Is T-Mobile bypassing local ad blockers when using your hotspot?

I hate online ads.

That could be the end of the post, but sadly it’s not.

I understand that sites need ad revenue to function. But the online advertising ecosystem is so fundamentally broken that I refuse to participate in it, even to the detriment of the sites whose content I value. It’s possible to run a sustainable business through ads without ruining your site visitors’ experience.

There are various tools you can use to block the most obnoxious ads, but I’ve taken a very direct, hands-on approach. I’ve actually taken the time to view source on sites I visit that go overboard with ads, identify the domain names of the ad servers, and add them to my Mac’s /etc/hosts file so they resolve to localhost (127.0.0.1), effectively killing any ads from those sources.

And it works. I often see “broken image” icons, but all of those hideous animated ads screaming at my eyeballs are gone.

Except when I’m tethering to my phone’s hotspot.

Somehow, when I use my T-Mobile hotspot instead of my home wifi, all of those ads come flooding back. What. Is. Happening.

The only explanation I can think of is that T-Mobile is somehow using a proxy to bypass my local hosts file, but I though the local file always trumped anything else.

I don’t have an answer. But I don’t like it. And perhaps more importantly, if T-Mobile is doing this, what else is it doing with the data I’m sending and receiving over its network?

Noted for future reference: Fixing a slow-to-boot Linux server

I have a few Ubuntu Linux VPSes that were originally spun up on the then-latest-and-greatest 16.04 LTS. Over the past year I’ve been belatedly upgrading them to 20.04 LTS. Almost without exception, all now have a really irritating flaw: where they used to reboot almost instantly — making me capriciously run OS updates involving reboots at any time of day, even on servers with a bunch of sites on them, since it only meant a blip of about 5 to 7 seconds — they were now consistently taking 2 to 5 minutes to reboot. Yikes!

Poking around, I learned that cloud-init was timing out, causing that delay, but since systems administration is just a small sliver of the work I do, I never had a chance to investigate why it was happening or really what cloud-init was even for. I just resigned myself to having to do those reboots in the middle of the night when no one would notice.

Well, I finally decided I needed to get an answer, and I found it. If I’m reading this correctly, cloud-init is really only needed during the initial creation of the VPS, and can safely be disabled after that. So, let’s do it!

touch /etc/cloud/cloud-init.disabled

I’m pleased to say, it works perfectly. I ran it on a test mirror of my biggest server and it worked, so I then applied it to the live server and… capriciously rebooted it, right in the middle of the day!

Shush. It worked.


Update (August 3, 2022): Mayyyyybe it’s a bit more complicated than what I described above. I just went into another server I had previously updated to 20.04, and I just went ahead and pre-emptively ran this before an update that required reboot. After the reboot, I could not connect to the server at all, other than through Digital Ocean’s direct access console. Thank goodness for that. It did not initially occur to me that this change might be why I couldn’t connect, but after trying a few other fixes without success, I just went in and deleted this new /etc/cloud/cloud-init.disabled file and rebooted, and everything came up just fine… and without any kind of delay on boot. Weird.

A lament for a lost video gaming era

For a long time I have been lamenting why computer games aren’t like the ones I loved to play in the late ’90s/early 2000s… games like SimCity 2000 and Age of Empires. Even the über-nerdy version of Scrabble I had on my computer back then, with tournament rules and rankings, etc.

Oh, the descendants of those games certainly exist, but they have lost all of the things that made them interesting to me. And there are no new games that really capture that spirit effectively anymore. (Even the new ones that ostensibly try to evoke that spirit… don’t. At least not for a purist like me.)

Finally today I realized why. Because back then the video game market was way smaller than it is now, and it only catered to hardcore nerds like me. These days, it’s so much bigger, and so much broader, that there’s (comparatively) no money to be made on the types of games I liked back then.

And the real tragedy for me is that I can’t even play those games I loved anymore, because I don’t have any hardware that can play them. There’s emulation, but these games seem to exist in a technological gap. Emulators are great for even older games, mostly console games, but I haven’t really been able to find a decent way to emulate these games that required more computing power. Then again… maybe I just haven’t been trying hard enough.

How to purge fake/bot WooCommerce customer accounts directly in the MySQL database

DANGER! If you don’t know the havoc one can wreak with a DELETE statement in MySQL, stop right here. I take no responsibility for what you might do with the information that follows.


Bots like to create fake customer accounts on WooCommerce (WordPress) sites, apparently. What they’re attempting to do, I don’t know. But if you don’t stay on top of things, you might find you have thousands of fake customer accounts in your site. Chances are they haven’t, won’t, and can’t actually cause any damage, but they’re cluttering things up, and any unnecessary user account in a WordPress database represents a potential future security risk.

On a particular client’s heavy-traffic WooCommerce site, I discovered that over the course of the site’s 7-year lifespan it had accumulated nearly 8,000 such accounts, and I wanted to be rid of them.

After carefully exploring the data surrounding a few of these obviously fake accounts, I determined a pattern, and came up with a fairly cautious set of conditions that, to me, indicated a customer was fake:

  1. They had the customer role.
  2. Their user account had nothing in the First Name or Last Name fields.
  3. Likewise, their user account had nothing in the Billing First Name or Billing Last Name fields. (If you’re feeling extra draconian, you might skip this one.)
  4. They had never placed an order while logged in — their user ID did not have an _order_count entry in the wp_usermeta table. Which is perhaps an obvious condition because…
  5. They had never logged in at all — their user ID did not have a wfls-last-login entry in the wp_usermeta table. This condition will only apply if your site uses WordFence.

You can test all of those conditions with a single SQL query:

SELECT DISTINCT `user_id`
FROM `wp_usermeta`
WHERE
`user_id` IN (SELECT DISTINCT `user_id` FROM `wp_usermeta` WHERE `meta_key` = 'wp_capabilities' AND `meta_value` = 'a:1:{s:8:"customer";b:1;}') AND
`user_id` IN (SELECT DISTINCT `user_id` FROM `wp_usermeta` WHERE `meta_key` = 'first_name' AND `meta_value` = '') AND
`user_id` IN (SELECT DISTINCT `user_id` FROM `wp_usermeta` WHERE `meta_key` = 'last_name' AND `meta_value` = '') AND
`user_id` NOT IN (SELECT DISTINCT `user_id` FROM `wp_usermeta` WHERE `meta_key` = 'billing_first_name') AND
`user_id` NOT IN (SELECT DISTINCT `user_id` FROM `wp_usermeta` WHERE `meta_key` = 'billing_last_name') AND
`user_id` NOT IN (SELECT DISTINCT `user_id` FROM `wp_usermeta` WHERE `meta_key` = '_order_count' AND meta_value > 0) AND
`user_id` NOT IN (SELECT DISTINCT `user_id` FROM `wp_usermeta` WHERE `meta_key` = 'wfls-last-login');

You may want to spot check some of these IDs in the wp_users table, or directly in the site admin, just to be sure everything looks right. Then you can turn the above into a subquery that will delete all of the matching users. Be sure to make a full backup of your database before doing this!

DELETE FROM `wp_users` WHERE `ID` IN (
SELECT DISTINCT `user_id`
FROM `wp_usermeta`
WHERE
`user_id` IN (SELECT DISTINCT `user_id` FROM `wp_usermeta` WHERE `meta_key` = 'wp_capabilities' AND `meta_value` = 'a:1:{s:8:"customer";b:1;}') AND
`user_id` IN (SELECT DISTINCT `user_id` FROM `wp_usermeta` WHERE `meta_key` = 'first_name' AND `meta_value` = '') AND
`user_id` IN (SELECT DISTINCT `user_id` FROM `wp_usermeta` WHERE `meta_key` = 'last_name' AND `meta_value` = '') AND
`user_id` NOT IN (SELECT DISTINCT `user_id` FROM `wp_usermeta` WHERE `meta_key` = 'billing_first_name') AND
`user_id` NOT IN (SELECT DISTINCT `user_id` FROM `wp_usermeta` WHERE `meta_key` = 'billing_last_name') AND
`user_id` NOT IN (SELECT DISTINCT `user_id` FROM `wp_usermeta` WHERE `meta_key` = '_order_count' AND meta_value > 0) AND
`user_id` NOT IN (SELECT DISTINCT `user_id` FROM `wp_usermeta` WHERE `meta_key` = 'wfls-last-login')
);

OK… so the users are gone. But each one has a bunch of records in the wp_usermeta table. Now that we’ve gotten rid of the users themselves, it’s easy to purge their associated meta data:

DELETE FROM `wp_usermeta` WHERE `user_id` NOT IN (SELECT `ID` FROM `wp_users`);

WordPress development in the Gutenberg era: Threading the needle of Block Theme development

As described in several of my recent posts here, I have been working for the past month or so on building my first “all-in” Block Theme for WordPress.

After nearly 4 years of adamantly resisting “Gutenberg” and the new Block Editor revolution — not because I disliked the block concept, but because I disagreed philosophically with the core team’s approach (to what constitutes a block, which types of blocks are important, and which technologies are used to manage the UI of the editing screen) — I am finally accepting that if I am to continue making a living primarily as a WordPress developer, I need to give up on my Classic Editor, Advanced Custom Fields “Flexible Content” approach, and embrace that the Block Editor is now The Way.

One of numerous challenges I’ve faced in this process (on top of the learning curve of a completely unfamiliar method of constructing themes, the dearth of adequate and up-to-date documentation, and the core team’s willingness to allow very unfinished versions of functionality to roll out in public WordPress releases) is figuring out the best way to approach some of the more complex design structures I am used to dealing with via ACF’s Flexible Content fields.

My biggest hurdle is recognizing that what I think of as a “block” is not what the WordPress core team thinks of as a “block”

Here I will admit this is a shortcoming of my own approach. I have been “opinionated” in my development approach (well, about everything, really), and created large-scale and complex “blocks” that, in Gutenberg/Block Editor terms, would really be “groups” or “block patterns,” not blocks. Gutenberg blocks are more granular.

Gutenberg blocks are also static, in that they generally do not interact with database content, or if they do, it is in very limited “bloggy” ways that don’t align with my use of WordPress as a general-purpose CMS.

So I’ve found myself falling back on ACF. I like its server-side approach. I’m more accustomed to dealing with PHP and MySQL. I use JavaScript (mainly jQuery) a fair bit for front-end interactive elements, but I don’t build complex functionality in JavaScript and I avoid AJAX if I can help it.

You can see I’m destined for a strained relationship with the Block Editor.

A concrete example: “Tiles”

One of the most idiosyncratic elements of my old ACF approach is the block I call “Tiles.” It’s a way of creating a set of small blurbs to link to other pages/posts. There are numerous options for appearance: number of tiles per row, relationship between the image and text in a tile, etc. And there are also numerous options for the content source: a specific page or post (with the title, featured image and excerpt automatically pulled in), a dynamic feed from the blog or a specific category (likewise with the info pulled in automatically, except this one auto-updates when there’s a new post), or a completely custom-built tile, where you manually select the image and enter the text and link.

Here are screenshots of the ACF-based editing tools for my Tiles block.

I tried recreating this entire setup as a new ACF-based Block field group. It worked, to be sure, but the complex ACF editing layout really did not feel right in the new Block Editor interface, either in the sidebar (eek) or in the main content area. It felt like a cop-out.

Then I considered creating a block pattern. I knew this would lack some of the benefits of my ACF-based Tiles block, but one in particular: the option to dynamically pull in the details of another page/post, rather than manually entering the text. But as a starting point, I decided to recreate the “custom” tile type.

That, also, worked. But it was finicky and didn’t apply well in a lot of different places. So I realized that instead of creating a Tiles block pattern, I needed to create a Tile block pattern. Just one tile. Instead of a monolithic block that was really a group, users would insert each individual tile into whatever larger structure they want (e.g. columns).

The end result was a block pattern that looked like this (screenshot instead of live code because, well, you really don’t want to use this):

I was really proud of myself for using the new “lock” feature to force the elements to stay in a particular order, but to allow the user to remove elements they don’t need, such as the image, lead-in text, or CTA button.

Still… I didn’t like it. And it didn’t allow for dynamic sources.

Along the way I also finally came to grasp another fundamental limitation of the Block Editor and Block Patterns. Since all of the parameters of the blocks are stored as HTML comments right in the post content, you (the theme developer) can’t update the design of a block pattern once it’s inserted into a page/post. The Block Editor isn’t dynamically inserting content into a template when the page is rendered. It’s making a one-off copy of the template and storing it right along with the content at the point when the content is inserted into the page/post.

This seems, to me, to be a fundamental flaw in the entire Gutenberg/Block Editor approach. It’s bad enough that if I were building it myself, I’d have stopped right there and taken a completely different direction. Maybe there’s a long-term plan to address this limitation, but for now it appears to be here to stay. Which led me back to ACF.

Threading the needle

And that was when I had my insight into the true nature of the problem. It was the fact that what I think of as a “block” is larger-scale than what the Block Editor treats as a block.

So I went back to my Tiles ACF field group, stripped out the repeater functionality and created a Tile ACF field group. Now you’re building one tile at a time, it has all of the previous benefits of ACF’s dynamic integration of content with template functionality, it seems to correctly fit the definition of a “block” in the WordPress sense, and you can still have a flexible presentation of multiple tiles in a group… just using Block Editor “groups” to achieve that.

I still have more to learn and questions to answer (followed by more questions to ask and then answer), but I feel like this was a major step forward in finding a way to merge the benefits of ACF with the… inevitability, I guess, of the Block Editor.