diff options
author | Bhasker Hariharan <bhaskerh@google.com> | 2019-05-30 10:47:11 -0700 |
---|---|---|
committer | Shentubot <shentubot@google.com> | 2019-05-30 12:08:41 -0700 |
commit | ae26b2c425d53aa8720238183d1c156a45904311 (patch) | |
tree | e7eff51fb10a29493586d4193acb3b1047ce1d57 /pkg/tcpip/tcpip.go | |
parent | 8d25cd0b40694d1911724816d72b34d0717878d6 (diff) |
Fixes to TCP listen behavior.
Netstack listen loop can get stuck if cookies are in-use and the app is slow to
accept incoming connections. Further we continue to complete handshake for a
connection even if the backlog is full. This creates a problem when a lots of
connections come in rapidly and we end up with lots of completed connections
just hanging around to be delivered.
These fixes change netstack behaviour to mirror what linux does as described
here in the following article
http://veithen.io/2014/01/01/how-tcp-backlog-works-in-linux.html
Now when cookies are not in-use Netstack will silently drop the ACK to a SYN-ACK
and not complete the handshake if the backlog is full. This will result in the
connection staying in a half-complete state. Eventually the sender will
retransmit the ACK and if backlog has space we will transition to a connected
state and deliver the endpoint.
Similarly when cookies are in use we do not try and create an endpoint unless
there is space in the accept queue to accept the newly created endpoint. If
there is no space then we again silently drop the ACK as we can just recreate it
when the ACK is retransmitted by the peer.
We also now use the backlog to cap the size of the SYN-RCVD queue for a given
endpoint. So at any time there can be N connections in the backlog and N in a
SYN-RCVD state if the application is not accepting connections. Any new SYNs
will be dropped.
This CL also fixes another small bug where we mark a new endpoint which has not
completed handshake as connected. We should wait till handshake successfully
completes before marking it connected.
Updates #236
PiperOrigin-RevId: 250717817
Diffstat (limited to 'pkg/tcpip/tcpip.go')
-rw-r--r-- | pkg/tcpip/tcpip.go | 19 |
1 files changed, 19 insertions, 0 deletions
diff --git a/pkg/tcpip/tcpip.go b/pkg/tcpip/tcpip.go index c8164c0f0..f9886c6e4 100644 --- a/pkg/tcpip/tcpip.go +++ b/pkg/tcpip/tcpip.go @@ -760,6 +760,25 @@ type TCPStats struct { // successfully via Listen. PassiveConnectionOpenings *StatCounter + // ListenOverflowSynDrop is the number of times the listen queue overflowed + // and a SYN was dropped. + ListenOverflowSynDrop *StatCounter + + // ListenOverflowAckDrop is the number of times the final ACK + // in the handshake was dropped due to overflow. + ListenOverflowAckDrop *StatCounter + + // ListenOverflowCookieSent is the number of times a SYN cookie was sent. + ListenOverflowSynCookieSent *StatCounter + + // ListenOverflowSynCookieRcvd is the number of times a valid SYN + // cookie was received. + ListenOverflowSynCookieRcvd *StatCounter + + // ListenOverflowInvalidSynCookieRcvd is the number of times an invalid SYN cookie + // was received. + ListenOverflowInvalidSynCookieRcvd *StatCounter + // FailedConnectionAttempts is the number of calls to Connect or Listen // (active and passive openings, respectively) that end in an error. FailedConnectionAttempts *StatCounter |