scald

It appears the call I was seeing originates in line 769 of the plugin:

function remote_login_js_loader() {
	global $current_site, $current_blog;

	if ( 0 == get_site_option( 'dm_remote_login' ) || is_user_logged_in() )
		return false;

	$protocol = is_ssl() ? 'https://' : 'http://';
	$hash = get_dm_hash();
	echo "<script src='{$protocol}{$current_site->domain}{$current_site->path}?dm={$hash}&action=load&blogid={$current_blog->blog_id}&siteid={$current_blog->site_id}&t=" . mt_rand() . "&back=" . urlencode( $protocol . $current_blog->domain . $_SERVER[ 'REQUEST_URI' ] ) . "' type='text/javascript'></script>";
}

I disabled Remote Login, cleared my caches in W3 Total Cache, and things seem to be back to normal. I can’t get it to occur after re-enabling Remote Login, so hopefully it won’t rear its head again…

I’m facing the same issue, but I think I’m closer to nailing down exactly why. So initially my MU setup was on an ip, and domain mapping was setup to use this IP. Then, I changed the WP siteurl to a domain, and only have had problems with mapped sites since. Sites that don’t have domain mapping work fine.

I noticed in the “network” tab of Chrome one particular request was holding everything up: a call to http://[ip-address]/?dm=daf25604da5c1e4b34a2f5ce3f1a7b30&action=load&blogid=7&siteid=1&t=331431286&back=http%3A%2F%2F%5Bsitename%5D.net%2F

Once that request times out the site loads properly — but after 10-15 seconds.

I figured it was a WordPress issue, and the IP was hanging around somewhere in the db, but I have changed every possible value in the tables that refers to the siteurl / domain.

Then it dawned on me that “?dm=…” is likely a domain mapping call, but I am at a loss as to where it is pulling this IP from.

Any ideas?