Dear List,
I recently updated my core router, which changed the MRTG 'target' IDs for
my interfaces. After roughly 10 days, I restarted the cron daemon which
runs MRTG.
Unfortunately, I discovered that before I updated my config files, the cron
job running MRTG had ran once after the MRTG workstation was rebooted. In
that one instance, the MRTG box believed the year was 2008. This created
quite a problem with the graphs. . .
To fix this, I copied the .old log files to .log and restarted MRTG (thus
the log files would be roughly 10 days late). This worked for the majority
of interfaces, but a few have a problem where the daily graph is working,
but the weekly, monthly, and yearly graphs are flat at roughly 60 kbps.
I have looked at the MRTG .log and .old files, but cannot find the problem.
I have also reviewed the MRTG Logfile Format FAQ.
I would like to know if there is a particular format that the .old and .log
files must follow in terms of weekly, monthly, and then yearly data
entries. I understand from the FAQ and the list archives that after some
time the daily 5 minute data is averaged and then removed; and so on for
weekly, monthly, yearly.
Last note: the particular log file in question does have many lines of data
at the end of the file with 0 0 0 0 for data, but with valid timestamps.
This seems odd.
Details are below. Any assistance would be greatly appreciated.
Sincerely,
-David C
~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~
Pertinent Details:
MRTG version 2.8.6 on Redhat Linux 6.0
(I know, I know; I have an upgrade scheduled for later this year)
tul1-s1 is a 1.5 Mbps T-1 link.
~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~
Excerpt from tul1.cfg.1
WorkDir: /home/httpd/html/mrtg
WriteExpires: Yes
WithPeak[_]:wmy
Options[_]: bits
#
#
Target[tul1-s1]: 2:railrd@10.250.8.1
MaxBytes[tul1-s1]: 192000
Title[tul1-s1]: tul1 : Serial0/0 - Sprint Frame Relay
PageTop[tul1-s1]: