On Tue, Jun 28, 2016 at 6:41 AM, Jim Jagielski <j...@jagunet.com> wrote:
> I am thinking of a T&R today... Anyone see or know of any
> reasons for not doing so?
>
Of the changes we just backported, there is one side effect, Jens wasn't
imagining things. From this query...
grep -E "^[ \t]*enable_.*=" configure | grep -v -E
"(yes|no|shared|static|module_default)"
I was looking for non-standard things, found some additional expected
cases, e.g.
enable_dumpio=most
enable_echo=maybe-all
... so these are valid (did not see 'all', or 'few', fwiw).
Discovered these explicit overrides, only one group concerns me...
enable_dtrace=$enableval
enable_so=$ac_cv_define_APR_HAS_DSO
enable_so=$enable_so
enable_session_cookie=$session_mods_enable
enable_session_crypto=$session_mods_enable_crypto
enable_session_dbd=$session_mods_enable
enable_lbmethod_byrequests=$proxy_mods_enable
enable_lbmethod_bytraffic=$proxy_mods_enable
enable_lbmethod_bybusyness=$proxy_mods_enable
enable_lbmethod_heartbeat=$proxy_mods_enable
enable_unixd=$unixd_mods_enable
enable_dav=$dav_enable
enable_dav_fs=$dav_fs_enable
but we've knocked out proxy_mods_enable as a switch (it only exists
now in co
I suppose this would have been the more accurate toggle, in the first place?
Any reason we would build lbmethods without balancer?
enable_lbmethod_byrequests=$enable_proxy_balancer
enable_lbmethod_bytraffic=$enable_proxy_balancer
enable_lbmethod_bybusyness=$enable_proxy_balancer
enable_lbmethod_heartbeat=$enable_proxy_balancer
Patch committed to trunk, r1750507, to depend-on proxy_balancer
and to use it's build flag as our default (static/shared or 'yes')
