HomeSearchPrevious dayNext day

00:10:57qw3rty__ joins
00:14:30qw3rty_ quits [Ping timeout: 268 seconds]
00:29:34Zachava quits [Read error: Connection reset by peer]
00:31:39Zachava (Zachava) joins
01:35:55Cuphead2527480 (Cuphead2527480) joins
02:05:10SootBector quits [Remote host closed the connection]
02:06:19SootBector (SootBector) joins
02:50:50Webuser351022 joins
02:53:21Webuser351022 quits [Client Quit]
03:45:42Cuphead2527480 quits [Client Quit]
03:54:14nukke quits [Quit: nukke]
03:57:18nukke (nukke) joins
04:40:31DogsRNice quits [Read error: Connection reset by peer]
06:53:20SootBector quits [Remote host closed the connection]
06:54:27SootBector (SootBector) joins
09:17:49Zachava quits [Read error: Connection reset by peer]
09:20:42Zachava (Zachava) joins
11:51:23SootBector quits [Remote host closed the connection]
11:52:31SootBector (SootBector) joins
14:10:35<justauser>Looks like US and CA S3 clusters have different configurations.
14:11:00<justauser>https://ia600207.us.archive.org/30/items/test-memac-index-test/ is an open directory while https://dn721909.ca.archive.org/0/items/picosong.com_finder/ is blank.
14:51:30<@JAA>That's not S3. http://ia600207.s3dns.us.archive.org/test-memac-index-test/ (yes, only HTTP) would be S3.
14:52:13<@JAA>(Well, also available over HTTPS, but not with a valid cert.)
14:53:32nukke quits [Quit: nukke]
14:56:51nukke (nukke) joins
15:16:54<IDK>hm, so its possible to force a specific server?
15:17:07<IDK>insteaed of s3.us.archive.org
15:23:59<justauser>My understanding is that s3.us is the one you are supposed to touch, and it'll redirect you as necessary.
15:24:29<IDK>nah, I was thinking I can point s3.us to a specific server
15:25:51pabs quits [Ping timeout: 268 seconds]
15:26:08<justauser>Others aren't guaranteed to work - they may not have a file even if they had it in the past.
15:27:10pabs (pabs) joins
15:30:58<klea>If you want, from the metadata endpoint you can get the server names which currently have the file, but you shouldn't need to do that.
15:41:06pabs quits [Excess Flood]
15:41:53pabs (pabs) joins
16:35:04Matthww quits [Quit: Ping timeout (120 seconds)]
16:36:04Matthww joins
16:47:47Matthww quits [Read error: Connection reset by peer]
16:47:50Matthww joins
17:28:40<@JAA>Yes, s3.us.archive.org is the one to use. It then redirects to s3dns on HTTP for bucket listing (and a few other things).
17:30:17<@JAA>Or well, considering the invalid certs etc., it's probably best not to use it at all except for uploading etc.
18:17:06KoalaBear84 joins
18:19:45KoalaBear quits [Ping timeout: 268 seconds]
18:20:39KoalaBear joins
18:22:18KoalaBear84 quits [Ping timeout: 268 seconds]
18:35:20nukke quits [Quit: nukke]
18:55:06nukke (nukke) joins
19:46:51<klea>So the ISP of where you upload files knows your s3 creds?
19:47:45<@JAA>No, the s3dns stuff for individual servers isn't used for uploads.
19:48:26<klea>Oh ok, that's good.
19:49:24<@JAA>But it's one reason why I haven't yet implemented retrieving multipart upload information from IA for resumption in ia-upload-stream. That endpoint isn't currently available over secure HTTPS.
19:57:34<klea>mhmm
19:58:35<klea>well, https://ia600207.s3dns.us.archive.org/test-memac-index-test/ at least seems to give a cert valid for *.us.archive.org, us.archive.org so you could make it check if it's valid for a subdomain of archive.org, based on the host?
20:03:05<@JAA>Yeah, but I'd rather not.
20:05:01<klea>Oh, it'd be very easy, just curl --resolve s3.us.archive.org:$(dig +short AAAA $server):443 -H "Host: $server" https://s3.us.archive.org/$endpoint, but yeah a bit hacky.
20:48:21linuxgemini quits [Quit: getting (hopefully fresh) air o/]

HomeSearchPrevious dayNext day