benbakelaar

Yep… we were on version 3.2.0.

I just set up a staging site and upgraded to 3.6.2, and it appears to be working now! I guess some JS must have changed between 3.2 and 3.6.2, since you were able to replicate the issue with your own testing which began with 3.6.2?

Btw I do see a sort of phantom / fast load (and close) of something when I click the mobile menu button and the SideMenu appears. I assume that’s the resize event that you mention above. It’s not an issue, just a note that it’s visible upon close inspection.

Tried again, but no luck! I tried turning off “Hijack mobile menu” and setting the SideMenu as the menu in the “Mobile” position/area in theme settings, but then it simply didn’t display SideMenu.

Let me submit this as a bug to Astra theme and see. There is currently no menu set for mobile in Astra, yet it generates an “all pages” menu on its own (which now shows up on live site after you close out SideMenu).

Thanks for your help, will keep you posted!

Hi! We are live now, at njrise.org, and the issue still occurs on mobile device. It can vary when you are on desktop and shrink the width to mobile, sometimes the 2nd “default” mobile menu does not appear. But on my Android device, it still reveals a mobile menu underneath the SideMenu.

Thank you very much, testing now!

Thanks so much for the quick update! I installed it earlier and cleared the server-side cache on the host, but so far it is still loading the double menu for me in Chrome on an Android phone (Motorola). It’s a little hard to clear cache on a mobile browser, so not sure if that’s it… especially if it’s some piece of Javascript that the browser is hanging on to.

Bump with notify

Bump

FINALLY fixed!

I started to narrow down on DNS issues as the cause. (duh)

I checked out /etc/resolv.conf, and was reading articles about editing it. But then I tried some “ping” commands to the sub-domains in question, and the pings worked. So it didn’t appear to be DNS issue.

Then I found this question, which pretty much exactly described the same problem I was having:

https://serverfault.com/questions/380288/curl-request-wont-resolve-to-localhost

No resolution there, but I took another look at the /etc/hosts file. Because I was able to “trick” FeedWordPress into importing my main website’s RSS feed via 127.0.0.1, I decided to add entries for the sub-domains in /etc/hosts under 127.0.0.1 (they were already entered as the external IP).

THIS SOLVED THE PROBLEM!

I did have to restart networking – on Ubuntu Linux, it is /etc/init.d/networking restart.

I hope this info ends up helping someone else some day. Really frustrating. I guess the answer is:

If you are running multiple virtual hosts on an Apache server, and want CURL/file_get_contents and similar functions to be able to access resources via FQDN, you need to associate the sub-domains with the localhost address (127.0.0.1) in your /etc/hosts file.

Yea I’m poking around for support for the FQDN/configuration error. I suppose I might eventually be able to figure it out but I just don’t understand every layer in enough detail to know how to fix or isolate the issue.

Re: my actual issues, I did manage to solve one of them. I tricked the FeedWordPress plugin by using https://127.0.0.1/blog/rss.xml as the feed address, rather than https://www.princetonlibrary.org/blog/rss.xml.

That won’t fix my problem with resource loading on the WP-to-iPad plugin though.

And I wonder if it has something to do with this error when I restart Apache?

root@PPLWebServer:/var/www/community# service apache2 restart
 * Restarting web server apache2                                                apache2: Could not reliably determine the server's fully qualified domain name, using 127.0.0.1 for ServerName
 ... waiting apache2: Could not reliably determine the server's fully qualified domain name, using 127.0.0.1 for ServerName

Also here is the Apache virtualhost configuration:

<VirtualHost *:80>
    ServerAdmin me@myserver
    DocumentRoot /var/www/community/
    ServerName community.princetonlibrary.org
    ServerAlias community
    RewriteEngine On
    RewriteOptions inherit
    CustomLog /var/log/apache2/community.log combined
    Options -Indexes FollowSymLinks
</VirtualHost>

Here are some details on the server from phpinfo:

Apache Version 	Apache/2.2.17 (Ubuntu)
Hostname:Port 	community.princetonlibrary.org:0
User/Group 	www-data(33)/33
Server Root 	/etc/apache2
HTTP_HOST 	community.princetonlibrary.org
SERVER_NAME 	community.princetonlibrary.org
DOCUMENT_ROOT 	/var/www/community/
allow_url_fopen	On	On
allow_url_include	Off	Off
cURL support 	enabled
cURL Information 	7.21.3

In tests using file_get_contents() and a curl() function, I can load https://www.google.com without issue. I assume this means I can load any other domain as well. But if I try to load a URL on the server itself, e.g. community.princetonlibrary.org/bookfestival/wp-content/plugins/wp-to-ipad/images/wpti-logo.png, or a URL on a different sub-domain on the same server, e.g. https://www.princetonlibrary.org/blog/rss.xml, the request times out with the “failed to open stream” error. Both of these URLs exist and can be loaded in a browser.

Obviously the server is having trouble referring to itself. Looking for ideas about why this might be. I think it is related to the htaccess rewrite code. I’ve been working with Apache/PHP/MySQL and WordPress for many years now, but only recently with multi-site. Everything functions normally, just this issue.

The plugins that are actually “causing” this issue are:
– WP-to-iPad (renders an iPad version of site)
– FeedWordPress (allows you to import RSS feeds)

But the problem has nothing to do with the plugins themselves, just the functions they are using to retrieve resources on the server.

>Is it possible to create a network for all my sites?

Yes, generally speaking. Depending on your webhost you may run into certain limitations or issues, but that’s the whole idea of WP multi-site: you can run multiple WP websites from a single codebase and database.

>Must all the websites be in the same hosting?

In the end, yes. Let’s say you have A.com, B.com, C.com, and D.com all on different hosts. You would create a new hosting plan, let’s say on E.com, install WP multi-site, and then migrate all 4 of your other websites into the new WP multi-site installation.

>Should all the websites use the same database?

Yes they will use a single database associated with the WP multi-site installation.

Here are some resources I’ve been poking around as I try to solve my own htaccess rewrite issues.

https://www.cheatography.com/davechild/cheat-sheets/mod-rewrite/

https://httpd.apache.org/docs/2.0/misc/rewriteguide.html

Hmm so you don’t think it is related to the .htaccess file? You tested on a multi-site install with sub-directories? I don’t think it’s the same environment if it’s two separate installed instances of WP regular. I’ve never had issues like this with WP regular, just multisite, which I know is a tricky beast.