[03:31:00] I'm pretty tired but I think something is up with https://phabricator.wikimedia.org/T217942 [03:31:11] the code looks okay [03:32:27] * paladox gives Krenair a energy drink :Zp [03:32:31] *:P [03:35:45] it's like PHP hasn't noticed that includes/ServiceWiring.php has changed to pass in those two new parameters [03:35:47] or something [07:11:15] <_joe_> Krenair: oh my, are you using php7 in beta? [07:11:25] <_joe_> if so, I think I know what might have happened [07:11:49] <_joe_> scap does invalidate the opcache in production, but I'm not sure scap configuration in beta does it [07:11:59] <_joe_> I assumed they were the same [07:12:25] <_joe_> we have moved to use opcache.validate_timestamps false [14:46:00] _joe_, I think the default PHP there is HHVM [14:46:11] But I could be wrong about where the web requests are getting sent [14:48:01] I do appear to have the PHP_ENGINE=php7 cookie [14:58:21] noted on the task [15:07:26] <_joe_> Krenair: ok so that could be it [15:07:44] <_joe_> can you check if scap.cfg in deployment-prep has php7-admin-port set or not [15:09:09] deployment-deploy01:/etc/scap.cfg has php7-admin-port: 9181 but only in the wmnet section, which won't apply [15:09:20] <_joe_> hah, I'm an idiot [15:09:37] <_joe_> ok this is at least easy to fix [15:10:50] we can just copy it to under the wmflabs section right? [15:39:01] <_joe_> yeah, or have revalidation happen in labs [15:39:09] <_joe_> which might be more fit to that use-case [16:27:23] might be, at the same time it might be better to have prod and beta behaviour match for this sort of thing