Walt Karas

OK I'll try to evaluate those, but I'm not sure when I'll have time.

> @shinrich Any update on the development of the issue? Susan will busy for a while with personal matters. You can submit a PR related to this issue without it...

I wonder if you used HTTP/2 to origin you would then not care about this.

Any way we could get an Au test for this plugin merged first?

> https://github.com/apache/trafficserver/blob/master/tests/gold_tests/cache/background_fill.test.py Sorry I'm not seeing where/how this tests the plugin.

Should we also KWF parentselectdefs.h ? The parent_select experimental plugin looks to be gone. @SolidWallOfCode ?

> Hmmm yeh why ? This seems like it’d make each binary unnecessarily large too ? Indont know about the dummy thing but this seems bad regardless. > > Why...

[approve ci autest]

Some results from Red Hat Enterprise LInux. ``` --- test_PluginDso --- ... ../src/proxy/http/remap/PluginDso.cc:135:load() plugin 'plugin_v1.so' failed to load: /tmp/sandbox_31leGz/runtime/plugin_v1.so: undefined symbol: pluginThreadContext ... --- test_PluginFactory --- ... ../src/proxy/http/remap/PluginDso.cc:135:load() plugin 'plugin_v1.so'...

``` wkaras ~/REPOS/TS/build/src/proxy/http/remap/unit-tests O$ nm -C -A test_RemapPluginInfo | grep -F mutexAllocator test_RemapPluginInfo:000000000052083c t _GLOBAL__sub_I_mutexAllocator test_RemapPluginInfo:0000000000622400 B mutexAllocator wkaras ~/REPOS/TS/build/src/proxy/http/remap/unit-tests O$ ```