tailscale

Commit Graph

Author	SHA1	Message	Date
halulu	874be6566d	netcheck: DERP latency over HTTPS when UDP is blocked * netcheck: DERP letency over HTTPS when UDP failed Updates #207 Signed-off-by: Zijie Lu <zijie@tailscale.com> * netcheck: async DERP latency check over HTTPS Updates #207 Signed-off-by: Zijie Lu <zijie@tailscale.com> * netcheck: DERP latency check over HTTPS: fix concurrent map Updates #207 Signed-off-by: Zijie Lu <zijie@tailscale.com> * netcheck: DERP latency check over HTTPS: some improvements Updates #207 Signed-off-by: Zijie Lu <zijie@tailscale.com> * netcheck: DERP latency check over HTTPS: use timeout context Updates #207 Signed-off-by: Zijie Lu <zijie@tailscale.com> * netcheck: DERP latency check over HTTPS: use report mutex Updates #207 Signed-off-by: Zijie Lu <zijie@tailscale.com> * netcheck: DERP latency check over HTTPS if UDP is BLOCKED Updates #207 Signed-off-by: Zijie Lu <zijie@tailscale.com> * netcheck: DERP latency check over HTTPS: new function measureHTTPSLatency Updates #207 Signed-off-by: Zijie Lu <zijie@tailscale.com>	4 years ago
Brad Fitzpatrick	44b07aa708	netcheck: STUN less aggressively to known distant servers If Australia's far away and not going to be used, it's still going to be far away a minute later. No need to send backup just-in-case-UDP-gets-lost STUN packets to the known far away destinations. Those are the ones most likely to trigger retries due to delay anyway (in random 50-250ms, currently). But we'll keep sending 1 packet to them, just in case our airplane landed. Likewise, be less aggressive with IPv6. The main point is just to see whether IPv6 works. No need to send up to 10 packets every round. Max two is enough (except for the first round). This does mean our STUN traffic graphs for IPv4-vs-IPv6 will change shape. Oh well. It was a weird eyeball metric for IPv6 connectivity anyway and we have better metrics. We can tweak this policy over time. It's factored out and has tests now.	4 years ago
Brad Fitzpatrick	73249f8032	netcheck: aggregate spammy logging onto one concise line	4 years ago
Brad Fitzpatrick	19a1704abd	netcheck: use best DERP server in past 5 minutes to avoid flip-flopping Fixes #162	4 years ago
Brad Fitzpatrick	42e98d4edc	Quiet two little log annoyances.	4 years ago
Brad Fitzpatrick	b9c6d3ceb8	netcheck: work behind UDP-blocked networks again, add tests Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>	4 years ago
Brad Fitzpatrick	afc3479d04	netcheck: fix data races for staggler STUN packets arriving after GetReport Fixes #179	4 years ago
Brad Fitzpatrick	0d3f42e1d8	netcheck: ignore IPv4 STUN failures if we saw at least one reply Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>	4 years ago
Brad Fitzpatrick	ed7e088729	netcheck: ignore IPv6 STUN failures Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>	4 years ago
Brad Fitzpatrick	01b4bec33f	stunner: re-do how Stunner works It used to make assumptions based on having Anycast IPs that are super near. Now we're intentionally going to a bunch of different distant IPs to measure latency. Also, optimize how the hairpin detection works. No need to STUN on that socket. Just use that separate socket for sending, once we know the other UDP4 socket's endpoint. The trick is: make our test probe also a STUN packet, so it fits through magicsock's existing STUN routing. This drops netcheck from ~5 seconds to ~250-500ms. Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>	4 years ago
Brad Fitzpatrick	5c1e443d34	wgengine/monitor: don't call LinkChange when interfaces look unchanged Basically, don't trust the OS-level link monitor to only tell you interesting things. Sanity check it. Also, move the interfaces package into the net directory now that we have it.	4 years ago
Brad Fitzpatrick	39c0ae1dba	derp/derpmap: new DERP config package, merge netcheck into magicsock more Fixes #153 Updates #162 Updates #163 Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>	4 years ago
Brad Fitzpatrick	2cff9016e4	net/dnscache: add overly simplistic DNS cache package for selective use I started to write a full DNS caching resolver and I realized it was overkill and wouldn't work on Windows even in Go 1.14 yet, so I'm doing this tiny one instead for now, just for all our netcheck STUN derp lookups, and connections to DERP servers. (This will be caching a exactly 8 DNS entries, all ours.) Fixes #145 (can be better later, of course)	4 years ago
Brad Fitzpatrick	ff6b3c2c38	netcheck: include two more DERP nodes, show more in CLI report	4 years ago
Brad Fitzpatrick	596fd449b9	netcheck: add TODO about alternate timing strategies	4 years ago
Brad Fitzpatrick	b27d4c017a	magicsock, wgengine, ipn, controlclient: plumb regular netchecks to map poll Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>	4 years ago
Brad Fitzpatrick	844d991baf	netcheck: add Report.Clone, return cloned report to avoid races on late replies	4 years ago
Brad Fitzpatrick	bcf3719b9e	netcheck: add hairpinning detection Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>	4 years ago
Brad Fitzpatrick	f3c0ea23d9	netcheck, stunner: reduce log spam Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>	4 years ago
Brad Fitzpatrick	14559340ee	Start of netcheck package & including network state in Hostinfo. * adds new packet "netcheck" to do the checking of UDP, IPv6, and nearest DERP server, and the Report type for all that (and more in the future, probably pulling in danderson's natprobe) * new tailcfg.NetInfo type * cmd/tailscale netcheck subcommand (tentative name, likely to change/move) to print out the netcheck.Report. Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>	4 years ago

20 Commits (874be6566db0f5747cbc24a3c9a5070f3eeb9f9e)