Bug 25610

For what it's worth, I have been getting these for a long time, long before updating inetd...

Can you find what pids it spawned in /var/log/secure?  (I don't have any 6.2
boxes around here right now...)

All of these (I checked about 6-7 of them, about one or two a day) were ipop3d.

Using:

imap-2000-2.6

Notes:  the client giving exit status 1 uses the service about 50-100 times for every exit.  He's using a 
microsoft mailer, probably messenger.  There are some people using RHL52 fetchmail to access ipop3d, who
haven't shown the symptoms, but the use is relatively less, so...

What I meant to ask was: Can you from information in one of the log files find
out what server maps to which pid?

Perhaps I don't still understand the question, but all 'exit 1' messages I had, 
have come from ipop3d (based on /var/log/secure and messages pids).

OK, as I understand it you can then match the exit status from "messages" with
the pids from "secure". Then it's useful debugging info - closing.

it sounds like this message is new as of some changes after rh6.2 --
'cause i wasn't seeing the log message before whatever version of inetd
was "most recent" on 6.2 prior to this bugfix.

regarding ipop3d -- if that wasn't tcp wrapped you wouldn't have a
/var/log/secure entry ... if the inetd log message contained a port
number that'd be sufficient for making it handle wrapped and unwrapped
services.

It's not new - there are only two changes in that RPM, the "don't leak file
handles patch" from pekkas and "spell RETVAL correctly" in the init script.

I am still getting tons of the following:

Feb 23 14:32:59 gonzo inetd[21492]: pid 5962: exit signal 13
Feb 23 14:33:03 gonzo inetd[21492]: pid 7123: exit status 1
Feb 23 14:33:09 gonzo inetd[21492]: pid 7125: exit status 1
Feb 23 14:33:17 gonzo inetd[21492]: pid 7130: exit status 1
Feb 23 14:33:22 gonzo inetd[21492]: pid 7132: exit status 1
Feb 23 14:33:27 gonzo inetd[21492]: pid 7133: exit status 1
Feb 23 14:33:33 gonzo inetd[21492]: pid 7134: exit status 1
Feb 23 14:33:38 gonzo inetd[21492]: pid 7135: exit status 1
Feb 23 14:33:43 gonzo inetd[21492]: pid 7136: exit status 1
Feb 23 14:33:48 gonzo inetd[21492]: pid 7137: exit status 1
Feb 23 14:33:54 gonzo inetd[21492]: pid 7138: exit status 1
Feb 23 14:33:58 gonzo inetd[21492]: pid 7139: exit status 1
Feb 23 14:34:02 gonzo inetd[21492]: pid 7140: exit status 1
Feb 23 14:34:08 gonzo inetd[21492]: pid 7141: exit status 1
Feb 23 14:34:15 gonzo inetd[21492]: pid 7142: exit status 1
Feb 23 14:35:27 gonzo inetd[21492]: pid 2184: exit signal 13

These are all from imapd.

This started the day I upgraded to inetd-0.16-7 (I have been running
imap-2000-2.6 for many months)


How is this not a bug??

thanks,

-Brian

I should note that many of our Outlook IMAP users have reported weird hanging
and slowdowns that seemed to have started after the upgrade to inetd-0.16-7

-Brian

Does anyone check this?

I've gotten that too, after and before upgrades, but I haven't heard any
complaints about it.

This appears to be debugging information (inetd.c):
---
  	          if (debug) {
                        fprintf(stderr, "pid %d, exit status %x\n", pid,   
  	                          status);
  	        }
---

I don't see logging non-zero exit codes as a bug in inetd - if you're not
interested, don't look for them or tell syslog not to care.