Topic: errors sent to admin/root user mailbox?

[Patrick Basile]

Hello everybody,

I know I've been posting a LOT of problems/questions these past few days...thanks for your patience and help/tips/ideas.  My e-smith 4.1.2 server is better due to all of your responses; however, there are still some issues based on what the admin/root user mailbox had in it when I logged in there tonight.  See the errors/messages below:

Message 1-
Cron /usr/sbin/sarg /etc/sarg/sarg.conf
SARG: No records found
SARG: End

Message 2-
errors rotating logs
errors occured while rotating /var/spool/fax/etc/xferfaxlog {
stat of /var/spool/fax/etc/xferfaxlog failed: No such file or directory

Message 3-
Cron /etc/startmail
fetchmail: another foreground fetchmail is running at 20030.
fetchmail: another foreground fetchmail is running at 20030.
fetchmail: another foreground fetchmail is running at 20030.
fetchmail: another foreground fetchmail is running at 20030.
fetchmail: another foreground fetchmail is running at 20030.
fetchmail: another foreground fetchmail is running at 20030.

Any ideas on why I'm getting these messages to the admin/root user?  If so, what are the fixes?  Thanks a lot.

Regards,
Patrick

[Patrick Basile]

I installed logcheck on the same machine, and came across hundreds of the following messages:

Oct 12 21:26:40 bcsrv1 dhcpd: BOOTREQUEST from 00:04:4d:9d:c1:40 via eth0
Oct 12 21:26:40 bcsrv1 dhcpd: No applicable record for BOOTP host 00:04:4d:9d:c1:40 via eth0
Oct 12 21:26:50 bcsrv1 dhcpd: BOOTREQUEST from 00:03:e3:b6:e2:40 via eth0
Oct 12 21:26:50 bcsrv1 dhcpd: No applicable record for BOOTP host 00:03:e3:b6:e2:40 via eth0
Oct 12 21:27:17 bcsrv1 dhcpd: BOOTREQUEST from 00:04:4d:9d:c1:40 via eth0
Oct 12 21:27:17 bcsrv1 dhcpd: No applicable record for BOOTP host 00:04:4d:9d:c1:40 via eth0
Oct 12 21:27:27 bcsrv1 dhcpd: BOOTREQUEST from 00:03:e3:b6:e2:40 via eth0
Oct 12 21:27:27 bcsrv1 dhcpd: No applicable record for BOOTP host 00:03:e3:b6:e2:40 via eth0
Oct 12 21:27:54 bcsrv1 dhcpd: BOOTREQUEST from 00:04:4d:9d:c1:40 via eth0
Oct 12 21:27:54 bcsrv1 dhcpd: No applicable record for BOOTP host 00:04:4d:9d:c1:40 via eth0
Oct 12 21:28:04 bcsrv1 dhcpd: BOOTREQUEST from 00:03:e3:b6:e2:40 via eth0
Oct 12 21:28:04 bcsrv1 dhcpd: No applicable record for BOOTP host 00:03:e3:b6:e2:40 via eth0
Oct 12 21:28:31 bcsrv1 dhcpd: BOOTREQUEST from 00:04:4d:9d:c1:40 via eth0
Oct 12 21:28:31 bcsrv1 dhcpd: No applicable record for BOOTP host 00:04:4d:9d:c1:40 via eth0
Oct 12 21:28:41 bcsrv1 dhcpd: BOOTREQUEST from 00:03:e3:b6:e2:40 via eth0
Oct 12 21:28:41 bcsrv1 dhcpd: No applicable record for BOOTP host 00:03:e3:b6:e2:40 via eth0

Does anyone know what might be going on here?

Thanks.

Regards,
Patrick

[Adam Rykala]

2 is logrotate complaining that a Hylafax log file doesn't exist

touch /var/spool/fax/etc/xferfaxlog

Will sort that out

3 is ossibly a large email coming in - its fetchmail complaining that another copy of itself is running while a scheduled email download is taking place


1 - not sure

Adam

[Duncan]

Message 2

This i found in the HylaFAX mailing lists

The file /etc/logrotate.d/hylafax is

/var/spool/fax/etc/xferfaxlog {
        rotate 3
        monthly
        compress
}

and should be

/var/spool/hylafax/etc/xferfaxlog {
        rotate 3
        monthly
        compress
}

Regards Duncan

[Patrick Basile]

Guys,

Thanks for your help - I made the change Duncan suggested, and will see whether that fixes the problem.

Regards,
Patrick

[Marcos Migliorini]

Patrick,

Regarding message 3 I have the same situation. Somebody answer you how to fix this?

Thank you very much in advance

Marcos

[Scott Smith]

First, the message is simply fetchmail letting you know that it is not starting another instance of itself since another one is already running. Not an error, just a polite notification -- so you can track down why

The most likely cause is that 1) mail retrieval is scheduled for a short interval, such as five minutes, and 2) fetchmail is still active from the previous session. The causes for #2 are several, but the likely candidates are:

-- slow connection, lots of messages to download
-- slow connection, large message to download
-- fast connection, and really really really lots of or large message(s)

I've also seen the problem occur due to DNS issues. One ISP was "rebuilding" their DNS, and this is where the e-smith server was pointing for primary DNS resolution (dial-up). The mail system would attempt DNS lookups on domains hosted by this "rebuilding" DNS server, and would sometimes take 2-3 minutes to resolve. Multiply this by even two or three messages in one session, and the fetchmail process would run 5-10 minutes. With mail pickup scheduled every 5 minutes, we were seeing these messages frequently! Plus messages in both /var/log/messages and /var/log/qmail/* regarding the DNS problems.

Scott

[Patrick Basile]

Scott,

Thanks for your response, the explanation certainly fits my setup.  Mail retrieval is setup for 5 minute intervals, so that probably explains it.  The server is connected via a frac T1 (192kbps CIR) so it's not a slow connection - but not super fast!  Maybe I'll just kick mail retrieval back to every 15 minutes.  Does that setting impact mail sent within my domain, or will that be delivered asap regardless of the retrieval setting?

I'm curious, how would I look to see if the DNS issue you mentioned is happening on my server?

Thanks for all your help guys.  BTW - I have solved all of the problems mentioned in my first/second post thanks to all the help/tips given here on the phorum.  

Regards,
Patrick

[Marcos Migliorini]

Scott,

Thank you for your explanation. You are right, I am retrieving with a 5 minutes interval!. The next interval value is 15 minutes but i think that for me 10 minutes will be ok so you know if is possible to set this manually?

Thanks

Regars

Marcos