00:13:19Arcorann (Arcorann) joins
00:20:04Arcorann_ joins
00:22:24Arcorann quits [Ping timeout: 240 seconds]
00:48:12Arcorann (Arcorann) joins
00:51:29Arcorann_ quits [Ping timeout: 265 seconds]
00:56:13Yg84WJchYH87d3A5tF quits [Client Quit]
01:06:21Arcorann quits [Ping timeout: 244 seconds]
01:30:19Arcorann (Arcorann) joins
01:38:44blue507 quits [Ping timeout: 240 seconds]
01:40:55nepeat quits [Remote host closed the connection]
01:41:01nepeat joins
01:43:56blue507 joins
02:25:22systwi quits [Read error: Connection reset by peer]
02:26:06systwi (systwi) joins
02:38:44TheFluffyCat joins
02:54:13<TheFluffyCat>Are we looking for a faster URL archiver than TTT2? I'm working on one written in C#, and so far it shreds right through bitly.
03:17:41hex joins
03:29:39@Fusl quits [Changing host]
03:29:39Fusl (Fusl) joins
03:29:39guybrush.hackint.org sets mode: +o Fusl
03:29:50@Fusl quits [Changing host]
03:29:50Fusl (Fusl) joins
03:29:50guybrush.hackint.org sets mode: +o Fusl
04:21:34atphoenix quits [Changing host]
04:21:34atphoenix (atphoenix) joins
04:23:25qw3rty__ joins
04:27:32qw3rty_ quits [Ping timeout: 265 seconds]
05:05:27TheFluffyCat quits [Remote host closed the connection]
05:23:42systwi quits [Read error: Connection reset by peer]
05:24:20systwi (systwi) joins
05:24:46nepeat quits [Client Quit]
05:24:54nepeat joins
05:34:10manyworlds joins
05:34:37manyworlds quits [Remote host closed the connection]
07:28:48HackMii quits [Remote host closed the connection]
07:30:39HackMii (hacktheplanet) joins
09:00:18britmob2 quits [Quit: britmob2]
09:34:27hex quits [Ping timeout: 265 seconds]
09:54:16hex joins
10:31:04hex quits [Client Quit]
12:13:29Arcorann_ joins
12:17:20Arcorann quits [Ping timeout: 265 seconds]
12:34:38Matthww quits [Quit: The Lounge - https://thelounge.chat]
12:57:08Matthww joins
13:24:31Datechnom quits [Ping timeout: 265 seconds]
13:34:23Datechnom (Datechnom) joins
13:44:55Yg84WJchYH87d3A5tF joins
13:50:06britmob2 joins
14:04:58Arcorann (Arcorann) joins
14:07:32Arcorann_ quits [Ping timeout: 265 seconds]
14:14:15OpSecNoob joins
14:14:35<OpSecNoob>Hey, whats the docker link for this project?
14:18:16<AK>atdr.meo.ws/archiveteam/urls-grab:latest
14:18:24<@JAA>No, that's another project.
14:19:36<AK>Oh poo good point, The warrior docker image is here: https://hub.docker.com/r/archiveteam/warrior-dockerfile/
14:20:00<@JAA>That's not the Docker image for this project though. :-P
14:20:27<OpSecNoob>JAA do you have it?
14:20:40<@JAA>atdr.meo.ws/archiveteam/terroroftinytown-client-grab:latest
14:20:57<OpSecNoob>thanks!
14:22:07<AK>Ahh I didn't know there was a separate image for this project, I'd thought this was the only one baked into warrior. Won't make that mistake again.
14:39:26Arcorann quits [Ping timeout: 265 seconds]
14:42:57<OpSecNoob>So this appears to be the linux image, is there one for windows (I'm getting: docker: image operating system "linux" cannot be used on this platform.)
14:53:34<OpSecNoob>nvm, got it :)
14:59:01AK quits [Client Quit]
15:00:19AK joins
15:08:55Matthww quits [Ping timeout: 265 seconds]
16:01:37Matthww joins
16:09:04Matthww quits [Ping timeout: 240 seconds]
16:12:50Matthww joins
16:44:44OrIdow6 quits [Ping timeout: 240 seconds]
16:45:23OrIdow6 (OrIdow6) joins
16:47:21OrIdow6 quits [Remote host closed the connection]
16:48:55OrIdow6 (OrIdow6) joins
17:42:48TheFluffyCat joins
18:08:16sierra_wright joins
18:10:50<sierra_wright>how long has the urlteam project been going on? and what % of the target spaces have been checked? out of curiosity
18:12:41<TheFluffyCat>At the current rate of scanning (650/sec), it has been running for about 3 and a half years.
18:13:54<TheFluffyCat>It's about time we upgrade the project's code, too... My machines are running at about 5% capacity at the moment.
18:15:16<sierra_wright>i get a LOT of no free items messages but i can clearly see that for example ow.ly isn't even out of the numbers in the most significant space yet
18:15:42TheFluffyCat68 joins
18:16:17<TheFluffyCat68>arkiver do you know the database schema/where we can get the crawl results? I'd like to make a compatible crawler that could replace the current one.
18:16:53<AK>sierra_wright: I think that's the tracker limiting how many items it gives out to avoid ddosing the target websites
18:17:34<@JAA>Correct
18:18:03dsi joins
18:18:27<sierra_wright>i figured that was part of it
18:18:39TheFluffyCat quits [Ping timeout: 244 seconds]
18:18:49<TheFluffyCat68>Ha. Bye former me.
18:18:56<@JAA>TheFluffyCat68: There is no DB. The output are simple text files (compressed), and you can find them here: https://archive.org/search.php?query=terroroftinytown
18:19:01<TheFluffyCat68>Thanks.
18:19:01<@JAA>The format should be quite self-explanatory.
18:19:21<TheFluffyCat68>I assumed it was Anti-DDoS- but even so, 600/second is not a lot, especially for the 20-or-so projects that are currently running.
18:19:55<TheFluffyCat68>At that rate, you wouldn't even net a network of people helping you- you can afford that hosting easily between 2-3 people.
18:20:46madmax (madmax) joins
18:20:49<@JAA>Only 2 or so projects are running currently I think.
18:20:55<AK>Some of the other projects have different numbers of people working on it, so it varies
18:21:30<@JAA>Ok, 4 it seems.
18:22:06<TheFluffyCat68>Also, seems that the site digbig.com is having issues with their shortener; all requests to the domain go on a 301 loop.
18:22:20<TheFluffyCat68>I noticed a few of my machines crawling it earlier today on TTT.
18:22:36<sierra_wright>i just got assigned urlteam by hitting the "archiveteam choice" project, so if the other ones have even less scanning...
18:22:58<TheFluffyCat68>Urlteam is the only one available on the Warrior.
18:22:58<madmax>i got a question: is there any reason to have maximum 6 concurrent items running per warrior?
18:23:11<TheFluffyCat68>I think it's just arbitrary. Frustrates me too, madmax
18:23:21<TheFluffyCat68>The rest are independent docker containers, sierra_wright
18:24:03<sierra_wright>i see
18:24:05<@JAA>That's mostly historic at this point. We used to run all projects through the warrior. Nowadays, this is the only one that still works.
18:24:07<TheFluffyCat68>You can find them on the website, alongside the project descriptions
18:24:34<@JAA>But back then, when we switched the default project, people would start running whatever we were working on. And many times, doing that at high concurrency was a Bad Idea.
18:24:41<madmax>ahh okay, JAA. it ended up that i run 17 docker containers now, each 6 items…
18:24:46<@JAA>Rate limits, disk space, etc.
18:25:12<@JAA>You can run the terroroftinytown-client-grab container or the scripts at up to 20 concurrency though if you want.
18:25:12<TheFluffyCat68>Think of it this way: If bit.ly did not want to be crawled, they would have released their database.
18:25:33<@JAA>Note that you'll anyway get limited to one item per shortener per IP because rate limits.
18:28:05<TheFluffyCat68>As a test I just downloaded 300 bit.ly URLs in quick succession; no rate limiting on there.
18:28:51<TheFluffyCat68>Especially with how lightweight serving a URL shortener is- and how many people likely use it per second-
18:28:59<TheFluffyCat68>I think it's worth it to speed up the process a little.
18:29:32<sierra_wright>yeah crank up the volume!
18:29:50<TheFluffyCat68>Wait- correction- 400 URLs.
18:29:56<madmax>TheFluffyCat68: same for me. i wrote a script to just fest the headers but there was definitely no rate limit, at least for bit.ly
18:30:10<TheFluffyCat68>I'll try tinyurl now, too.
18:30:23<AK>The risk is they do spot it, implement rate limiting or just start banning ips
18:31:26<@JAA>You can get banned from bit.ly quite easily.
18:31:54<@JAA>Also, I don't think we can control the items/IP limit on a per-shortener basis.
18:32:31<@JAA>The better approach is to just spread things out by adding more shorteners.
18:32:52<TheFluffyCat68>Heh... are you suggesting I make a URL shortener?
18:33:02<@JAA>But well, someone has to do it, monitor it, adapt the rules as needed, etc.
18:33:26@JAA slaps TheFluffyCat68 around a bit with a large trout
18:33:40<madmax>JAA: i totaly agree. would be too hard to have separate rates for each shortener. bit.ly has published the rates for api access
18:33:51<TheFluffyCat68>You use the API?
18:33:55<AK>Doh just realised the log messages for ttt are different to the other projects, can't collect them in the same way
18:33:58<TheFluffyCat68>Wait, no.
18:34:01<madmax>1000 calls per ip per hour
18:34:03<@JAA>I'm pretty sure we don't.
18:34:16<madmax>no in a custom script
18:34:21<@JAA>AK: Yes, this shares nearly no code with the other projects.
18:35:03<TheFluffyCat68>Have we ever tried... asking them... for a list of their URLs?
18:35:12<@JAA>Another thing that's been on the todo list for a loooooong time is writing WARCs. Now we're just throwing away much of the information, and the redirects don't go into the Wayback Machine either.
18:35:37<@JAA>Bitly is a member of 301Works. They know about efforts like this.
18:35:59<@JAA>They're just arseholes and don't share their DB.
18:36:26<TheFluffyCat68>Why don't we ask nicely
18:36:35<@JAA>I'm certain IA has done just that years ago.
18:37:04<@JAA>The whole concept behind 301works is that shorteners give their mappings to IA for safekeeping, and once the shortener ceases to operate, they're made public.
18:37:24<TheFluffyCat68>"Dear Sir or Madam, we are a group of rogue internet archivists. For a while, we've been crawling your site, and seeing which shortened links go where. This is very slow, however, and I think it'd be better for the both of us if we could just have a list of your shortened URLs. Sincerely, Terror of Tiny Town."
18:37:56<TheFluffyCat68>*No one*'s tried that?
18:38:11<@JAA><@JAA> I'm certain IA has done just that years ago.
18:39:58<TheFluffyCat68>Hmm, okay.
18:40:53<TheFluffyCat68>(Off topic) While we're here, has anyone considered re-making Warrior with Lua as the primary environment? Would be nice to centralize projects again.
18:45:26<@JAA>A new warrior is planned, yes.
18:45:51<TheFluffyCat68>Where can I find development? #warrior?
18:48:07<@JAA>Nothing's public yet, but yes, discussion will likely be there once things start moving.
19:23:40TheFluffyCat68 quits [Remote host closed the connection]
19:23:49<dsi>Hey everyone, has anyone gotten the standalone to work through Azure Containers? I can get warrior to work (az container create -g ArchiveTeamWarrior --name warrior --image archiveteam/warrior-dockerfile --ip-address public --port 8001) but not so much luck with the standalones.
19:47:02<madmax>dsi: i don't know what does not work for you. but a container in azure would be way to expensive. i have tried it some time ago for another project, one instance costs us 30€/month
19:47:37<madmax>a vps which can handle several docker containers is cheaper (approx. 10€/month or less)
19:56:52<dsi>madmax i get a lot of free Azure credit that i'm not currently using.
19:57:21<dsi>but that does sounds about right; each warrior container is ~1 a day.
20:05:53blue507 quits [Remote host closed the connection]
20:10:59madmax quits [Client Quit]
20:11:57sliccricc quits [Changing host]
20:11:57sliccricc (sliccricc) joins
20:12:30sliccricc quits [Client Quit]
20:12:45sliccricc joins
20:13:05sliccricc quits [Changing host]
20:13:05sliccricc (sliccricc) joins
20:13:26sliccricc quits [Remote host closed the connection]
20:13:40sliccricc (sliccricc) joins
20:26:37atphoenix quits [Quit: Leaving]
20:27:02atphoenix (atphoenix) joins
21:00:31madmax joins
21:15:05madmax quits [Client Quit]
21:32:35inf33ct joins
22:06:01ragu quits [Read error: Connection reset by peer]
22:06:27ragu joins
22:55:41OpSecNoob quits [Remote host closed the connection]
23:21:52Arcorann (Arcorann) joins