Forum Replies Created

Viewing 5 replies - 1 through 5 (of 5 total)
  • Thread Starter Andres

    (@andresb)

    Clorith, I can’t thank you enough for this. I’m importing everything with wp-cli, but in a secondary WP copy that will hold the archive.

    Ipstenu: The main blog is a fast-moving beast and I can’t stop it for enough time to allow for copy, import and SQL substitution + moving images, etc. The way I did it ends up with a dynamic front and a semi-static archive, allowing for design changes and stuff, but 99% of the time pregenerated via caching plugins.

    Thread Starter Andres

    (@andresb)

    Yes! Thanks for your answer. I figured this out this morning.
    I somehow thought the setting would impact WP’s internal blog resolution/redirection, now I know that either setting only affects the “new site” form for future additions.

    But my problem still stands:
    I set domain.com/blogname as URL and path on the site settings (update siteurl and home, etc.).
    Then I go to domain.com/blogname and WP redirects me to domain.com/check-out-blogname-our-new-blog, a post on the main blog.

    I refreshed (saved) the permalink structures at both blogs (/year/month/post-name), cleaned any rewrites/redirections in the nginx config, disabled caching plugins, etc. I’m positive I’m hitting /index.php every time and WP should do its magic, but it’s not happening.

    Does this mean that the blog at domain.com/ needs a /blog path in front of it to avoid conflicts between blogs vs posts URLs?

    I think I ran out of things to examine.

    Thanks in advance.

    Thread Starter Andres

    (@andresb)

    OK, I’m updating this as I’m seeing there are many more responses here and around the web than when I started looking. I’m experiencing this problem on and off since a couple of years ago. I gave up around august last year.
    I’m using the Yoast SEO plugin, but with the option to cleanup query strings OFF.
    I tried the two main cache plugins and some homebrew solutions (caching proxy on the same server, etc), to no avail.
    I truncated the domain_mapping_logins table expecting to delete repeated records that interfered with the queries. The problem stays.

    What I’ve seen in sites that use this: remote login allows to have users logged in on mapped domains. While the login occurs at blog1.network.com, the users are shown as logged in in blog1.com, the mapped domain, and the comment forms show their registered name and url.
    If you look at the code it’s simply a cookie exchange between the subdomain and the mapped domain via a javascript-injected page reload.

    Thread Starter Andres

    (@andresb)

    To expand: I tried to write a dump with all variables every time it happens, nothing -unusual- found. I tried extending the 120 seconds to more time just in case it was deleting logins too soon. I tried redirecting to the login page but that gets the user into a redirect loop because he’s correctly logged-in.

    I’m wary of deleting the 40K records in the database to start fresh because I don’t want to piss off our readers by killing their logins.

    Any recommendation will be very welcome.

    Thanks again,
    Andres

Viewing 5 replies - 1 through 5 (of 5 total)