TCP. |
|
tcpRtoAlgorithm | The algorithm used for retransmit timeout - should display 4 ( Integer )
The algorithm used to determine the timeout value used for retransmitting unacknowledged octets.(Integer)
other (1) none of the following,
constant (2) a constant rto ,
rsre (3) MIL-STD-1778, Appendix B ,
vanj (4) Van Jacobson's algorithim **( Solaris) |
tcpRtoMin | The minimum value permitted by a TCP implementation for the retransmission timeout, measured in milliseconds. (Integer , default is 400 ms )
More refined semantics for objects of this type depend upon the algorithm used to determine the retransmission timeout. In particular, when the timeout algorithm is rsre(3), an object of this type has the semantics of the LBOUND quantity - described in (RFC) 793
see: # ndd /dev/tcp tcp_rexmit_interval_min |
tcpRtoMax | The maximum value permitted by a TCP implementation for the retransmission timeout, measured in milliseconds. (Integer , default is 60000ms )
More refined semantics for objects of this type depend upon the algorithm used to determine the retransmission timeout. In particular, when the timeout algorithm is rsre(3), an object of this type has the semantics of the UBOUND quantity described in (RFC) 793
see: # ndd /dev/tcp tcp_rexmit_interval_max |
tcpMaxConn | The total number of TCP connections the entity can support. In entities where the maximum number of connections is dynamic, this object should contain the value -1.(dynamic) |
tcpActiveOpens | The number of times TCP connections have made a direct transition to the SYN-SENT state from the CLOSED state. The outgoing calls you make are called Active Opens. You decide who to call and why. An outgoing rlogin, or Web browser request causes an active open. (counter) |
tcpPassiveOpens | The number of times TCP connections have made a direct transition to the SYN-RCVD state from the LISTEN state. (counter )
The incoming calls are called Passive Opens as they occur without any activity on your part. In this case, you just have to have a program running that is waiting to pick up the phone. For exampl,e a Web server listens on port 80 for any incoming requests. |
tcpAttemptFails | The number of failed connection attempts. (counter)
The number of times TCP connections have made a direct transition to the CLOSED state from either the SYN-SENT state or the SYN-RCVD state, plus the number of times TCP connections have made a direct transition to the LISTEN state from the SYN-RCVD state. The tcpAttemptFails increments on outgoing connections when a reset is received and we have sent a syn, or if the connection times out after a syn is sent.
On an incoming connection, it increments if we receive a syn and then send one, but do not get a response, or get a reset. |
tcpEstabResets | The number of connection resets. (client ungracefully shut downs ). (counter)
The number of times TCP connections have made a direct transition to the CLOSED state from either the ESTABLISHED state or the CLOSE-WAIT state. |
tcpCurrEstab | The number of connections open, ESTAB or CLOSE_WAIT, (Gauge)
Each connection can last a few milliseconds (for a Web server on a local LAN running a benchmark) or many seconds or minutes (for a large transfer on a slow Internet connection). Each established connection will use up a process or a thread in your system. The number of TCP connections for which the current state is either ESTABLISHED or CLOSE- WAIT. see:
ESTABLISHED displayed with # netstat -P tcp |
tcpOutSegs | The total number of segments sent, including those on current connections but excluding those containing only retransmitted octets. (counter) |
tcpOutDataSegs | The number of TCP segments sent. (counter)
The total number of segments (tcpOutSegs) is mostly made up of tcpOutDataSegs and tcpOutAck, although the numbers don't add up exactly. The outgoing data is divided into segments, where each segment corresponds to an Ethernet packet. |
tcpOutDataBytes | The number of data bytes sent. (counter) |
tcpRetransSegs | The total number of segments retransmitted is the number of TCP segments transmitted containing one or more previously transmitted octets. If no acknowledgment is received, the segment is retransmitted. (counter)
see Doc 1009001.1 Note: An excessive number of TCP retransmissions indicates that client accessing your server have slow or error-prone connections. Retransmissions consume bandwidth that might otherwise go to serving new connections. |
tcpRetransBytes | The number of data bytes retransmitted. (counter) |
tcpOutAck | The number of acknowledgement only packets sent. (counter) |
tcpOutAckDelayed | The number of delayed acknowledgements sent. (counter)
see: # ndd/dev/tcp tcp_deferred_ack_interval |
tcpOutUrg | The number of packets sent with the URGENT flag. (counter) |
tcpOutWinUpdate | The number of window update only packets sent. (counter) |
tcpOutWinProbe | The number of window probes sent. (counter) |
tcpOutControl | The number of control - SYN, FIN, RST packets sent. (counter) |
tcpOutRsts | The number of connection resets . client requesting non-listening port. (counter)
The number of TCP segments sent containing the RST flag. |
tcpOutFastRetrans | The number of TCP segments sent with the URGENT flag only. (counter) |
tcpInSegs | The number of segments received on TCP connections. (counter)
The total number of segments received, including those received inerror. This count includes segments received on currently established connections. |
tcpInAckSegs | The number of ack only packets recv. (counter) |
tcpInAckBytes | The number of bytes ack by received ack packets. (counter) |
tcpInDupAck | The number of duplicate ack only packets recv. (counter) Almost every dropped packet will trigger some duplicate acks. and you get them if packets arrive out of order |
tcpInAckUnsent | The number of acknowledgements received for unsent data. (counter) |
tcpInInorderSegs | The number of packets received in sequence. (counter) |
tcpInInorderBytes | The number of bytes received in sequence. (counter) |
tcpInUnorderSegs | The number of packets received out of order. (counter) |
tcpInUnorderBytes | The number of bytes received, out of order pkts. (counter) |
tcpInDupSegs | The number of completely duplicate packets. this is a counter. |
tcpInDupBytes | The number of completely duplicate bytes received. (counter) |
tcpInPartDupSegs | The number of packets w some duplicate data. (counter) |
tcpInPartDupBytes | The number of dup bytes received w some duplicate data. (counter) |
tcpInPastWinSegs | The number of segments received after window. (counter)
It is the number of segments received outside of the window -- when the window is non-zero. Note: Segments received when the window is zero are considered probes.
A high number indicates this system has peers that aren't very good at arithmetic. |
tcpInPastWinBytes | The number of bytes received after window. (counter)
A large value could actually mean that the peer isn't following the tcp window size of this system. It likely means that your peers don't implement TCP properly. |
tcpInWinProbe | The number of window update packets received. (counter) Segments received when the window is zero. |
tcpInWinUpdate | The number of window update packets received. (counter) |
tcpInClosed | The number of connections closed, including drops. (counter) |
tcpRttNoUpdate | The number of times unsuccessful getting RTT. (counter) |
tcpRttUpdate | The number of times successful getting round trip time. (counter) |
tcpTimRetrans | The number of retransmission timeouts. (counter) |
tcpTimRetransDrop | The number of connections dropped due to retrans timeout. (counter).
The tcpTimRetransDrop parameter shows the number of aborts since boot time due to abort timer expirations. This count includes both SYN requests and established TCP connections. |
tcpTimKeepalive | The number of of keep alive timeouts. (counter) |
tcpTimKeepaliveProbe | The number of keep alive probes sent. (counter) |
tcpTimKeepaliveDrop | The number of connections dropped in keep alive. (counter) |
tcpListenDrop | The number of connections refused due to backlog full on listen.
the listen queue length is the minimum of tcp_conn_req_max_q and the qlen value passed to listen().
see: # ndd /dev/tcp tcp_conn_req_max_q |
tcpListenDropQ0 | The number of connections refused due to half-open queue (q0) full
see: # ndd /dev/tcp tcp_conn_req_max_q0 |
tcpHalfOpenDrop | The number of connections dropped from a full half-open queue (q0)
see above |
tcpOutSackRetrans | The number of retransmitted segments by SACK retransmission
if sack enabled
see: # ndd /dev/tcp tcp_sack_permitted |