Strange behavior after server migration with admin section (EE 5.4)
News and General
We moved to Rackspace, there were a couple bumpy spots that were strange, but the site itself seems to be running properly.
However, the admin control panel area has some strange behavior.
- Caching of entry lists - and entry changes or adds to a channel don’t show in the list until you refresh the page
- File upload list when inserting images in an entry (RTE) doesn’t update so you can’t see the newly added images.
- Multi-Site: Changing from one site to another also requires a refresh, but doesn’t not always work which causes you to be stuck in one site with no way to get out without clearing caches locally.
- The Channel selection in the Options panel are no longer showing similar channels - so I can’t push from channel A to Channel B like I did previously.
All but #4 seem to be some sort of local caching issue that I can’t seem to understand. Browser version or type doesn’t seem to change anything, multiple machines in different parts of the country having the same issue so it’s not a local setting.
Server: Running the same PHP 7.4 version as previously run, and MariaDb on CentOS. The only change was MariaDb from the previous host. Same EE cache settings with Redis, CloudFlare is handling DNS and only caching images, CSS and JS (no change there either)
UPDATE This is actually happening on the front end as well, I have to refresh the page if I have been there recently to see the new entries. This wasn’t happening before the switch so I am at a loss… I also put Cloudflare on Development mode and it didn’t help any.
Hi Ccbangler,
Thanks for getting in touch about this.
My first thought was Cloudflare as well. On the frontend side of things, if you have another caching add-on etc in play, I’d recommend testing with that off (and clearing caches in the CP).
Beyond that, It sounds like it’s likely environment related. I’m not sure if you have a support plan or not. But if you do, we’d be happy to jump in and see what we can figure out! We’d likely need access to the server as a whole.
Only other thought I have is that it might be worth mentioning in the Slack Group) theres about 7,000 of us in there. I wouldn’t be surprised if someone else had another thought or idea before looking into the environment end of things.
Thanks, -Tom Jaeger
I did get 1-3 narrowed down to something / someone from the previous host adding come aggressive cache directives in htaccess - I didn’t really dive into it because I haven’t modified that file since switching hosts and I didnt have an issue previously. Why it didn’t effect the site previously is beyond me… very strange.
Number 4 is still a mystery - I have to figure that one out still.
Thanks!