Disqus - Latest Comments for ax0n

Re: Tinkercad: How to Split an Object in 4 Easy Steps

ax0n — Wed, 10 Jul 2024 10:00:11 -0000

I ran into the same problem. I solved it by first selecting the main shape I want to split, then clicking the "Repeat and duplicate" button (looks like 3 overlapping squares next to copy and paste icons). This made a perfect copy of my base work that was exactly in the same spot, so when I join one of the "hole" boxes, it leaves the duplicate there. Then when I click the light bulb to show the other "hole" box, I select what's left of my model on the other side and join those. Then I'm left with one half from the original, one half from the replicated part, and in total I have a split object.

Re: What is wardriving? How to prevent it

ax0n — Mon, 29 Jan 2024 12:51:59 -0000

Hi! Wardriving is actually a harmless pastime of many technology enthusiasts. Despite the aggressive-sounding name, wardriving is completely passive -- we do not connect to any of the networks we identify. The goal is not to highlight insecure networks, attack anything or anyone. All wireless networks transmit signals that can be observed by passersby, and we have made a game out of it. It's a little bit like geocaching or playing Pokemon Go. Taking note of these signals is not cybercrime, nor does it aid criminals.

Wireless networks not intended for public use should still be encrypted with WPA2 or WPA3, and if you, as a user, decide to use public Wi-Fi, you might opt to use VPN as one additional layer of security on top of HSTS and TLS that is practically used everywhere to protect all users' data from bad actors. None of these things will "prevent wardriving," and it's not really something that needs to be prevented.

Re: http://www.h-i-r.net/2008/02/tinkering-with-combogard-2.html

ax0n — Wed, 01 Nov 2023 17:48:25 -0000

I really have no idea but I'd start by hooking up an ammeter in series with the 9v battery to see how much draw it's getting. Ultimately you might need to crack open the lock body itself and look for damaged electronic components on the board. You can also carefully spray it with upside down air duster just enough to cause light frost to form on the board and see where it evaporates quickest to help identify hot spots

Re: The Google Cemetery

ax0n — Thu, 22 Nov 2018 12:15:00 -0000

I'm also self-hosting Tiny-Tiny RSS. It feels a lot like Google Reader, and with everything (your internal labels, starred articles, etc) being exposed as a public RSS feed with UUID-style 'difficult to guess' URLs, I'm mixing it up with other things like IFTTT to re-share interesting stuff to my social feeds.

Re: Dissecting a Simplex lock

ax0n — Thu, 31 May 2018 00:12:41 -0000

I don't have access to this lock anymore, but I did try that. If I recall correctly, pressing 2 and 4 at the same time, for example,
pawls 1 and 2 increment twice. Pawls 3 and 4 increment once.

Re: Flyback transformers and CRT discharge. OF DEATH.

ax0n — Sat, 12 May 2018 16:43:57 -0000

See Jay Fawkes' comment in this same thread. He got shocked. Don't do it.

Re: Raspberry Pi Zero inside a Lapdock 100

ax0n — Thu, 15 Feb 2018 13:01:12 -0000

You plug the micro-USB cable into the USB port (not the power port) on the Zero. It will feed power to the Pi Zero over the USB port (and it works fine) and it will then show up to the Pi as a USB hub with the trackpad, keyboard and two USB ports on the back.

Re: DPFE Sensor and EGR Information

ax0n — Mon, 29 Jan 2018 12:57:05 -0000

Losing power at high revs could also be a fuel filter or fuel pump problem.

Re: OpenBSD vmm hypervisor: Part 2

ax0n — Sat, 27 Jan 2018 20:07:54 -0000

I hunted down the change. It's this thread: https://www.mail-archive.co...

The solution is to set up bridge0 to connect to vether0, so make a file called /etc/hostname.bridge0 and inside, the contents are:

add vether0

now, reboot or re-run netstart:

doas sh /etc/netstart

This will bring up bridge0, which will be connected to vether0 (which is your nat segment, using your built-in DHCP environment, like I configured in this article).

Now, this is the new switch config in vm.conf:
switch "local" {
interface bridge0
}

Now, any vms that you have configured will have their tap(4) allocated to the bridge matching your vm.conf bridge. My VMs are all working again now.

Re: OpenBSD vmm hypervisor: Part 2

ax0n — Fri, 26 Jan 2018 22:54:07 -0000

I recently upgraded snapshots on my laptop and noticed this as well. It looks like switch entries in VM.Conf have to be tied to bridge(4) interfaces now. I think. I'm seeing the same problem, and I haven't fully solved it.

Re: Building a honeypot army with Pi, EC2 and MHN

ax0n — Wed, 15 Nov 2017 22:40:50 -0000

Also, would you be willing to share access to your MHN in France, if I also provide access to ours here in Kansas City? I would happily trade HPFeeds credentials with you :D

Re: Building a honeypot army with Pi, EC2 and MHN

ax0n — Wed, 15 Nov 2017 22:39:34 -0000

Brandon took Rob Scanlon's Encom Boardroom visualization and did a lot of changes, including building some custom APIs to get the data to the dash. Honestly, he's way more skilled at front-end and middleware software development than I am. I believe most of his changes are in his Git Repo, but there may be a few missing pieces to get it all working. https://github.com/bwinches...

Re: Handbrake adjustment - Focus Hacks

ax0n — Sun, 08 Oct 2017 21:44:44 -0000

Ages ago, in a few forum posts, this option came up. Several folks ended up breaking the cheap plastic frame around the handbrake boot. If you're careful and don't mind the risk, it's an option.

Re: Logical Domains on SunFire T2000 with OpenBSD/sparc64

ax0n — Fri, 02 Jun 2017 10:41:18 -0000

The system itself is pretty anemic despite the massive-sounding number of cores. Part of this is that the ALU and FPU functions are divvied up and there aren't actually 32 of those to go around. Before I upgraded to 6.1 and split this thing into LDOMs, I had run John The Ripper with 32 parallel threads. Each thread was insanely slow, at around 270 C/s (basically, password-guesses-per-second). That put the entire system at about 8500 C/s. That's only marginally faster than each of the 4 cores in my 2010-era Clarksfield i7 laptop, which was ripping through close to 29,000 C/s.

Raw compute power aside, though, with LDOMs, you can have a decent number of almost-bare-metal 4- or 8-core systems running general-purpose stuff (web, mail, database, file shares) on a single 2U chassis, which more than makes up for the ~300W of juice these things consume. It's a nice addition to my home lab, if nothing else, though I still haven't dared run it around the clock for more than a few days straight.

Re: OpenBSD vmm hypervisor: Part 2

ax0n — Tue, 30 May 2017 10:38:43 -0000

Thanks for the heads up! I've moved the pfctl reload down a few spots in the walk-through.

Re: Raspberry Pi Zero inside a Lapdock 100

ax0n — Wed, 26 Apr 2017 00:17:57 -0000

Yep. The LapDock supplies power into the micro USB plug. This usually charges the phone while docked, but in this case, it's powering the Pi through the same port it's using for data.

Re: PHP/MySQL Walk-Throughs updated for OpenBSD 6.1

ax0n — Mon, 24 Apr 2017 10:22:49 -0000

nginx on OpenBSD 6.1: http://www.h-i-r.net/p/open...

Re: PHP/MySQL Walk-Throughs updated for OpenBSD 6.1

ax0n — Sat, 22 Apr 2017 22:50:43 -0000

I'll see if I can revise the base-httpd walk-through to work for nginx. The formatting in the existing page is a nightmare and it needs to be replaced from the ground up. Give me a few days and I'll see about keeping it alive.

Re: Raspberry Pi Zero inside a Lapdock 100

ax0n — Sat, 04 Feb 2017 20:58:31 -0000

FYI, you'll also need completely different cables for any of the other Pi aside from the Zero. You can find a whole lot of how-to's for hooking up full-size Pi to a lapdock. I'd recommend looking up one of those.

Re: Make your own tint-safe window cleaner solution

ax0n — Wed, 01 Feb 2017 16:13:19 -0000

I've been using cheap 32-ounce spray bottles from home depot for years. I'll give good ol' Dentless Dave a quick kick in the nuts for being so vague, though.

Re: PHP/MySQL on OpenBSD's relayd-based httpd

ax0n — Sun, 15 Jan 2017 19:11:25 -0000

I tested this walk-through on OpenBSD 6.0 in my lab. What part wasn't working for you?

Re: Programming new keyless remotes

ax0n — Tue, 03 Jan 2017 16:00:34 -0000

Are you talking about the remote buttons on the key to unlock the car, or getting a new key to start the 2012 Focus?

Re: DPFE Sensor and EGR Information

ax0n — Mon, 15 Aug 2016 14:37:30 -0000

It might not have enough force to keep the vacuum pressure from actuating the EGR valve, but if you're having the problem, give it a shot and let me know how it works.

Re: DNS Tunneling Part 1: Intro and Nameserver setup

ax0n — Sun, 22 May 2016 13:04:00 -0000

You need to at the very least set up a subdomain and an NS record for it, yes. If you can get UDP packets to port 53 on any host in the world, you can probably just use Iodone in raw UDP mode. Otherwise, for recursive DNS queries to work in highly-restricted environments, the DNS server on the LAN you're trying to get out from needs to have an authoritative DNS server to connect to -- hence, the NS record and subdomain.

Iodine and DNSCAT2 didn't exist when I wrote this series, but they both have some serious advantages over the tools I wrote about back in 2010. You should check them out. They both work best if you have an NS record though.

Re: PHP/MySQL on OpenBSD's relayd-based httpd

ax0n — Tue, 26 Apr 2016 16:49:00 -0000

FYI: Updated for 5.9 as of a few days ago. I'm running WordPress on a few of my web servers. I'm also running tiny-tiny RSS.

The benefits of allowing WordPress to update itself, in my opinion, outweigh the risk that an exploit will allow an attacker to write files to your web directory. Set the ownership of the WordPress files to "www" and allow the owner write access (mode 644 should do nicely). You may also opt to use group www ownership (mode 664) instead. Wordpress' automatic updates are pretty slick.

As for PHP apps encountering trouble communicating out, you likely need to get enough data into /var/www chroot to do DNS lookups. Make a directory called /var/www/etc and copy (don't link) your /etc/hosts and /etc/resolv.conf files there. The ownership and permissions should be the same there as they are in /etc (root:daemon, mode 644)

Note that there are a lot of things that change a bit when you run in a chroot environment. Anything that requires access to files, binary executables, or filesystem sockets has to be copied into the matching directory structure under /var/www. Some other things that cause hang-ups are programs that want to send mail by directly calling /usr/sbin/sendmail (you'll need to copy /usr/sbin/sendmail to /var/www/usr/sbin/sendmail and any necessary libraries from /usr/lib to /var/www/usr/lib for example)

Chroot keeps most successful Remote Code Inclusion/Execution attacks from impacting the entire system. They limit damage to the /var/www directory structure.