From: Hunter, Mark (Mark.Hunter@ANHEUSER-BUSCH.COM)
Date: Wed Jun 04 2003 - 15:52:43 EDT
CRC and Alignment errors are normally hardware related.
1. Even though your machines are at 100FD fixed, verify the switch is not set to
auto negotiate the ports.
2. Have the cables checked. Many times a slightly loose cable can do this.
3. Run advanced diagnostics (with wrap plugs) on your Ethernet adapters to
eliminate them as a suspect.
4. Have your network people run diagnostics on the switch ports. Hopefully they
have a Fluke unit to check the line or something.
Good Luck.
Mark Hunter
Anheuser-Busch
-----Original Message-----
From: Thomas Blaney [mailto:trb811@HOTMAIL.COM]
Sent: Wednesday, June 04, 2003 2:35 PM
To: aix-l@Princeton.EDU
Subject: Ethernet CRC and Alignment Errors
I'm experiencing some problems with poor network performance between 2 B50
servers on the same lan segment/subnet. They are both fixed at 100FD and
are connected via a Nortel Networks BayStack 450-24T Switch. They both also
have the same type of ethernet adapter installed -- IBM 10/100 Mbps Ethernet
PCI Adapter (23100020). No problems are showing up in either server's error
log.
entstat on one of the servers is showing increasing error counts for both
"CRC" and "Alignment"
# entstat -d ent0 | grep Errors
Transmit Errors: 0 Receive Errors: 0
No Carrier Sense: 0 CRC Errors: 970
Lost CTS Errors: 0 Alignment Errors: 414
Max Collision Errors: 0 No Resource Errors: 0
Late Collision Errors: 0 Receive Collision Errors: 0
Deferred: 0 Packet Too Short Errors: 0
SQE Test: 0 Packet Too Long Errors: 0
Timeout Errors: 0 Packets Discarded by Adapter:
0
No mbuf Errors: 0
No Receive Pool Buffer Errors: 0
How do I reduce/eliminate these errors? Is there a way to tell whether they
are hardware and/or software related?
The odm settings I have on each adapter are identical:
# lsattr -l ent0 -EH
attribute value description
busio 0xbfe800 Bus I/O address
busintr 6 Bus interrupt level
intr_priority 3 Interrupt priority
tx_que_size 16384 TRANSMIT queue size
rx_que_size 256 RECEIVE queue size
rxbuf_pool_size 2048 RECEIVE buffer pool size
media_speed 100_Full_Duplex Media Speed
use_alt_addr no Enable ALTERNATE ETHERNET address
alt_addr 0x000000000000 ALTERNATE ETHERNET address
ip_gap 96 Inter-Packet Gap
poll_link no Enable Link Polling
poll_link_timer 500 Time interval for Link Polling
The network settings on each server are identical as well:
# no -a |egrep "sb_max|tcp_"
sb_max = 1310720
tcp_ndebug = 100
tcp_ttl = 60
tcp_sendspace = 221184
tcp_recvspace = 221184
tcp_keepintvl = 150
tcp_keepidle = 14400
tcp_mssdflt = 512
tcp_keepinit = 150
tcp_pmtu_discover = 1
tcp_timewait = 1
tcp_ephemeral_low = 32768
tcp_ephemeral_high = 65535
tcp_newreno = 0
tcp_nagle_limit = 65535
Any suggestions would be appreciated.
Thanks.
_________________________________________________________________
MSN 8 helps eliminate e-mail viruses. Get 2 months FREE*.
http://join.msn.com/?page=features/virus
This archive was generated by hypermail 2.1.7 : Wed Apr 09 2008 - 22:16:54 EDT