This looks like it's higher in the pipeline.  tcpdump doesn't show any
traffic until after the timeout threshold has been exceeded.

Now I guess the question is this: why do those packets get queued for so
long?  There *is* a txqueuelen: 1000 on the interface, but the same is
set on the 5 VMs that work.

It looks like this is off topic now, so I'll leave it at this.

Cheers,

C.J.

On Sun, 2007-03-18 at 09:42 -0700, C.J. Adams-Collier wrote:
> Hey all,
> 
> I posted this to the IRC channel, but it seems folks are snoozin'.  I
> thought I'd put it in front of a few more eyes.
> 
> Since I posted this, I've removed the IP address specifier
> from /etc/default/snmp (these are debian machines).  Therefore all of
> the config files are exactly the same across all VMs.
> 
> Thanks in advance for any help you might be able to provide,
> 
> Cheers,
> 
> C.J.
> 
> 
> 08:59 < cj> I've got the snmpd running okay on a few hosts, but my name 
> servers 
>             don't seem to be responding correctly to the requests sent by the 
>             following:
> 08:59 < cj> cfgmaker --enable-ipv6 --ifref=ip --noreversedns --community 
>             colliertech --output ~/tmp/mrtg.cfg --dns-domain colliertech.org 
> ns1
> 09:00 < cj> I did the same with my mx0 and mx1, web server and database server
> 09:01 < cj> they all seem to be functioning correctly, and their respective 
>             /etc/snmp/snmpd.conf and /etc/default/snmp are basically the same
> 09:01 < cj> the difference is that each of them is listening on a different 
> IP 
>             address.
> 09:01 < cj> when I run the above command, I get the following in my syslog:
> 09:01 < cj> Mar 18 15:55:36 ns1 snmpd[1071]: Connection from UDP: 
>             [66.152.65.3]:32851 
> 09:01 < cj> Mar 18 15:55:36 ns1 last message repeated 92 times
> 09:02 < cj> but not until after the 10 seconds required for cfgmaker to give 
> up
> 09:03 < cj> the memory on the name servers is a bit more constrained than on 
>             most of the other systems, so I think that *might* be it
> 09:03 < cj> the only vm with less memory than the name servers is the mx1, 
>             which gets a lot less traffic
> 09:04 < cj> anyone know what might be causing this?
> 09:31 < cj> bumping up the memory didn't do it
> 
> 
> 


-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
Net-snmp-users mailing list
[email protected]
Please see the following page to unsubscribe or change other options:
https://lists.sourceforge.net/lists/listinfo/net-snmp-users

Reply via email to