Yes and no. It works, but there are some problems. I know I ran into problems when trying to login using the link in the blog. Posting comments may have done something strange, too. Give it a try and see if you can live with the drawbacks.

I believe the bugs (urm, "non-critical problems") can be avoided, depending on HOW one accesses the blog on the subdomain.

If under "access parameters", one chooses "absolute URL", specifying the subdom (i.e., http://subdom.site.com/) ...

Then selecting the "preferred access type" to be "automatic detection by index.php"

and using "index.php" to launch the blog on the subdom ... then everything should be fine.

*i think* (your smileage may vary)

Hope this helps.

I've done more with testing using multiple domains (and I haven't done much with that.) I should really test this out sometime. :oops:

More information, as I recently folded an ( old, separate-subdomain install of b2evo, but sharing one DB ) into v1.6.

What a mine field!

First ... Easiest was Absolute URL (e.g., http://sub-dom.mainDom.com/) with "Automatic detection by index.php". (Then putting an index.php file in the root folder of sub-dom.mainDom.com. I modded the index.php require statement to a hard-coded reference to the mainDom main-inc.php file).

It's important to use the sub-dom ROOT folder as the kicking-off point. (I first tried a folder or two deeper, which yielded a blog, but the nav_links() tried to point to http://mainDom.com/folder1/folder2/index.php?blah=blah ... instead of the subDom ... odd).

I got a stub file to work (but I ended up having to put the stub file in the root folder of BOTH the subDom and the mainDom to satisfy something, I can't remember what, now) :|

Lastly ... permalinks failed for "nice urls" (which is kind of a deal-breaker for us, considering the number of links we have to our main blog, which has been using the extra-path info setting for some time). As a temporary work-a-round, i added

<?php $Settings->set('links_extrapath','0');       // Turn off clean urls ?>

to the _main.php template for the subDom skin, which effectively turns-off nice urls for just the subdom blog.

So ... I got the thing working, but it seems to me that there remains work to bring subDom support up to snuff. I'm sure that the dev team will get this stuff ironed out for the beta release.

Hope this helps.

Thanks for the answers...
I am almost there.

stk - I get an error and I think I have narrowed it down to something you mentioned in your post:

First ... Easiest was Absolute URL (e.g., http://sub-dom.mainDom.com/) with "Automatic detection by index.php". (Then putting an index.php file in the root folder of sub-dom.mainDom.com. I modded the index.php require statement to a hard-coded reference to the mainDom main-inc.php file).

My error is on line 20 of my index.php

How do I hard-code reference to the mainDom main-inc.php file as you metioned?

Line 20 in my index.php reads this:

require_once dirname(__FILE__).'/evocore/_main.inc.php';

What exactly would I change? (sorry I am fairly new at all this...)
Thanks in advance for your help.

p3bundles,

require_once basically means "load this file once" ... and the dirname(__FILE__) is a fancy way of saying (start with the directory name of THIS file) ...

at least, that's my warped way of understanding it.

Problem is, your index file is on the subDom and the main.inc.php file that you REALLY want to load is on the mainDom.

So, I literally, changed :

<?php require_once dirname(__FILE__).'/evocore/_main.inc.php'; ?>

to:

<?php require_once '/home/mysite/public_html/dirTo/evocore/_main.inc.php'; ?>

Once I pointed to the correct mainDom b2evo install, then things started working.

Note: Someone else has suggested requiring the mainDom index.php file, which might be another option. I haven't tested that.

Hope this helps.

PS - I noted, however that my index.php file doesn't have that line you claim around line 20, but rather requires "_blog_main.inc.php" last, around line 67.

stk,

Thanks for the help.
I tried it but for whatever reason it keeps looking into the subdomain folder even though I tell it to go to the main domain...

Additionally it tells me to go to line 108 on the index.php which reads:

require( dirname(__FILE__).'/'.$core_subdir.'_blog_main.inc.php' );

Any thoughts on this?
Thanks.

stk,

I am using Phoenix 1.6 and I use subdomains and want to use your hack to deactivate clean urls with the proposed solution:

Lastly ... permalinks failed for "nice urls" (which is kind of a deal-breaker for us, considering the number of links we have to our main blog, which has been using the extra-path info setting for some time). As a temporary work-a-round, i added
PHP:
<?php Settings->set('links_extrapath','0'); // Turn off clean urls ?>
to the _main.php template for the subDom skin, which effectively turns-off nice urls for just the subdom blog.

I does not work for me..could you provide me with more information how these lines are added to the _main.php?

Update: I solved it. The code should look a little bit different:

<?php $Settings->set('links_extrapath','0'); // Turn off clean urls ?>

marcok,

You're right. My BAD.

In my defense, I was looking at the pretty, colored code that the |php| tag renders so nicely, that I didn't spot the missing "$". :|

(I'll go correct it now).

-stk

stk,

have you figured out a different solution for using b2evolution with two different subdomains? If clean urls would work with both blogs this would be perfect...

Cheers

Marco

marcok,

No, unfortunately, I haven't. But I haven't been trying either, because the old subdom blog I wanted to roll into my main one, lacks much content and is rather "dead". (I barely have time, it seems, to keep the mainDom blog active, let alone add to this subDom blog).

But ... that's my problem, not yours and I only use it as an excuse to say ... I haven't gotten a round tuit yet. ;)

HOWEVER ... I'm HOPING that the answer doesn't lie with me, but rather, with the b2evo developers. v1.6, which is an "alpha" release, was their first foray into subdom support and you can see that it's rather acting like an ALPHA release should (it doesn't all work). I'm speaking way out of turn here, but I'm hoping that the development guys and gals have addressed these issues in the BETA release (v1.8 CVS+ +?) .

Lots 'o folks have been waiting for the BETA release, since some rumor shortly after ALPHA was released, that it would be coming "shortly". (That was the end of November '05 ... some solid 3+ months ago.) :|

Mind you, many had been waiting for ALPHA for a long time too. ;)

Anyway, to make a short story even longer than it needs to be ... I'm hoping (since they're re-writing a lot of core functions) that you'll have your perfect solution once BETA comes out. My advice for now, would be just to use "unclean" URLS with oodles of queries and parameters (uck), till the production version comes out (how much longer can that be?), then, when it does ... remove that bit of code, blammo ... you got clean URLs.

As far as any links, you can take two approaches. (1) Use redirects so they don't break and point to the NEW, CLEAN urls, to keep folks happy ... or be like me and say (2) "Fa ghetta bout it." Let them break! Maybe they'll come 'round, visit again & relink things. Maybe not. Either way, it's less work for me. :|

PS ... If you wanted to have a play, I'd imagine that the fault lies with the function that gens up the permalink (but that's only a guess). Haven't looked at it. Send me a couple round tuits and maybe I can! :D

Cheers,

stk,

thanks for your effort..this is another reason why I like b2evolution. It's not just a piece of software but a bunch of people who care!

Cheers,

Marco

stk wrote:

marcok,

HOWEVER ... I'm HOPING that the answer doesn't lie with me, but rather, with the b2evo developers. v1.6, which is an "alpha" release, was their first foray into subdom support and you can see that it's rather acting like an ALPHA release should (it doesn't all work). I'm speaking way out of turn here, but I'm hoping that the development guys and gals have addressed these issues in the BETA release (v1.8 CVS+ +?) .

Lots 'o folks have been waiting for the BETA release, since some rumor shortly after ALPHA was released, that it would be coming "shortly". (That was the end of November '05 ... some solid 3+ months ago.) :|

Mind you, many had been waiting for ALPHA for a long time too. ;)

As far as any links, you can take two approaches. (1) Use redirects so they don't break and point to the NEW, CLEAN urls, to keep folks happy ... or be like me and say (2) "Fa ghetta bout it." Let them break! Maybe they'll come 'round, visit again & relink things. Maybe not. Either way, it's less work for me. :|

PS ... If you wanted to have a play, I'd imagine that the fault lies with the function that gens up the permalink (but that's only a guess). Haven't looked at it. Send me a couple round tuits and maybe I can! :D

Cheers,

Hi,

I'm just trying to set up b2evolution with multidomain support for a site.
I already had to modify ldap-authentication, and some other modifications are also coming (for example, Create New Blog feature in the admin toolbar for every user, and possibly n-to-n user-group relations..)

Question is: is it worth to try to set up ("correct") the link system, or maybe I should wait a few weeks for beta coming out?

It will be a blog site for around 1000 people in a university dormitory.

aadaam,

I say go for it. I think the only issue is clean URLs and now that I think about it again ... I believe it's a NON-issue.

IF the hosting server is Windows-based ... go for it (as it doesn't allow clean-URLs anyway).

IF clean URLs aren't important to you ... go for it (as URL parameters, as ugly as I think they might look, are valid, permanent and work just fine).

IF clean URLs ARE important to you ... go for it (I was writing about doing redirects and things so that the OLD, ugly links still point to the clean ones, later ... but you know what? Those old, ugly links will ALWAYS WORK. It's the clean ones that would "break" if they were moved, or changed, but the ugly ones contain URL parameters that the software interprets ... so it'll always understand all that stuff.

Besides, the correction is only a one-line mod to a skin file, so when you upgrade ... no core hacks to lose, everything will still work and THEN, when you do want to switch to clean URLs ... just a back office change and one-line delete from as many skin files as you have.

Easy.

Hope this helps.