NixOS Binary Cache Self-Hosting | 170 Members | |
| About how to host a very large-scale binary cache and more | 58 Servers |
| Sender | Message | Time |
|---|---|---|
| 5 Mar 2024 | ||
In reply to @edef1c:matrix.orgWill Fastly always chunk up the requests into small range requests, even if the user's Nix requests the whole NAR, or only if the end user requests a range? | 04:41:58 | |
| eg it is a humongous pain in the rear to collect the narinfos, we basically have custom tools to rapid-fire pipelined S3 fetches | 04:42:04 | |
In reply to @nh2:matrix.orgi don't recall right now, sorry | 04:42:13 | |
In reply to @edef1c:matrix.orgBecause that could indeed inflate the IOPS, though Ceph has readaheads of configurable size, so it could be worked around that way | 04:43:07 | |
| Fastly segmented caching docs (https://docs.fastly.com/en/guides/segmented-caching#how-segmented-caching-works)
| 04:43:09 | |
| critical part being the final sentence | 04:43:23 | |
In reply to @edef1c:matrix.org And the beginning of the paragraph:
This suggests "no range request by nix" => "no range request by Fastly to upstream" | 04:45:12 | |
| So it should be a quite rare case | 04:45:32 | |
| out of ~1B 2xx responses, ~25% are 206 Partial Content responses, ~75% are 200 OKs | 04:46:45 | |
| so not that rare | 04:47:06 | |
In reply to @nh2:matrix.orgSorry, I had misread that sentence: I thought you wrote "mean 16MiB, median 14MiB" for file size. But it was throughput. | 04:47:48 | |
In reply to @edef1c:matrix.orgInteresting, I wonder why it's that many, at least in my nix use it is very rare to interrupt downloads | 04:48:21 | |
| we have users like. everywhere | 04:48:35 | |
| edef: Do you know total number of files? | 04:48:39 | |
| i've seen countries i'd never even heard of in the fastly logs | 04:48:48 | |