Following up on this topic:
Wordpress Jetpack: Wpcom Connection Test Error
I’ve had the env SERVER_PORT 443 in my php block and it has worked well but stopped in the last month. I think the last jetpack point upgrade killed it.
Anyone else having issues again?
@mt-dev is yours working? If so, Are you on current jetpack also?
I have SERVER_PORT set at 80 as my backend services are unencrypted. All WP sites are OK and connected to Jetpack. Is it WordPress Site Health that’s reporting the issue? If you’ve made no changes to the Caddy environment, I wonder if opening a thread in the Jetpack forum might be a better way to go.
Here’s the funny thing: all the jetpack traffic and services are fine. It’s only remote
Admin from the wordpress site or wordpress apps. And the wpcom connection test
I’m wondering if I’m suffering the blog token issue. Or - more likely, if latest jetpack is making more assumptions about server ports.
Unless Site Health is reporting it, I doubt it. What exactly is WP Site Health saying?
Wpcom connection failure: 200
Recommends reconnecting - Which I’ve done many times now.
It’s strange stuff. The wordpress iOS app, when trying to fetch posts complains of the malformed answer. This was the same symptom as last year with the server port issue. That’s what makes me think the server port issue is around again. But now the environment variable isn’t enough to fix it.
Not necessarily. That message suggests that access to xml-rpc.php is being blocked. Jetpack needs access to this file to operate properly.
That’s what I thought as well but using the tester on their support page shows it works.
It sounds like you’re doing all the right things. So, if I’m clear on what you’re saying, Tools > Site Health in your WP instance says there is a problem, but the Jetpack Debug tool says everything is fine? Have you also worked through the Jetpack troubleshooting tips?
I migrated to caddy 2.3.0 (from 2.1.1) and updated my php from 7.4.8 to 7.4.16 last night also. I wanted to be sure the trailing slash on xml reqs fixed in 2.2 wasn’t it.
After several hours of compile (homebrew hates me), the issue is the same:
All extended tests on site health pass (jetpack connected, blog token, sync, etc) except the wpcom connection.
In the iOS app, it shows me a somewhat verbose error:
Been scouting that error num and best I can figure is either cloudflare or my caddy.conf
Every other test I run is fine so I’ll get jetpack and cloudflare to look. They can’t blame versions at this point
I notice that you added an entry in the wiki article Example: configure WordPress with a static cache. I must admit to being in awe at the clever stuff done in that article and don’t fully understand it all at this stage.
In relation to the issue you’re having here though, I’m just wondering if you temporarily comment out the import you mentioned in your post, does this problem disappear?
Good thinking, but no effect.
This topic was automatically closed after 30 days. New replies are no longer allowed.
