Commit Graph

16 Commits (f9949cde8bba1156aaccc189e1632bf9a1478444)

Author SHA1 Message Date
Jordan Whited a228d77f86
cmd/stunstamp: add protocol context to timeout logs (#13422)
We started out with a single protocol & port, now it's many.

Updates #cleanup

Signed-off-by: Jordan Whited <jordan@tailscale.com>
3 months ago
Jordan Whited 95f0094310
cmd/stunstamp: cleanup timeout and interval constants (#13393)
Updates #cleanup

Signed-off-by: Jordan Whited <jordan@tailscale.com>
3 months ago
Jordan Whited 1fc4268aea
cmd/stunstamp: increase probe jitter (#13362)
We've added more probe targets recently which has resulted in more
timeouts behind restrictive NATs in localized testing that don't
like how many flows we are creating at once. Not so much an issue
for datacenter or cloud-hosted deployments.

Updates tailscale/corp#22114

Signed-off-by: Jordan Whited <jordan@tailscale.com>
3 months ago
Jordan Whited 1dd1798bfa
cmd/stunstamp: use measureFn more consistently in naming/signatures (#13360)
Updates #cleanup

Signed-off-by: Jordan Whited <jordan@tailscale.com>
3 months ago
Jordan Whited 6d6b1773ea
cmd/stunstamp: implement ICMP{v6} probing (#13354)
This adds both userspace and kernel timestamping.

Updates tailscale/corp#22114

Signed-off-by: Jordan Whited <jordan@tailscale.com>
3 months ago
Jordan Whited 7aec8d4e6b
cmd/stunstamp: refactor connection construction (#13110)
getConns() is now responsible for returning both stable and unstable
conns. conn and measureFn are now passed together via connAndMeasureFn.
newConnAndMeasureFn() is responsible for constructing them.

TCP measurement timeouts are adjusted to more closely match netcheck.

Updates tailscale/corp#22114

Signed-off-by: Jordan Whited <jordan@tailscale.com>
4 months ago
Jordan Whited 218110963d
cmd/stunstamp: implement HTTPS & TCP latency measurements (#13082)
HTTPS mirrors current netcheck behavior and TCP uses tcp_info->rtt.

Updates tailscale/corp#22114

Signed-off-by: Jordan Whited <jordan@tailscale.com>
4 months ago
Jordan Whited 20691894f5
cmd/stunstamp: refactor to support multiple protocols (#13063)
'stun' has been removed from metric names and replaced with a protocol
label. This refactor is preparation work for HTTPS & ICMP support.

Updates tailscale/corp#22114

Signed-off-by: Jordan Whited <jordan@tailscale.com>
4 months ago
Jordan Whited ddf94a7b39
cmd/stunstamp: fix handling of invalid DERP map resp (#12679)
Updates tailscale/corp#20344

Signed-off-by: Jordan Whited <jordan@tailscale.com>
5 months ago
Jordan Whited 94415e8029
cmd/stunstamp: remove sqlite DB and API (#12604)
stunstamp now sends data to Prometheus via remote write, and Prometheus
can serve the same data. Retaining and cleaning up old data in sqlite
leads to long probing pauses, and it's not worth investing more effort
to optimize the schema and/or concurrency model.

Updates tailscale/corp#20344

Signed-off-by: Jordan Whited <jordan@tailscale.com>
5 months ago
Jordan Whited 0d6e71df70
cmd/stunstamp: add explicit metric to track timeout events (#12564)
Timeouts could already be identified as NaN values on
stunstamp_derp_stun_rtt_ns, but we can't use NaN effectively with
promql to visualize them. So, this commit adds a timeouts metric that
we can use with rate/delta/etc promql functions.

Updates tailscale/corp#20689

Signed-off-by: Jordan Whited <jordan@tailscale.com>
5 months ago
Jordan Whited 6e106712f6
cmd/stunstamp: support probing multiple ports (#12356)
Updates tailscale/corp#20344

Signed-off-by: Jordan Whited <jordan@tailscale.com>
6 months ago
Maisem Ali 4a8cb1d9f3 all: use math/rand/v2 more
Updates #11058

Signed-off-by: Maisem Ali <maisem@tailscale.com>
6 months ago
Jordan Whited cf1e6c6e55
cmd/stunstamp: fix remote write retry (#12348)
Evaluation of remote write errors was using errors.Is() where it should
have been using errors.As().

Updates tailscale/corp#20344

Signed-off-by: Jordan Whited <jordan@tailscale.com>
6 months ago
Jordan Whited ba0dd493c8
cmd/stunstamp: validate STUN tx ID in responses (#12339)
Extremely late arriving responses may leak across probing intervals.

Updates tailscale/corp#20344

Signed-off-by: Jordan Whited <jordan@tailscale.com>
6 months ago
Jordan Whited d21c00205d
cmd/stunstamp: implement service to measure DERP STUN RTT (#12241)
stunstamp timestamping includes userspace and SO_TIMESTAMPING kernel
timestamping where available. Measurements are written locally to a
sqlite DB, exposed over an HTTP API, and written to prometheus
via remote-write protocol.

Updates tailscale/corp#20344

Signed-off-by: Jordan Whited <jordan@tailscale.com>
6 months ago