by YEO ANN KI » Sun, 17 Jul 1994 11:17:55
We keep getting the message: "attempt to xmit but no xmit i/o buffer"
for our bf0 (FDDI) interface.
I have changed "bufhwm" in /etc/system from the default 2% to 10% of
the RAM (512MB). The message is still coming, sometimes right after
"boot -r" when there is only a few login users.
This happens at our Sparc1000 with 8CPU and 512MB running 2.3 at 101318-42
patch level.
Anybody knows how to fix this problem? how to monitor how much buffer
has been used?
--
Ann-Kian Yeo
1. ? attempt to xmit but no xmit i/o buffer
I'm getting a message in /var/adm/messages:
"attempt to xmit but no xmit i/o buffers"
Environment:
server: sparc 10 96mb, solaris 2.3 - standard patches,
sun's fddi, nfs
client: sun ipx, ethernet, nfs mounted fs from the server,
running an intense read test (100MB/10Krecs).
I'm testing concurrent reads from 12 ipx clients attached to the
fddi ring via 12 ethernets. the clients start reading and the
messages come out in the servers message file... and all slows
to a crawl.
Any ideas on what parameter(s) to change to allow more xmit buffers?
3. token ring message - tr0: xmit ret_code: 22 xmit error code: 00
4. /bin/passwd
5. Token Ring error xmit ret_code: 22 xmit error code: 88
7. xmit hung ?
10. using sockets to sniff/xmit
11. 'xmit on stopped card' problem
12. [re-xmit] kprobes for 2.5.58
13. Socket xmit out of sequence?