Hi everyone,
We are running the pine imapd server (version 12.252, ~Jun 1999) on a
Alphastation 255 running DU4.0d. Recently, we've been having problems
with a imap client program called Simeon (v4.12 and 4.15) which would
complain of IMAP timeouts when fcc'ing to sent mail folder. The problem
is confusing, and am not sure if client or server problem. I am describing
the problem below. I am primarily interested if anyone else has seen anything
similar or knows of any issues with 4.0d or imapd that might cause this sort
of behavior.
Netscape 4.5 imap client successfully saves the fcc w/out error. Simeon
clients connecting to the same imapd running on a Sun also works.
I captured some of the sessions with a sniffer, but that didn't clarify
anything. I'm a novice at protocol debugging, and can clearly see where the
problem is in the session, but don;t know why. I am suspicious that problem
is on server side. Sessions follow the pattern:
1) Finish smtp connection to send the mail
2) client: IMAP Request: Append to sent-mail
server: IMAP Response: + Ready for data
client sends some IMAP Requests with the data, server ACKs
client: sends an empty IMAP request (presumably to indicate end of data)
3) In the successful cases, the server ACKs the empty IMAP request then
sends an IMAP Reponse: Append completed.
4) In the cases with timeout errors, the server ACKs the empty request, and
nothing else happens for 2 minutes, at which time the client FIN, ACKs. I
had been shutting the browser down after the timeout, and see the IMAP logout
requests, and only after a bit of traffic I think is logout related does the
server respond with Append completed. (The fcc actually generally is
successful, but users are frozen for 2 minutes until get a timeout message.)
I am having trouble pin-pointing when this behavior began, (users only started
complaining of it recently, but say the behavior went back awhile). Neither
the client software or the imap daemon has changed recently, and only change
that is close to coincident is the OS upgrade of the server running imapd from
4.0b to 4.0d (some users claim problem preceded this, but not sure how reliable
their report is).
Again, I am curious if anyone else has seen any behavior like this, or know of
any issues that might cause such. Thanks in advance.
Tom Payerle
Dept of Physics payerle_at_physics.umd.edu
University of Maryland (301) 405-6973
College Park, MD 20742-4111 Fax: (301) 314-9525
Received on Wed Jan 12 2000 - 16:24:04 NZDT