From: (no name) (no email)
Date: Fri Feb 27 2004 - 10:29:26 EST
On Fri, 27 Feb 2004, Joshua E Warchol wrote:
> T A 5 10 20 40 80 160 320 320+
> TOTAL 249 5 0 1 6 17 15 18 3 184
> cargo.com.sv 14 0 0 0 0 0 1 0 0 13
> hkelleher.com 8 0 0 0 0 0 0 0 0 8
> gse.harvard.edu 7 0 0 0 0 0 1 4 0 2
> osu.edu 6 0 0 0 0 0 1 0 0 5
> guaranteedrate.com 6 0 0 0 0 0 0 0 0 6
> weitzlux.com 5 0 0 0 0 0 0 0 0 5
> grinteksa.com.ar 5 0 0 0 0 0 0 0 0 5
> ibolaw.com 5 0 0 0 1 0 0 0 0 4
> peoplepc.com 4 0 0 0 0 0 1 0 0 3
> cyber.net.pk 4 0 0 0 0 0 0 0 0 4
> kmcnetwork.org 4 0 0 0 0 0 0 0 0 4
> padway.com 4 0 0 0 0 0 0 0 0 4
> triumphhospitality.com 4 0 0 0 0 0 0 0 0 4
> gruporemor.com.sv 4 0 0 0 0 0 0 0 0 4
> lbcclaw.com 3 0 0 0 0 0 0 0 0 3
> edit.nydailynews.com 3 0 0 0 0 0 0 0 0 3
> domain.com 3 0 0 0 0 0 0 0 1 2
> mindspring.com 3 0 0 0 0 0 1 0 0 2
> ...
No obvious spikes. 60 messages deferred in last 320 minutes, 184 in last
5 days, so the errors are transient (the long term deferral rate is ~10%
of the short term rate). What is the total number of "timeout" defer
reasons in the "mailq" output? Are these concentrated among just a few
domains. It is possible that the destination in question is having email
congestion problems.
To rule out network problems look for TCP retransmission with "tcpdump",
which is different from SMTP timeouts of a healthy TCP session (your
packets are ACKed, but the remote SMTP server does not respond).
-- Viktor.
|
|
|