| 00:10:21 | | mls (mls) joins |
| 00:21:13 | | mls leaves |
| 00:35:39 | | Guest58 joins |
| 03:22:59 | | Guest58 quits [Client Quit] |
| 05:17:01 | | @hook54321 quits [Read error: Connection reset by peer] |
| 05:17:23 | | hook54321 (hook54321) joins |
| 05:17:23 | | @ChanServ sets mode: +o hook54321 |
| 05:53:41 | | sec^nd quits [Remote host closed the connection] |
| 05:54:12 | | sec^nd (second) joins |
| 08:06:13 | | FatalBulletHit joins |
| 08:23:02 | <FatalBulletHit> | Hello! :) |
| 08:23:02 | <FatalBulletHit> | Using VirtualBox, I've been running 4 Warriors in the background for a couple of weeks now and I would like to give some feedback regarding the Warrior, since I believe the following additions could aid archiving throughput: |
| 08:23:02 | <FatalBulletHit> | - an indication to see which projects have stuff to do in the "Available projects" tab |
| 08:23:02 | <FatalBulletHit> | - an option to automatically switch projects if n items failed or no items are available for a certain amount of tries or time |
| 08:23:02 | <FatalBulletHit> | - an option to work on multiple projects at once |
| 08:24:38 | | bladem (bladem) joins |
| 08:31:38 | | BornOn420 quits [Remote host closed the connection] |
| 08:32:10 | | BornOn420 (BornOn420) joins |
| 10:10:24 | | Guest58 joins |
| 10:13:36 | <anonymoususer852> | FatalBulletHit: I do believe the last option is reserved for advanced users. Reasons are that it is more likely to do with having an entry in the FAQ section about the warrior taking up too much bandwidth, and unnoted outlier cases like warrior VM reaping OOM. |
| 10:17:57 | <anonymoususer852> | s/FAQ/Troubleshooting/ https://wiki.archiveteam.org/index.php/ArchiveTeam_Warrior#The_warrior_is_eating_all_my_bandwidth! |
| 10:23:52 | | Guest58 quits [Read error: Connection reset by peer] |
| 10:24:51 | | Guest58 joins |
| 12:33:26 | | Guest58 quits [Client Quit] |
| 12:37:48 | | Guest58 joins |
| 12:45:12 | | Guest58_ joins |
| 12:48:59 | | Guest58 quits [Ping timeout: 260 seconds] |
| 13:05:06 | | zhongfu (zhongfu) joins |
| 13:08:00 | <FatalBulletHit> | anonymoususer852 I think the last point alone would circumvent the issue where one cannot get items or is rate limited. If implemented, it should probably keep the limit of max 6 simultaneous connections total, 🤔 |
| 13:09:20 | <FatalBulletHit> | what I'm looking for is to have it run in the background and be constantly be useful, rather than hitting an API and getting 'you're rate limited' for 5 consecutive hours 😅 |
| 13:24:14 | <anonymoususer852> | I don't believe it is that sophisticated enough to have such foresight. On potentially saturated projects, that can work well, but if users chooses projects that aren't rate-limited, e.g. telegram or urls it may still cause issues with various constraints. Last I know, URLs was pretty CPU/RAM heavy for a short period of time, but then again, that's not really a project recommended for anyone. |
| 13:26:30 | <anonymoususer852> | The rate-limiting can sometimes be a good thing. You probably wouldn't want users complaining that they've been IP banned or received abuse notices for trying to otherwise have malicious intents. |
| 14:01:24 | <FatalBulletHit> | anonymoususer852 I mean, this is just from my limited and short experience so far, but e.g. Meta will allow you to scrape the ad API for quite some time and then start rejecting further calls for a temporary ban duration. I don't tend to check in on my Warriors while my PC is running the entire day, so after a productive 30 minutes stint or so, the |
| 14:01:24 | <FatalBulletHit> | Warrior kinda just gets depressed by all the rejection by the API. |
| 14:01:24 | <FatalBulletHit> | On the other hand, there are projects, where I've 6 simultaneous tasks running for it and are all waiting for the tracker, since there is a rate limit enforced. |
| 14:01:24 | <FatalBulletHit> | So, instead of having 6 tasks failing every 2 seconds because of a temp ban or idle indefinitely because of a rate limit, it'd be more much more efficient if the 6 tasks where split between up to 6 projects OR the Warrior would switch from one project to another after a certain amount of failed tasks. |
| 14:06:50 | <anonymoususer852> | FatalBulletHit: I suppose having faster internet connectivity may lead to your IP being banned. I am also running meta but I haven't witnessed being temporarily banned. Though projects like USgov is heavily throttled, but that's likely due to being saturated (with downloaders). |
| 14:08:17 | <anonymoususer852> | Like I said, it's more of seesaw lacking more sophisticated heuristics to know when to switch projects automatically and such. |
| 14:49:06 | | Guest58_ quits [Client Quit] |
| 14:49:44 | | glassy quits [Ping timeout: 260 seconds] |
| 15:25:34 | | sec^nd quits [Ping timeout: 264 seconds] |
| 18:09:35 | | thenes quits [Remote host closed the connection] |
| 18:09:52 | | thenes (thenes) joins |
| 19:19:17 | | sec^nd (second) joins |
| 19:30:09 | | sec^nd quits [Remote host closed the connection] |
| 19:30:33 | | sec^nd (second) joins |
| 20:00:30 | | Guest58 joins |
| 22:01:46 | | FatalBulletHit quits [Quit: Ooops, wrong browser tab.] |
| 22:04:52 | | sec^nd quits [Remote host closed the connection] |
| 22:05:08 | | sec^nd (second) joins |
| 22:12:58 | | thenes quits [Ping timeout: 264 seconds] |
| 22:14:53 | | thenes (thenes) joins |
| 23:13:58 | | @Fusl quits [Quit: K-Lined] |
| 23:16:06 | | Fusl (Fusl) joins |
| 23:16:06 | | @ChanServ sets mode: +o Fusl |
| 23:17:18 | | @Fusl quits [Client Quit] |
| 23:19:04 | | Fusl (Fusl) joins |
| 23:19:04 | | @ChanServ sets mode: +o Fusl |
| 23:20:21 | | @Fusl quits [Client Quit] |
| 23:20:38 | | Fusl (Fusl) joins |
| 23:20:38 | | @ChanServ sets mode: +o Fusl |
| 23:24:08 | | @Fusl quits [Client Quit] |
| 23:24:23 | | Fusl (Fusl) joins |
| 23:24:23 | | @ChanServ sets mode: +o Fusl |
| 23:25:58 | | @Fusl quits [Client Quit] |
| 23:26:15 | | Fusl (Fusl) joins |
| 23:26:15 | | @ChanServ sets mode: +o Fusl |
| 23:59:10 | | thenes quits [Ping timeout: 264 seconds] |