Showing results for 
Search instead for 
Did you mean: 


Amanda Deer
Frequent Advisor


The clu_bdmgr.log file on our cluster is getting quite large. Is it ok to null this file - if you view the file it says not to edit?
Venkatesh BL
Honored Contributor

Re: clu_bdmgr.log

The log file captures the log about the clu_bdmgr command execution. I am not sure why you see "DO NOT EDIT" in this file. Can you ensure that line was not from an output?.

I think you should be able to zero the file without any impact.
Ivan Ferreira
Honored Contributor

Re: clu_bdmgr.log

As far i know, the only relevant file generated by this command is the /cluster/members/{memb}/boot_partition/etc/clu_bdmgr.conf file, the configuration of the CNX partition.

So, you can safely remove the file, if you want to "feel" that you are protected, make a tape backup of the entire system, then remove the files that you don't need.
Por que hacerlo dificil si es posible hacerlo facil? - Why do it the hard way, when you can do it the easy way?

Re: clu_bdmgr.log

Normaly system open log-file using the open-option : APPEND

The open command finds the (End Of File)EOF-pointer and start to fill the file.

Usually you don't get more free space if you remove already open file because the owner prosess append deleted file it remember the EOF-pointer ;-)

Best way remove log file is to use command: cat /dev/null > ../log.file

The command "cat..." not change inode-number which is the key to find this ../log.file, it's only clear file contents and keep the file.

Johan Brusche
Honored Contributor

Re: clu_bdmgr.log

When you use the command "clu_bdmgr -d dskXh"
to display the content of the cnx h-partition from a bootdisk or quorum disk, the same output that you see on the screen is appended to /cluster/admin/clu_bdmgr.log .

That's why you see "DO NOT EDIT" in the file.

On our cluster clu_bdmgr.log is 105Kb and begins in Jan 2003.

If your file is much larger, then maybe you have a script that regulary runs clu_bdmgr...
The IMPORTANT info in the logfile are thos lines with:
::TYP:m:CFS:/dev/disk/dsk58h:LSM: .... or
::TYP:q:CFS:/dev/disk/dsk58h:LSM: .... etcetera.
They help to see if someone tried to change the cnx partition, or help to recup your cluster.
So you migth 'grep TYP clu_bdmgr.log > cnx_info', to save if before cleaning the logfile.

___ Johan