On 9/15/05, Paul Taylor <PaulTaylor at winn dash dixie dot com> wrote:
> 
> Anyhow, sometime yesterday mini-http died for my captive portal...  The only
> other unusual thing I see on the whole machine are these two messages from
> yesterday afternoon in the System Log:
> 
> 12:32:14 xl0: transmission error: 90
> 12:32:14 xl0: tx underrun, increasing tx start threshold to 120 bytes
> 

those are normal, actually.  As you increase load on some NIC's, xl
included, they basically "self-tune".  On higher network load machines
you'll see this immediately after boot and never again.  On machines
with less load, you'll eventually see it if/whenever it gets to the
point where it's needed.


> 
> Have there been other people with issues with httpd stopping?  (With recent
> betas?)  The machine had only been up and active for about a week prior to
> this problem.
> 

Yes, I believe prior to 1.2b9 there were issues with it dying when
running HTTPS, but that was reported to be fixed in b9.  IIRC...  it
was sometime during the 1.2 betas, I might have my numbers a bit off
(check the release notes).

-Chris