High TCP segment retransmit rate Post: 302602819

9 More Discussions You Might Find Interesting

1. UNIX for Advanced & Expert Users

Sun: High kernel usage & very high load averages

Hi, I am seeing very high kernel usage and very high load averages on my system (Although we are not loading much data to our database). Here is the output of top...does anyone know what i should be looking at? Thanks, Lorraine last pid: 13144; load averages: 22.32, 19.81, 16.78 ...

2. Programming

Segment Fault

When run it, segment fault. What is wrong? #include <stdio.h> #include <stdlib.h> const int max =20; //**************************************************** // Input Matrix //**************************************************** void inMatrixAA(int *AA, int row, int col)...

3. UNIX for Dummies Questions & Answers

code segment

how do i close a do code segment? od?

4. Programming

How can I know where the segment of memory is all Zero?

I mean, I malloc a segment of memory, maybe 1k maybe 20bytes.. assume the pointer is pMem How can I know the content pMem refered is all Zero or \0 . I know memcmp but the second parameter should another memory address... thanx

5. Programming

Data segment or Text segment

Hi, Whether the following piece of code is placed in the read-only memory of code (text) segment or data segment? char *a = "Hello"; I am getting two different answers while searching in google :( that's why the confusion is

6. Red Hat

apache high cpu load on high traffic

i have a Intel Quad Core Xeon X3440 (4 x 2.53GHz, 8MB Cache, Hyper Threaded) with 16gig and 1tb harddrive with a 1gb port and my apache is causing my cpu to go up to 100% on all four cores heres my http.config <IfModule prefork.c> StartServers 10 MinSpareServers 10 MaxSpareServers 15...

7. Programming

why segment fault,

I always get segment fault, why? can sb help me and modify it, I have spend on much time on #include <pthread.h> #include <stdio.h> #include <stdlib.h> #include <sys/time.h> #include <string.h> #define MAX 10 pthread_t thread; void *thread1() { int *a; int i, n; ...

8. Shell Programming and Scripting

what would a script include to find CPU's %system time high and user time high?

Hi , I am trying to :wall: my head while scripting ..I am really new to this stuff , never did it before :( . how to find cpu's system high time and user time high in a script?? thanks , help would be appreciated ! :)

9. Solaris

Too much TCP retransmitted and TCP duplicate on server Oracle Solaris 10

I have problem with oracle solaris 10 running on oracle sparc T4-2 server. Os information: 5.10 Generic_150400-03 sun4v sparc sun4v Output from tcpstat.d script TCP bytes: out outRetrans in inDup inUnorder 6833763 7300 98884 0...

LEARN ABOUT REDHAT

tc575_cb

3C575_CB(4)						     Kernel Interfaces Manual						       3C575_CB(4)

NAME

       3c575_cb - 3Com 3c575 Etherlink XL 10/100 CardBus device driver

SYNOPSIS

       insmod 3c575_cb.o [debug=n] [rx_copybreak=n] [max_interrupt_work=n] [use_hw_csums=n] [down_poll_rate=n] [auto_polarity=n]

DESCRIPTION

       3c575_cb  is  the  low-level  Card  Services  driver for the 3Com 3c575 EtherLink XL CardBus series ethernet adapters.  When this driver is
       attached to a card, it allocates the next available ethernet device (eth0..eth#).  This information for this card will be reported  in  the
       kernel log file, and the device name passed on to cardmgr(8).

       The  driver uses the N-Way autonegotiating transceiver to negotiate the speed and duplex of the network link.  If the link partner does not
       support negotiation only the link speed is detected.

       If the on-card EEPROM is programmed for a full-duplex (FD) link, only FD media types are advertised and the link is  assumed  to  be  full-
       duplex.

PARAMETERS

       rx_copybreak=n
	      Set  the breakpoint for copying incoming frames into reduced-sized packet buffers.  The default value of 200 bytes is reasonable for
	      most systems.

       max_interrupt_work=n
	      This option selects the maximum amount of work handled during each interrupt.  Each received packet counts as one unit of  work,	as
	      does  updating  statistics  counters  and  handling  errors.   The default value of 32 should only need to be increased on very slow
	      machines.  An occasional "too much work in interrupt" message is not a problem.

       use_hw_csums=n
	      Controls use of an adapter's on-board TCP/IP checksum verification for incoming packets.	This is on by default; setting to  0  dis-
	      ables  hardware  checksums.   It may be useful to disable hardware checksums if it is suspected that something is corrupting data on
	      its way from the card to the CPU.

       down_poll_rate=n
	      For Cyclone and Tornado chipsets, this is the rate at which the card polls the transmit ring for new packets.  The value is in units
	      of  320 ns; a value of 0 disables polling.  The default is 20 (6.4 us), and the maximum allowed value is 127.  Polling is more effi-
	      cient but can consume a significant amount of PCI bus bandwidth.

       This flag can be used to control the transceiver's auto polarity
	      detection capability.  Some Cisco Catalyst switches are known to interact badly with this feature in 10baseT mode.  It defaults to 1
	      (enabled).

AUTHOR

       This driver: Donald Becker - becker@scyld.com

       Linux PC Card control software: David Hinds - dahinds@users.sourceforge.net et al.

SEE ALSO

       cardmgr(8), pcmcia(5).

pcmcia-cs							2001/06/22 01:12:38						       3C575_CB(4)