Hi
Can I set the $_SERVER vars somewhere manually?
please consult the documentation for Apache / mod_php. This is entirely something in the responsibility of the web server.
Hi
Can I set the $_SERVER vars somewhere manually?
please consult the documentation for Apache / mod_php. This is entirely something in the responsibility of the web server.
Hi
as expected the value inside $_SERVER['HTTP_HOST'] is incorrect, instead it is provided inside $_SERVER['HTTP_X_FORWARDED_HOST']. Unfortunately I cannot help with that, as I know next to nothing about Apache and how one would go on changing the configuration for those Dockerfiles (in fact they look uncessarily complex to me, I prefer the stock php:apache / nginx + php:fpm Dockerfiles).
Hi
Where does the BB get localhost:8905from?
it's taken from the HTTP_HOST value as provided by the Webserver (in your case the Apache in Docker). You might want to add a file with a <?php phpinfo(); inside the Docker container and provide me a link to it (you can send a conversation). I'll have a look at the values and may be able to suggest a fix.
Hi
That's really weird: https://localhost:8905
this indicates an issue passing along the original Host header to the Apache in Docker. I am not proficient in configuring Apache, but you might want to check whether RequestHeader set Host "my.domain" (from the configuration you posted above) is the correct way to do this.
Hi
can you try dumping a request and response using curl: curl -v http://example.com?
Hi
what exactly is the issue? While I have Docker at hand I don't have an Apache 2 Reverse Proxy at hand and I might be able to pinpoint the issue if you explain what does not work.
Hi
remove line 10 and change the loop in line 12 to: while (($laravelUser = $statement->fetchArray())). You need to call fetchArray once per row, not once per query.
Hi
Jupiter I successfully received the files. From a quick look they are looking good. I'll look deeper at them at a later point.
Hi
How do I get the files to you? I just tried to attach them to a conversation, but got a message saying the files were too big.
Depending on the size you can either mail it to me at duesterhus@woltlab.com or upload it on your webspace and send me the link. I'll inform you once I've downloaded it so you can delete it. And of course it will not be shared and will be deleted once I finished working with it.
Hi
Although I lost some data like member passwords and polls, most stuff was migrated to my new woltlab forum. Thank you!
I'm aware of some issue with the polls, but could not fix it yet, because I don't had a proper database export to reproduce the issue.
Would you consider providing your source database of the vBulletin 5 if you still have it for me to fix the issues with the polls and passwords? While it would not provide you with a benefit after the the fact it would prevent others running into the same issues. If you're lucky I might even be able to provide you with a database query to fix the passwords.
Hi
They never made any significant changes to the URLs, these links are still valid. Tim Düsterhus might know more
Unfortunately I don't.
As it's now been a month since I changed over to the Woltlab software, would you still recommend I amend the .htaccess file as recommended in that guide?
If correctly done this cannot harm, you still might be able to lessen the impact of the migration. But you must make sure that the redirects are correct
Hi
I just checked the functionality in both a WCF 2.1 and WSC 3. It worked fine in both versions. Please check that the link has not been mangled / cut off by your mail program.
Hi
could it be that some kind of antivirus software “helpfully” clicked the links before showing the in order to check them for malware? Could it be that the email was delayed and the link expired?
Hi
TimWolla brought up https://www.percona.com/doc/pe…online-schema-change.html and https://githubengineering.com/…migration-tool-for-mysql/, both sites are about online schema changes that do not require a full table lock. I've never done this myself nor can I tell you if they work properly, but I thought it makes sense to share it with you.
be extra cautious about the Percona one, it most likely breaks, because in MySQL the combined behaviour of FOREIGN KEYs and TRIGGERs is pretty broken. In any case: Make sure that your backups are good, before proceeding with something like that.
Hi
I was not able to determine a possible solution based on looking at the code alone and without that output I cannot further act on this issue. Thus I marked it as „Cannot reproduce“.
Hi
are you running a dedicated server? Could you please provide me with the output of SHOW ENGINE InnoDB STATUS?
Hi
take a look at the cacheQuotes method: https://github.com/WoltLab/WCF…ghter.class.php#L255-L263
You would perform the proper highlighting inside the anonymous function.
Hi
part of the issue is solved (the 416 is gone), but it should be discussed whether images on the same domain should be fetched using the Image Proxy.
Hi
Are you saying that ANY user logging in will trigger ALL queued notifications?
Correct. To be absolutely correct: Even guests and Google will trigger the email sending for all the emails that are due.
Is there any negative effect from running that cron job every, say, 30 seconds? Presumably if there is no queue to process, the resource usage is negligent?
You can run it every 30 seconds just fine.
Hi
independently from SoftCreatR I fixed the issue in a similar fashion in git: https://github.com/WoltLab/WCF…fbed7bc7a9ea031c683566573