Rick's b.log - 2024/11/17
It is the 19th of November 2024 You are 18.225.92.95, pleased to meet you!

Congratulations Amazon, your bot is the first one blocked

So, no, I don't bother to keep up with all the bots and block them. It's really not worth my time. Plus, who'd be dumb enough to consider anything that I write to be in any way authoritative? I know Jack Shit (he's much nicer than his name would imply, his wife Jill, on the other hand, is a bitch) and this blog is a mixture of the things that I do know, the things that I think I know, the things that it turns out that I didn't actually know, and some random dribble about the mess that gets called "politics". Oh, and a lot of it is a cross between catharsis and my attempt to figure out life and my part in it.

So this morning my blog carried an apology that comments could not be added. This happens when the server gets 'full' and, you know, when disc allocation runs out, it runs out.

I have...had...two log files, generated by PHP. One was on this blog, that would record an IP address and what page was selected. It looks like this:

2024/11/17 10:54:33  ?=20210516         54.36.xxx.xxx
2024/11/17 10:54:54  ?=20230411         54.36.xxx.xxx
2024/11/17 10:55:21  ?=20210404         54.36.xxx.xxx
2024/11/17 10:55:47  ?=20170917         54.36.xxx.xxx

People using OVH France IP addresses (all different) looked at those pages at those times. I was going to write a program to read the IPs and plot them on a map list countries in order of access, and also to collate the pages viewed in order of popularity, but I never got around to it so what actually happens is every so often the log just gets deleted.

Don't panic about this, the server access logs record a lot more, but I don't know how to access those, so a couple of lines of PHP did the job with much less hassle...

To give you an example, here's a snippet of my WebJames access log for my personal server:

hostglobal.plus - - [13/Nov/2024:12:35:38 +0000] "GET /app_dev.php/_profiler/open?file=app/config/parameters.yml HTTP/1.1" 404 343 "" "Mozilla/5.0 (Linux; Android 4.05; Galaxy Nexus Build/IMM76B) AppleWebKit/535.25 (KHTML, like Gecko) Chrome/17.0.1025.133 Mobile Safari/536.19"
hostglobal.plus - - [13/Nov/2024:12:35:38 +0000] "GET /app_dev.php/_profiler/phpinfo HTTP/1.1" 404 346 "" "Mozilla/5.0 (Linux; Android 7.1.1; N9560 Build/NMF26F) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.109 Mobile Safari/537.36"
hostglobal.plus - - [13/Nov/2024:12:35:40 +0000] "GET /application/.env HTTP/1.1" 404 333 "" "Mozilla/5.0 (compatible; MSIE 10.0; Windows NT 6.2; Trident/6.0; Touch; MASMJS)"
hostglobal.plus - - [13/Nov/2024:12:35:40 +0000] "GET /apps/.env HTTP/1.1" 404 326 "" "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_12_6) AppleWebKit/603.3.6 (KHTML, like Gecko) Version/10.1.2 Safari/603.3.6"
hostglobal.plus - - [13/Nov/2024:12:35:41 +0000] "GET /backend/.env HTTP/1.1" 404 329 "" "Mozilla/5.0 (Linux; U; Android 2.2.2; en-us; HUAWEI-M835 Build/HuaweiM835) AppleWebKit/533.1 (KHTML, like Gecko) Version/4.0 Mobile Safari/533.1"

The domain hostglobal.plus is 81.53.136.197, and these are obviously attempts to hack my server. Well, it's like nothing else (a custom-written server, not an Apache clone) running on RISC OS, so none of the usual exploits work. I find the differing user agents to be an interesting touch.

But, whatever, that's an example of server level logging. Every server does this.

As a side note, I see a number of attempts to pass ='nvOpzp; AND 1=1 OR (<'">iKO)) as a parameter. SQL injection and XSS exploit attempt?

The second, and much older, log was attached to the main index page of my site. It would create a log that looked like this:

10:54:00 17/11/2024 (35.170.205.140, 35-170-205-140.crawl.amazonbot.amazon)
10:54:01 17/11/2024 (34.234.197.175, 34-234-197-175.crawl.amazonbot.amazon)
10:54:01 17/11/2024 (3.218.35.239, 3-218-35-239.crawl.amazonbot.amazon)

This does not track you around the site. It only records who gets the main index. If you come here to the blog, nothing is recorded there. If you go look at /software, nothing is recorded there. It also, very specifically, does not record any URL parameters, as I know that Google at one point had a tendency to leak search phrases. The idea, also, was to plot a chart of where my visitors were coming from but, again, I never got around to writing the software necessary to do it.

That, main log, was an astonishing 1.9GB. I, obviously, deleted it. By the time it had been deleted and I'd gone to refresh the AndFTP display, the log was 9K. Now I know I'm not that popular!

It was a simple task of commenting out one line of code to disable the logging. So I did, that log is no longer maintained. All these years and I never looked at what it actually said, so might as well just turn it off.

But before I did...

10:54:00 17/11/2024 (35.170.205.140, 35-170-205-140.crawl.amazonbot.amazon)
10:54:01 17/11/2024 (34.234.197.175, 34-234-197-175.crawl.amazonbot.amazon)
10:54:01 17/11/2024 (3.218.35.239, 3-218-35-239.crawl.amazonbot.amazon)
10:54:03 17/11/2024 (35.169.240.53, 35-169-240-53.crawl.amazonbot.amazon)
10:54:03 17/11/2024 (3.209.174.110, 3-209-174-110.crawl.amazonbot.amazon)
10:54:04 17/11/2024 (34.234.200.207, 34-234-200-207.crawl.amazonbot.amazon)
10:54:05 17/11/2024 (54.198.33.233, 54-198-33-233.crawl.amazonbot.amazon)
10:54:06 17/11/2024 (54.210.155.69, 54-210-155-69.crawl.amazonbot.amazon)
10:54:07 17/11/2024 (54.225.98.148, 54-225-98-148.crawl.amazonbot.amazon)
10:54:09 17/11/2024 (52.22.87.224, 52-22-87-224.crawl.amazonbot.amazon)
10:54:09 17/11/2024 (107.22.208.39, 107-22-208-39.crawl.amazonbot.amazon)
10:54:10 17/11/2024 (3.221.156.96, 3-221-156-96.crawl.amazonbot.amazon)
10:54:11 17/11/2024 (3.232.39.98, 3-232-39-98.crawl.amazonbot.amazon)
10:54:13 17/11/2024 (44.205.120.22, 44-205-120-22.crawl.amazonbot.amazon)
10:54:13 17/11/2024 (52.202.52.82, 52-202-52-82.crawl.amazonbot.amazon)
10:54:14 17/11/2024 (18.235.158.19, 18-235-158-19.crawl.amazonbot.amazon)
10:54:15 17/11/2024 (34.225.138.57, 34-225-138-57.crawl.amazonbot.amazon)
10:54:17 17/11/2024 (52.22.87.224, 52-22-87-224.crawl.amazonbot.amazon)
10:54:17 17/11/2024 (54.235.158.162, 54-235-158-162.crawl.amazonbot.amazon)
10:54:18 17/11/2024 (18.213.27.222, 18-213-27-222.crawl.amazonbot.amazon)
10:54:19 17/11/2024 (3.209.174.110, 3-209-174-110.crawl.amazonbot.amazon)
10:54:21 17/11/2024 (50.16.248.61, 50-16-248-61.crawl.amazonbot.amazon)
10:54:21 17/11/2024 (100.28.44.58, 100-28-44-58.crawl.amazonbot.amazon)
10:54:22 17/11/2024 (52.1.106.130, 52-1-106-130.crawl.amazonbot.amazon)
10:54:23 17/11/2024 (52.54.157.23, 52-54-157-23.crawl.amazonbot.amazon)
10:54:25 17/11/2024 (18.204.152.114, 18-204-152-114.crawl.amazonbot.amazon)
10:54:25 17/11/2024 (54.164.106.236, 54-164-106-236.crawl.amazonbot.amazon)
10:54:26 17/11/2024 (44.217.255.167, 44-217-255-167.crawl.amazonbot.amazon)
10:54:27 17/11/2024 (52.71.218.25, 52-71-218-25.crawl.amazonbot.amazon)
10:54:29 17/11/2024 (34.238.45.183, 34-238-45-183.crawl.amazonbot.amazon)
10:54:29 17/11/2024 (34.195.60.66, 34-195-60-66.crawl.amazonbot.amazon)
10:54:30 17/11/2024 (18.211.148.239, 18-211-148-239.crawl.amazonbot.amazon)

It goes on and on like that. Over 110 entries in a minute and a half, Amazon spamming me from all different IP addresses. More specifically, Amazon spamming me by fetching the exact same page - I repeat again, the log only recorded fetches of /index.html, nothing else.

Which, of course, earns Amazon the destinction of being the very first bot to annoy me enough to add a couple of lines to my robots.txt file:

User-agent: Amazonbot
Disallow: /

I don't know if the bot is already obeying this, or will carry on spamming me until the next time it bothers to fetch the robots file. I mean, if it's fetching robots and the index, that's - what - some two hundred (known) fetches per minute? What the hell, Amazon, makes you think that clobbering a site like that is acceptable? Note, also, the vast number of IP addresses used. It's like they know it's a crappy thing to do but don't want to trigger any anti-spam measures that might block the same IP address from connecting too frequently.

Not cool, Amazon, not cool at all.

CreArt

Like Generative AI, it can sometimes do some really weird things like people with additional limbs, and asking for a person swimming wearing a snorkel can get it comically wrong. However, it's a simple matter to tap the button for a different image, or to tweak the description and try again.
This is where having limits with too few images can be a problem, because it may take a fair bit of fiddling to get an image that looks sort of what you want.
Okay, one should cough up money for the use of the app, but at €4,99/week, €39,99/year, or €94,99/forever autorenewing, it's something that would need to be budgeted for rather than being an impulse purchase. Sure, a fiver a week doesn't sound bad. But if you forget about it, well, it's a shocking two hundred and sixty a year. The forty/year option is better, but this is an amusing toy. Is it worth that to me? I used to put up with Genie and its adverts because it allowed me a simple trade of one-picture-per-advert. When they started adding such strict limits, that's when it became useless to me.
Like I said, AI images does tend to need a fair bit of tweaking to get what you actually want. You'll see some pictures below. On my phone I have over a hundred, and those are the ones that I kept and didn't instantly reject. AI is not a tool that spits out perfect pictures every time. And sometimes when a picture looks just right, you notice the three legs and/or seven fingers and... well... you get the idea.

If you want to play, CreArt is available on the usual app stores for the usual mobile devices.

Since it is AI you can do all sorts of interesting things.

For example, what would this place be like if it was in an animé?

If this were an animé...

I should point out that the photo was taken just now, with a dreary not-quite-drizzle grey sky and leaves falling from the trees.

Alright then, how about my car if it was knitted?

If my car were knitted...

Since it's AI, you can easily create scenes that don't exist, so long as you can describe them well enough - that's the tricky part.

Very few, possibly only one, of you will know what this following picture is about.

Which one is Belinda?

Let's take an equally dreary photo looking up the driveway and animéify it. This time, I'll give the dual before-and-after output.

The photo and the animé version.

I asked for a miniature of a flying train on a table in a rural kitchen. I nearly wet myself when I saw this. Well played, AI bot, well played.

Look carefully, you'll see it. 😂

And, finally, let's get interactive. Here's one for you to colour in yourselves.

Your HeyRick Colouring Book, sort of.

If you don't fancy downloading and colouring a reduced-size image, I have made a PDF version available (828.29K). Have fun!
BTW: If you scan and send me your colourings, I'll feature them here.

Lemon cake

...it really isn't. It's chilly and bleak and dark and unpleasant. Zombies and vampires and boars, oh my!

So what better way than to appreciate that little dose of sunshine that goes by the name of "lemon". If "sun" could have a taste, that taste would be it. And what better way to do that, since teaching your granny to suck lemons is a waste of time, than to present lemon in the form of cake?

Everything is better with cats cake.

A slice of lemon cake.

Betty Crocker "Zesty Lemon Cake" with a layer of lemon créme in between. I think that's a sort of lemon curd that's trying to be a spread. Either way, it works well here. It had better, there's half a jar between the parts of that cake!

I don't know what I want for dinner. I'm leaning towards burger & chips. Simple enough to make, so I can crawl into my room, fire up the heated blanket (that I first turned on last night when the outdoor dropped to two (yikes!)) and maybe try to finish Everything Everywhere All At Once even though it is annoying having to have the French subs so I have some sort of idea what the Chinese parts are about. Thanks, Netflix, for not including other subs...

Your comments:

A tree-dwelling mammal, 17th November 2024, 19:28

Probably best I don't try colouring in. I'd be too tempted to eat the crayon. *sits in the corner dribbling*    (It's been one of those days. I may have Covid again. Or it might just be man-flu. Either way, have slept most of the day.)