> On 18 May 2017, at 12:46, Jim Jagielski <[email protected]> wrote:
>
> Based on feedback from various sessions:
>
> o A new-kind of "hot standby" in mod_proxy which kicks
> in whenever a worker moves out of the pool (ie, doesn't
> wait until all workers are out)... ala a redundant
> hard drive.
>
> o Look into AAA and mod_cache; eg: "bolt in at the end"
>
> o HTTP/2 no longer experimental!
>
> o balancer-manager more scriptable (move to REST for realz?)
>
> o When restarting w/ persistent balancer data, warn if
> config files mtime is newer.
>
> o Warn if the trailing '/'s don't match in ProxyPass/Reverse
> directives (eg: ProxyPass /foo http://www.example.com/foo/ )
>
> All I can recall at present...
o Investigate mod_cache performance when bypassing cache (e.g.
`Cache-Control: no-cache'). In benchmark testing for my
talk,
(1) no mod_cache
vs.
(2) cached responses
vs.
(3) cache enabled but requests specify `Cache-Control: no-cache'
showed that (3) was significantly slower than (1). I would expect
that no-cache would short-circuit/bypass mod_cache and have
results similar to (1), but there is *something* going on in
there. I have not investigated the code...just noted the results.
