exxos blog - random goings on

[JezC]

Interesting...when (or maybe if) I get some retro time I'll try the booster plugged into one of my STEs and the HDD emulator...

Then I'll know if I need one!

[exxos]

Found another bug in Eagle

When you create a "circle" and align a part to it, then zoom right it, the shape of the "circle" then goes weird on the vertical axis

1.PNG (22.71 KiB) Viewed 2443 times

2.PNG (24.28 KiB) Viewed 2443 times

[exxos]

.. and there Eagle goes up its own arse again.

Capture.PNG (47.83 KiB) Viewed 2432 times

2.PNG (23.32 KiB) Viewed 2432 times

[exxos]

STE526 030 booster for the STE finally got into my PCB cart! What a nightmare!

A few headers/pins are not going to be assembled at factory because is just taking way too much time up having to continuously search for parts. I really cannot be bothered to keep doing this any more.

The PLD will actually be assembled, just doesn't show up in preview. The PGA socket will be assembled either.

But overall this will hopefully give me a huge head start in final assembly of the boards when they arrive. I'm only getting five made this batch as they are prototypes. I hope 4 will get into the store. Though at this point I am not really planning on making any more after these. They take a huge amount of time and expense and I don't think there really is much demand for these anyway.

ce9c882b0c444d1395b27c5da2777496_T.png (898.4 KiB) Viewed 2376 times

3d1a558a52074e909e214ed4bbdf3520_B.png (745.69 KiB) Viewed 2376 times

[alexh]

Looking good.

If I was in the market for one this would be the design I chose.

I would like an accelerator for my STe but I am holding out for a PiStorm. The current STe PiStorm adapter boards are not quite there for me yet. They need to support disable (e.g. fall back to the onboard processor) for compatibility, fit in such a way they allow the case to close. Ideally in disable mode operate as an STBook IDE controller.

[exxos]

The drama continues, albeit partly my fault it seems

The original alignment of the PLD is correct, but then JLC "corrected it" and now its wrong. For some reason the pin 1 dot wasn't exported in the silk screen for that package But then surely JLC would have literally had to have guessed at it, rather than emailing and asking, it is certainly not good!

The only saving grace in all this, was that I had mislabelled a capacitor for assembly on the bottom layer which was actually supposed to be on the top layer. This held up the order thankfully!!

If it was not for that capacitor, the PLD would have been soldered incorrectly and I can pretty much guarantee I would never get time to correct them all.

Capture.PNG (74.1 KiB) Viewed 2325 times

In the mix of all this, I have not had any emails from JLC for some time this was because I was getting spam from "mandrillapp" , so I blocked it a few weeks back. Unfortunately JLC also use mandrillapp

EDIT:

OK so JLC just emailed asking about that alignment!

[Steve]

Looking forward to more updates on this

[exxos]

I'm swamped with multiple family crisis going on at the moment. I just haven't had time for anything the past week and likely won't this week either unfortunately. I'll probably need another week after that to recover from all the stress.

I think JLC shipped the boards, but ive ended up with like 10 stalled projects now due to lack of time.

What time I had this week I've been fighting with more server attacks. Some right strange sh*t going on and various attempts at SQL injection. Trying to write files to temp folders in the system, to many things to list. The attacks have been ongoing for the past month it seems. I spent a lot of time debugging it all and nothing was compromised from what I can see.

I've done a massive firewall update to deal with all these new attacks. Yesterday the scripts had clocked 1,409 IP addresses all attacking the server in multiple ways. These were from IPs all over the world. But could well be proxy servers from a lower number of actual hackers. I almost missed a lot of them as they were delaying requests by 15 seconds each time which made actually finding all the common attacks a bit problematic. Each log file is 100MB so that's a LOT of requests to look though.

It really is crazy what servers get attacked with. Just when you think you have seen it all, several more things appear. There's some right weird stuff going on. Like SQL injection attempts pretending to come from Google search which are attacking all index.php files. It's either a really badly confused attack bot or something really ingenious which makes no sense to me.

I also found a odd nginx "bug" . I have caches enabled and expires after 10 days. So the browser should the download the files again. But that wasn't always happening. I don't think the forum was effected. Only some pages which use server aliases. Like if you use an alias and a cache, it screws up and throws 404 errors even though the files are there. Remove the cache and it works fine, also works fine without a alias . It should be fixed now anyway.

I will try and keep an eye on the logs over the next few days. I don't think I've missed any attack attempts. But with literally millions of requests each week, it's not simple to find them all.

[exxos]

Reprinting and rebadging EEPROMs is getting worse Different prints, even though look the same sort of numbers. Totally different devices!

IMG_2328.JPG (215.95 KiB) Viewed 1510 times

[exxos]

Got myself one of these...

Place your bets on how long it will be before I find something wrong with it...

Capture.PNG (141.04 KiB) Viewed 1320 times