|
|
311c612f2e
|
Only flag stillAlive on successful write
|
2019-07-17 23:23:19 +01:00 |
|
|
|
307b24d8cb
|
Fix Conn.Read/Conn.Write behavior after Conn.Close, get rid of second TUN/TAP conn reader goroutine, no longer use deadlines
|
2019-07-17 21:42:17 +01:00 |
|
|
|
1bf1c6eb36
|
Revert "Remove stillAlive code from TUN/TAP conn as it is no longer required with the new deadlines"
This reverts commit eec70bf2f2.
|
2019-07-17 19:43:29 +01:00 |
|
|
|
eec70bf2f2
|
Remove stillAlive code from TUN/TAP conn as it is no longer required with the new deadlines
|
2019-07-17 13:53:16 +01:00 |
|
|
|
7d1c03d2ac
|
Only call stillAlive if channel read succeeds
|
2019-07-17 12:07:16 +01:00 |
|
|
|
747b50bb7c
|
Try to improve handling of timeouts
|
2019-07-17 11:13:53 +01:00 |
|
|
|
d34600b5f9
|
Try to fix TUN/TAP conn reader leakage
|
2019-07-17 10:12:10 +01:00 |
|
|
|
9e086e70f0
|
Don't indefinitely block TUN/TAP reader goroutine when a conn error happens
|
2019-05-30 12:44:47 +01:00 |
|
|
|
0096d1ae3e
|
Re-add ICMPv6 packet too big handling
|
2019-05-29 20:16:17 +01:00 |
|
|
|
b2513fce56
|
have the tunConn close things after a 2 minute timeout
|
2019-05-28 18:35:52 -05:00 |
|
|
|
5ea864869a
|
don't spam searches for unused connections. todo: timeout old connections somehow
|
2019-05-23 20:27:52 -05:00 |
|
|
|
522ed147b1
|
use the subnet derived ID/mask when creating a connection based on a subnet address, fix a potential blocking channel send in tuntap/conn.go, and get debug.go compiling well enough to profile things (the sim is currently still broken)
|
2019-05-15 18:01:26 -05:00 |
|
|
|
efdaea1b5e
|
fix some races and GetBytes/PutBytes usage, but this still seems to deadlock somewhere in iperf tests
|
2019-05-02 17:37:49 -05:00 |
|
|
|
5f66c4c95c
|
Try using separate workers for each TUN/TAP connection (sometimes produces duplicate packets when communicating with both the node address and a subnet address, sometimes also can't Ctrl-C to quit)
|
2019-04-28 17:14:09 +01:00 |
|
