> On Jan 30, 2017, at 4:17 PM, Jim Jagielski <[email protected]> wrote:
>
>>
>> On Jan 30, 2017, at 4:04 PM, Jacob Champion <[email protected]> wrote:
>>
>> On 01/28/2017 07:22 AM, Jim Jagielski wrote:
>>> In your scenario does "old mode" == "old" Apache or non Apache?
>>
>> "Old" Apache.
>>
>>> My idea was to send FCGI data such that PHP-FPM doesn't use *any*
>>> Apache-related fixups. In other words, httpd sends "what it should"
>>> and PHP-FPM "uses what it receives"
>>
>> Right. My point is that without simultaneous modifications to PHP-FPM, I see
>> no way to do this without breaking deployed use cases, because of how easily
>> the fixup code is triggered. And therefore IMO we shouldn't push that change
>> in 2.4.26, but wait (for a reasonable amount of time) for both sides to
>> agree on the way forward.
>>
>> I'll try to review Eric's patch later this week, since it would help ease
>> the migration pain (and give other users a path forward even with broken
>> FCGI backends).
>>
>
> Looking over fpm_main, Apache is detected iff PHP sees the proxy:balancer
> and/or
> proxy:fcgi prefix. Looking at the logic paths related to 'apache_was_here',
> it looks like it just works around stuff that we *CAN* fix. For example, one
> code path is due to "mod_proxy_fcgi and ProxyPass, apache cannot set
> PATH_INFO",
> but we can, and do now....
Plus, other than the "additional" way of detecting apache_was_here, that
file has been very stable with no real changes:
https://github.com/php/php-src/commits/master/sapi/fpm/fpm/fpm_main.c
since
https://github.com/php/php-src/commit/e6d93a11ad343efdc42315f7f69ed82515c9f374#diff-624bdd47ab6847d777e15327976a9227
