From: Joe Dierker (ja_dierker@hotmail.com)
Date: Wed Apr 09 2003 - 13:06:50 EDT
I have an E6500 running Solaris 8 in a production environment. We have
recently begun to encounter some problems with significant packet loss
(20-25%) during the overnight hours, at about the same time our backups are
kicking off. The backup software is Legato Networker version 6.1.1. The
network interface is a single hme.
Netstat -k output shows that we are getting significant increases in two
categories. Both nocanput errors and norcvbuf errors are plentiful during
the 30-60 minute problem period and non-existent during the rest of the day.
I have found little information from Sun or elsewhere on how to address
these issues. Most of the docs I find are third party. Recommendations
include increasing the value of sq_max_size to address the nocanput errors,
and adjusting the tcp driver parameters tcp_conn_req_max_q and
tcp_conn_req_maxq0 to address the norcvbuf values. However, I have found
limited docs on these parameters, suggested values or ramifications of such
adjustments.
Specific questions in order of importance to me at this time:
1. Can anyone provide insight into these parameters, ie. what they do -
what they should be set to - what the downsides may be in adjusting them?
2. Can anyone point out good documentation on these params?
3. What other things might we look at in terms of problem identification
and resolution?
Thanks in advance. I will summarize.
Regards,
Joe Dierker
_______________________________________________
sunmanagers mailing list
sunmanagers@sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers
This archive was generated by hypermail 2.1.7 : Wed Apr 09 2008 - 23:26:09 EDT