Hi @mrwordpressuk,
I take it you tried the steps highligted in the provided ticket such as checking site URLs in WordPress configuration etc.?
I checked our inbox too to see if you’ve sent us a diagnostic report but couldn’t see one, so in order to see your configuration and whether our services can connect to your site it’d be good to start there. Can you send one to wftest @ wordfence . com? You can find the link to do so at the top of the Wordfence > Tools > Diagnostics page. Then click on “Send Report by Email”. Please add your forum username where indicated and respond here after you have sent it.
NOTE: It should look as follows – Screenshot of Tools > Diagnostic > Send by Email
Thanks,
Peter.
Yes I checked if the wordfence IPs are blocked or not. They are not blocked. I’ve sent the diagnostics email from wordfence just now.
P.s. I’m using lightsail (bitnami stack) with a cloudfront distribution in front of it.
P.p.s scans work fine when they are manually initiated but not the system initiated ones. N.B. I have moved wp-cron to a linux cron job that runs every 5mins – I don’t know if that makes a difference.
Hi @mrwordpressuk, thanks for sending your diagnostic and letting me know about the cron change.
I did look into this and reply yesterday, but I’m now guessing my account became signed out between opening the ticket and hitting submit without me noticing so sadly I’ve lost the message.
The specific error you’re seeing doesn’t crop up much across our customer base but looking into another case shows that something was making a request that’s not an http OR https request, but we usually need detailed PHP or server error logs matching timestamps of the scan failures to know what exactly was making that request.
What’s encouraging though is that the previous case I referenced was fixed by a server cron job being switched to cURL. It certainly seems to be related to the way crons are run so let me know how this change goes for you.
Thanks,
Peter.
Thanks for getting back to me. Yes that logout issue happens to the best of us 🙂
So not sure what I should do exactly to introduce a curl expression – I simply copy and pasted a trick to reduce server load by moving the wordpress cron to the system cron like so:
*/5 * * * * su daemon -s /bin/sh -c “cd /path/wordpress; /path/php/bin/php -q wp-cron.php”
How should i break this up or is there a specific curl line i can add to just trigger wordfence?
To be frank, it actually used to work. If I could think of an event it would be that I put AWS CDN in front of the lightsail instance. LIghtsail used to have its own ssl cert but it didnt play well with CDN so i removed it. The ssl cert is on the CDN now. CDN and lightsail are communicating over http. The issue is wordfence doesn’t like to run the scan by itself over http but it works fine when its manually initiated. Do you still suggest an additional line in crontab for curl’ing some wordfence command – if so what would that be?
Hi @mrwordpressuk,
DISABLE_WP_CRON and ALTERNATE_WP_CRON are disabled, so you should be running a cron job on the server to process wp-cron.php. In the case I mentioned before, they were running that server cron job via PHP. Your last message highlights a cron job trying to run wp-cron on the command line but we recommend instead creating one that runs via cURL to see if that fixes it.
If the HTTP(S) error still occurs after changing how you run cron, Wordfence > Tools > Diagnostics > Debugging Options > Enable Debugging Mode and sending us a scan log to wftest @ wordfence . com (with your forum username in the subject line) after the next failed scan might help.
Thanks,
Peter.
