Re: OE6 does not connect with pop3 server
From: Robert Aldwinckle (robald_at_techemail.com)
Date: 03/20/04
- Next message: Richard G. Harper: "Re: microsoft eyeing aol"
- Previous message: Jim Pickering: "Re: Question Marks - ???? - in Subject & Address"
- In reply to: Brian Summers: "Re: OE6 does not connect with pop3 server"
- Messages sorted by: [ date ] [ thread ]
Date: Fri, 19 Mar 2004 19:02:53 -0500
> > > The log does not show that it tried to communicate.
> >
> > Try to record an approximate timestamp for these diagnostics
> > so we can see how this message relates to the rest of the log.
> > (It may also be useful for you to keep a log of these with timestamps
> > to present to your ISP when you think that you have sufficient evidence
> > to toss the ball their way.) Also tell us how long your timeout is set to
> > for each server (Properties, Advanced tab)
> My time out is set for two minutes
...
> I don't know how to show the failing attempts.
Let's use one of your previous posts as an example:
< news:ejWg8kQDEHA.2800@tk2msftngp13.phx.gbl >
Subject: Re: OE6 does not connect with pop3 server
Date: Thu, 18 Mar 2004 09:51:14 -0700
<summary of connect requests>
Ok this really getting on my nerves. OE is set to check in every 10 minutes.
Here is a copy of my log that shows it connected on its own automatically:
POP3: 08:27:19 [rx] +OK NGPopper vEL_5_7 at earthlink.net ready
...
POP3: 08:46:07 [rx] +OK NGPopper vEL_5_7 at earthlink.net ready
...
POP3: 08:56:10 [rx] +OK Qpopper (version 4.0.4) at vmbsd09.viewmark.com
...
POP3: 08:56:39 [rx] +OK NGPopper vEL_5_7 at earthlink.net ready
</summary of connect requests>
Can we infer from this that you got a timeout on the EarthLink account
at about 08:38? (08:36 + 2)
What happened to the other account? Was it timing out approximately
2 minutes after each of the successful connections on the EarthLink
account? IOW the 08:56 record for it was the only success of 4 tries?
BTW please clarify the timeout symptom. Are you getting that prompt
that I mentioned (choice of Wait or Stop) or something else?
My point about making an approximate timestamp is just to look at the
clock and type it in whenever you capture a timeout message.
I agree again that it is unhelpful that there is no better way of recording
this information.
As an alternative to tracing with netcap you might be able to glean some
clues using netstat. E.g. after the timeout does this show you any
residual information about the attempt:
netstat -a -n -o
The -o switch shows you the PID of the originating task.
You could use tasklist to show you the PID of your OE task:
tasklist /fi "IMAGENAME EQ MSIMN.EXE"
Then use that number to filter out the netstat output.
For purposes of example let's say that you found out msimn's PID
was 1234. Then the pipeline for the filter would be:
netstat -a -n -o | find "1234"
You might also be able to use netstat's statistics report to bracket those
scheduled connections. E.g. as near as possible before the connection
attempt occurs enter:
netstat -s >before.txt
and again after the connection
netstat -s >after.txt
To make the data useful you would want at least 2 sets--the connection
working normally and the connect request timing out. Temporarily stop
the second account so that you are collecting statistics for only the one
problem account and save the statistics only for the (normal) case where
no messages are received.
But if you are going to go to that much effort I think it would be easier
and potentially much more informative to also have a trace of the actual
packets involved by starting and stopping netcap on the same basis.
I.e., as near as possible before the connection attempt is expected
start a netcap capture, then after you get the timeout message,
stop the capture. I'm not sure but it may even be possible to leave
netcap running with some kind of wrapped trace active so that all you
would have to do then is stop the trace after a timeout occurred. TBD.
To summarize and analyse the data you capture you could use
fc to compare the two netstat reports. E.g.
fc before.txt after.txt >diff.txt
notepad diff.txt
You'd probably want to do a diff of the diffs too, in order to help see
the differences between your normal case and the failing case.
Netcap's .cap files can be formatted by Ethereal or netmon (if you happen
to have access to Windows server). Netcap itself apparently is installed
when you install the XP Support tools.
HTH
Robert
--- "Brian Summers" <tinman_63@earthlink.net> wrote in message news:Oq9Y2gdDEHA.3788@TK2MSFTNGP10.phx.gbl... > Robert, > My time out is set for two minutes, I have increased it from the one minute > default setting. When this thing acts up, it doesn't matter whether it is an > auto connect or a manual. I have manually tried to connect for as much as 4 > hours before with no luck. I didn't have this problem until last week and I > have been using Norton AV and Norton Internet Security Pro since December. > As for time stamps, The log I included shows when the program connects to > the server, but I don't know how to show the failing attempts. If you have > any info on that I would greatly appreciate it. POP.Viewmark will not admit > it is their problem. They tell me I am the only one experiencing this > problem with them. > Thanks, > Brian > "Robert Aldwinckle" <robald@techemail.com> wrote in message > news:%23%23CibbWDEHA.3408@tk2msftngp13.phx.gbl... > > "Brian Summers" <tinman_63@earthlink.net> wrote in message > > news:OgOdy5PDEHA.624@TK2MSFTNGP10.phx.gbl... > > > PA Bear, > > > As usual it worked all day yesterday but, today it failed. Here is the > error > > > message I received: > > > A time-out occurred while communicating with the server. Account: > > > 'pop.viewmark.com', Server: '216.38.205.239', Protocol: POP3, Port: 110, > > > Secure(SSL): No, Error Number: 0x800CCC19. > > > The log does not show that it tried to communicate. > > > > Try to record an approximate timestamp for these diagnostics > > so we can see how this message relates to the rest of the log. > > (It may also be useful for you to keep a log of these with timestamps > > to present to your ISP when you think that you have sufficient evidence > > to toss the ball their way.) Also tell us how long your timeout is set > to > > for each server (Properties, Advanced tab) > > > > > > In any case I think that I have a similar intermittent problem. > > Also on XPsp1. I reported it to my ISP and they claimed that it was > > a "known problem" with the XP firewall. I disabled the XP firewall > > and didn't notice sufficient change to justify leaving it disabled. > > > > I deal with the problem by having the server timeout as quickly as > possible > > (30 seconds). That allows me to either retry if I want or more usually > > just Stop the operation and retry manually later. Since I usually > combine > > connecting with sending and receiving my speculation would be that > > the timeout could be associated with the link not being up completely. > > I don't think it ever happens when I retry manually after the link is up. > > > > It isn't enough of an annoyance for me to bother diagnosing it further > > but if I was going to investigate it I would use netcap instead of OE's > log. > > As you have observed the latter doesn't record failed connection attempts. > > > > > > HTH > > > > Robert Aldwinckle > > --- > > > > > > > My POP.viewmark account is configured to be the first one connected too. > > > My outlook express is configured to check two accounts. > > ... > > > > > > > >
- Next message: Richard G. Harper: "Re: microsoft eyeing aol"
- Previous message: Jim Pickering: "Re: Question Marks - ???? - in Subject & Address"
- In reply to: Brian Summers: "Re: OE6 does not connect with pop3 server"
- Messages sorted by: [ date ] [ thread ]
Relevant Pages
|
