• Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 361
  • Last Modified:

POP3 Keeps timing out

i have a wierd issue where my pop3/sendmail server keeps dropping pop3 connections.  If i do a telnet to it on port 110, half hte time it will come up (sometimes very fast, sometimes very slow) and the other half it will time out.  I did a 'ps aux' on the box and i found a few entries like this:

1176      6774  2.0  0.3  4152 1768 ?        S    15:58   0:00 ipop3d
1468      6776 14.5  0.3  4044 1608 ?        D    15:58   0:00 ipop3d
1320      6779  4.0  0.3  4060 1652 ?        S    15:58   0:00 ipop3d
1429      6658  0.2  0.0     0    0 ?        Z    15:57   0:00 [imapd <defunct>]
root      6685  0.0  0.2  4028 1468 ?        S    15:57   0:00 ipop3d

not sure what the defunct means but this is causing enough of a problem for all of my users to notice.  smtp still responds though it is a tad slower too (it always responds, it just sometimes takes a while).

i've no idea what the hell is going on but the box is using a load average of 2 or so so its not under that high a load.  any ideas on how i can troubleshoot this?  
0
onsite_tech
Asked:
onsite_tech
  • 2
1 Solution
 
PsiCopCommented:
POP3 and sendmail are completely unrelated. POP3 is a client/server protocol for a mail reader program to download mail from a mailbox. Never talks to sendmail - sendmail doesn't know what POP3 is.

Sendmail is a mail router that uses SMTP (usually, but can use UUCP and a few other things) to route mail between servers. Clients don't use SMTP to access their mailboxes.

What is the OS and hardware? I'm guessing a *NIX.

What is the POP3 server? I'm guessing UofW PINE is where it came from.

Have you eliminated layer 1/2 issues, such as bad NICs, cables, switch ports, duplex settings, etc. ?
0
 
onsite_techAuthor Commented:
yeah, i tracked down the issue.  ironically, the reason the pop3 was freaking out was because the server had some downtime the day before (because of smtp issues) and the clients were hammering the box out of paranoia.  the default inetd.conf settings for how many connections you can do are a little low i guess cause i upped those and everything came up fine and happy.  i guess we've been skirting the limit for a while now but never had *all* of our users on at once like that day.

Here are the lines i added to the ipop3 file:
instances       =150
cps             =70 30


Everything is fine now and its actually very happy a box...
0
 
onsite_techAuthor Commented:
can you PAQ this and throw my last comment on as the accepted comment?  thats what fixed it...
0
 
moduloCommented:
PAQed with points refunded (500)

modulo
Community Support Moderator
0

Featured Post

Technology Partners: We Want Your Opinion!

We value your feedback.

Take our survey and automatically be enter to win anyone of the following:
Yeti Cooler, Amazon eGift Card, and Movie eGift Card!

  • 2
Tackle projects and never again get stuck behind a technical roadblock.
Join Now