:49, Paul Wouters <paul at nohats.ca> wrote:

> On Wed, 18 Aug 2021, Andrew Cagney wrote:
>
> > Subject: [Swan-dev] TCP: retransmit skipped because TCP is handling
> retransmits
> >
> > Is there a bug to track this?
>
> I didn't create a bug because I didn't think there was one. I figured
> all we can do is wait, and that the retransmit-timeout value would
> still be reached and cause the proper failure/tear down to happen.
>

It needs to schedule something some retransmit-timeout seconds into the
future that triggers the dead IKE code.
However it shouldn't schedule individual retransmits (even when it ignores
them).

If nothing else it stops us having to deal with questions about why
--statestatus shows:
    TCP; ... RETRANSMIT in 0.5s
when it should probably show:
    TCP; ... TIMEOUT in 60s



> Paul
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.libreswan.org/pipermail/swan-dev/attachments/20210820/1c2f15b5/attachment.html>