| 00:40:42 | <h3ndr1k> | flashmeow: That is true for almost all projects. People need to remember that the goal isn't to increase your score on the tracker. Shorteners can just rate limit your ip and you could produce invalid results in the worst case. |
| 00:45:21 | <flashmeow> | h3ndr1k: fair enough. I do find it a little frustrating though when I see my workers idling while seeing the same names go by again and again on the tracker. (i know they're probably running from multiple servers with different ips) |
| 00:46:08 | <Krownest> | I assume the big names run 10s if not 100s of machines, or have a bunch of IPs hanging around. |
| 00:46:35 | <Krownest> | I'm content with my 2 concurrency amongst two machines since as long as it gets done any participation is progress. |
| 00:49:38 | <flashmeow> | It would be pretty neat to get some sort of round-robin queuing based on ips and services to distribute the work between servers faster |
| 00:50:57 | <flashmeow> | but I don't know much about how the redis queue is designed (or how to keep open connections between the tracker and client) |
| 00:54:20 | <flashmeow> | ...something something websockets are the future |
| 03:27:00 | | qw3rty_ joins |
| 03:30:38 | | qw3rty__ quits [Ping timeout: 258 seconds] |
| 03:33:28 | | Boltie2 (Boltie) joins |
| 03:36:46 | | Boltie quits [Ping timeout: 258 seconds] |
| 04:32:28 | <@JAA> | flashmeow: On this project, the tracker will only ever hand out one item per shortener per IP to prevent you from getting banned even at high concurrencies. |
| 04:33:27 | <@JAA> | Basically, if you see the 404 message and your workers aren't doing anything, this simply means that we have way more workers than items, which leads to an individual worker only rarely receiving items. |
| 04:34:10 | <@JAA> | In an ideal world, we'd have more shorteners enabled on the tracker, but that means someone has to enter them first... |
| 04:35:26 | <flashmeow> | If a client makes a request for another item while the first is still in progress, will the tracker try to give an item from a different service (if available)? |
| 04:41:10 | <@JAA> | I think so, but I don't remember the details. The code's on GitHub if you want to dig around. |
| 04:42:58 | <@JAA> | But yeah, when a new shortener is added, its items are gobbled up in a couple seconds or less. We have way more workers than items on this project. |
| 08:06:56 | | billy549 quits [Ping timeout: 250 seconds] |
| 08:13:33 | | billy549 (Billy549) joins |
| 12:09:00 | | Daloader joins |
| 18:23:03 | | flashmeow quits [Client Quit] |
| 18:25:23 | | flashmeow (flashmeow) joins |
| 20:30:32 | | Daloader quits [Ping timeout: 250 seconds] |
| 21:27:31 | | @OrIdow6^2 is now known as @OrIdow6 |
| 22:57:09 | | Boltie2 quits [Client Quit] |
| 22:57:23 | | Boltie (Boltie) joins |
| 23:38:07 | | Boltie2 (Boltie) joins |
| 23:41:35 | | Boltie quits [Ping timeout: 258 seconds] |
| 23:54:12 | | Boltie2 quits [Ping timeout: 250 seconds] |