There's more deep rooted issues - https://github.com/heiglandreas/authLdap/issues/194 . The package itself is wrong and unfortunately wp admin suggests it as an update. Breaking a lot of things.

@girish Nope, tried to debug it. Tried to even repair it. In the end a reboot fixed the problem, but that was pretty weird.

@jodumont I think just having any recommended snippets in the docs would suffice for most, no? Then just copy/paste any you want to use using the File Manager (after making a backup in case it breaks something).

@dreamcatch22 You might have to restart the app after making php.ini changes. Apart from that, you can install something like https://wordpress.org/plugins/php-info-wp/ which should give you phpinfo output to verify is the extensions are enabled.

Hm surfer will send cache-control: public, max-age=0 so I am not sure why it would cache as such, the etag also should have changed if the source content was updated.

@moocloud_matt Thanks for the heads up. The project is still early enough where I can make the switch.

The amazing self solving topic

Right, for the moment it accepts everything till 403

I agree with both of @girish's assessments of the file for the time being. But that doesn't mean that when Cloudron does support multi-site in the future, we can't assess multisite constants. So thanks for bringing it up! It lead me to that really interesting and eye-opening wp-config.php on Github.

@ketchalegend I see. With the above htaccess, I got curl -X POST -H "Authorization: Bearer 1234567890" https://wp.domain.com/wp-json/jwt-auth/token {"success":false,"statusCode":403,"code":"jwt_auth_invalid_token","message":"Wrong number of segments","data":[]} So looks like the route is working.

Follow up from the customer: "The issue here turned out to be that in Wordpress, WP Rocket caching plugin was used. This plugin automatically starts to preload the cache of each page once something in the site has been updated. The preload itself causes some stress on the CPU and maybe some other processes. Turning off the plugin, the products were sent for less than 2 mins." They are working with the WP Rocket team to find a workaround.

@Lonk said in WordPress Developer vs LAMP PHP 7.4: could test it in the same conditions yes, I got that after, I just simply did'nt actived it inside WordPress but at the end it is doing what the WordPress_Dev container doing manually.

@d19dotca Wordpress is my favorite and Cloudron is my new development passion so I’m happy to troubleshoot WP stuff to find edge cases that Cloudron may need to account for. I am very for making WP the best it can be on Cloudron. ️ I’m happy I’ve been able to help troubleshoot with you!

@imc67 said in Wordpress (unmanaged) CRON interrupts product loading: If you set it to 5 then WP site check is complaining that there are missed events. Because either WP or plugins can create 1 minute tasks. Yes, cron scheduled Wordpress can be scheduled down to the minute. But they're expected to be missed. The default WP-Cron interval is "whenever the next person visits the site, run all the scheduled events that are in the past and haven't been run" (the one that Cloudron rightfully uses the WP-CLI and disables the native WP-Cron). That's pretty horrible, but that's what Wordpress installs expect the environment to be at default. So, the WP Site Check says that's notable, that wp-crons don't get run at the minute they're specified? If the WP Health Status Page actually says that, then Wordpress devs clearly think 1 minute (or a hugely popular site) is the best interval.

Yeah, making them read only makes the most sense. It's not like URLs change that often but you certainly wouldn't want to change it in WP or Cloudron would freak.

@girish said in Is it possible to configure the Apache Logs in Wordpress (Developer)?: Currently, you cannot change this, no. You have to use the LAMP app if you want to change the apache configs. In the LAMP app, there is an app.conf which gives you control over logging. You'd expressed interest in more fine-control over logging in another thread. If that's still true, I'll deal with the Apache over-Logging for now looking forward to a time where we figure out a solution for a way to best configure logs for the end-user. Going to LAMP would just feel like I'd be contributing less and less to the Wordpress (Developer) and (Managed) codebase(s) and that's my real goal. To make Wordpress (Developer) the best it can be in Cloudron. So maybe, one day, those access logs will be configurable.

@girish said in Can I send the Wordpress Debug Log to the Cloudron Logs: I agree we should visit this at some point for better logging This fixes the Wordpress "custom log path issue". But I do think we should revisit in the future about other apps and how we might be able to symlink the logs together. Anyway, you got me 75% there on this one, so thank you for that! ️

@girish That makes sense why it wasn’t in mine. I’m planning to reinstall anyway for LDAP support so I’ll let your install script add the custom COOKIEHASH for me. ️

I'm gonna look into this to get a definitive answer on if WP-DEBUG errors get piped to Cloudron's logs (which would be super useful). I'll post back guys (and if it doesn't, I'll see if I can make it do so and share the code).

Can confirm while converting the CRON Job to support multisite, it's still 1 minute and has been since I've been working on it.