Check_logfiles for status files

Icinga Modules
1

We have an application, which rewrites the log file upon each run. New messages are not appended, but the old messages are cleared before new one are written. How do we monitor this situation? The check_logfiles writes a seek file (recording the already seen log file lines), hence the new entries are not evaluated.

1st try: check_logfiles --logfile ... --rotation virtual : did not give results.
2nd try: check_logfiles --logfile ... --allyoucaneat --nosavestate : unfortunately “nosavestate” is not an command line option.

How do you handle the situation? Except writing a post-script, which deletes the seek file after each run.
Currently the log is not huge, hence --allyoucaneat would be an option. Thanks!

2

Seems, we were looking at the wrong place. The old file is rotated correctly, keeping its i-node number. The file is then re-written, but with a new i-node number. AFAIK check_logfiles is able to handle this situation and it seems the LogMon did not work due to another mistake.
We’ll prove that and mark the thread as solved as soon as this is evident.