Alright, I get the point. Making the connection more stable isn't going
to work or happen inside of Apache. So my question is probably answered,
you can close this case. (Oh, no case management in place, just a bunch
of comments.)
More off-topic: Mailing list defenders usually can't come up wi
On Mon, Feb 12, 2018 at 10:24 AM, Yves Goergen
wrote:
> One more of those unpractical mailing lists... (I'm already deleting 99% of
> what I receive to ask a single question. This is more than the spam.)
One more of those users who think their issues should be resolved the
way they think...
>
>
One more of those unpractical mailing lists... (I'm already deleting 99%
of what I receive to ask a single question. This is more than the spam.)
Okay, so I'm not going to solve this problem server-side. The client
that gets the error will have to retry on its own more often and even
resend it
If your upgrade procedure is indeed "a second", I'd try to DROP with iptables
SYN packets from proxy to appserver for this period, proxy will likely retry
connection attempts. Depends on many factors, but at least worths a try.
> Where can I request that feature to be considered for adding?
On
From getting zero replies I think this isn't possible with Apache.
Where can I request that feature to be considered for adding? It can
increase the availability of websites. Doesn't that sound interesting?
What solutions do you have for this issue? Or am I just thinking too
small and everybody
Hello,
I've configured Apache as a reverse proxy for my application server.
When the backend server isn't available, Apache quickly responds the
browser with an error 503.
There are short downtimes while I'm upgrading the application server.
But Apache doesn't even care to retry and make the
