Repository navigation
Feat: iOS POC - #1
Merged
Merged
Conversation
Merged
NathanSusser
pushed a commit
to NathanSusser/rust-sdks
that referenced
this pull request
Sep 17, 2026
Pings to both SFU addresses held 0.81 s and released together (623/425/219 ms, none lost); 27 media packets lost; Host A's path and modem log flat. Host B's DLF: one second of broad MAC/L1 reporting drop, no RRC, no beam-management records, no RACH. Not a handover, not a beam-failure recovery; cause unnamed. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
NathanSusser
pushed a commit
to NathanSusser/rust-sdks
that referenced
this pull request
Sep 17, 2026
…livekit#1-livekit#5 The 10-11 Sep collapse did not recur: 0 collapse seconds in 53 cycles, Host A target min >= 1.955 Mbps in 52 cycles (one brief 1.698 dip in c002), both hosts' uplinks stepped down together from ~13:30Z (same cell), floor ~25 Mbps single. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
NathanSusser
pushed a commit
to NathanSusser/rust-sdks
that referenced
this pull request
Sep 17, 2026
Whether Host B could capture packets before today has now been asserted three different ways within one day: that B could not (correct, in the original design), that B could all along and a bad sudo -n gate hid it (wrong, and I committed it), and that B could not after all (correct again, from the operator's own timeline). Memory is evidently not sufficient here, so the evidence goes in the document. stat /usr/bin/tcpdump shows mtime 2025-09-10 08:28:48 against ctime 2026-09-17 11:55:25. setcap rewrites a security xattr, which bumps ctime while leaving file contents and mtime untouched, so that pair is the signature of the capability grant rather than of any change to the binary. It lands 393 seconds after commit 33224fc, which listed this very setcap as prerequisite livekit#1, and matches the operator confirming it directly afterwards. So: before 11:55:25 on 2026-09-17, Host B could not capture, and Host A's long-standing comment gating B to h264 cells was correct when written. The sudo -n gate remains a real defect, but a forward-looking one -- it would skip B's capture today, now that the capability exists. Verified on this host rather than accepted from the peer session that found it, which is the habit the whole day argued for. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.