log file location?

your commands are setting the log level for NDMP.

You will get these log lines using the server_log server_2 command

Almost all relevant Celerra logging is access via the "server_log" command and not via direct file access.

What information are you trying to gather, as this could generate a lot of mostly useless data..

we are trying to diagnose our NDMP issue.

working with our backup vendor, they requested that we enable those logging levels.

it also states to use:

server_log server_2 -a -s >server_2.log

where would it put the .log file?  when i set the NDMP and PAX logging level back, will this log file still keep growing?  I do not know what that line will do - is there a second line i need to run to clear what i did.

Thanks.

The ">" operator is a standard unix redirection.  All that will do is take the output of the command "server_log server_2 -a -s" and rather than display it on your screen send it to  file called server_2.log in whatever directory you are in when it was ran.  Alternately you can specify a full path like this "server_log server_2 -a -s > /path/to/file/server_2.log".

The -a argument forces server_log to show the complete log file and -s formats the date/time.

When you are done you'll want to set logging back to default. Just replace the LOG_DBG2 with LOG_ERR in your .server_config commands.

thanks for the update.

do i need to undo this command somehow?

server_log server_2 -a -s > /path/to/file/server_2.log

or when will it it stop logging to that file?

Thanks.

Do a „man server_log“ – it explains the command and the switches

if you want to watch the log "live" (same as tailing a file), you can run server_log server_2 -f . If you simply want to redirect output of what's already been written to the log then you run the command Mark provided. If you do not specify any path it will be sent to a file in your current path.

thanks everyone for the details.

BJC,

I'm not certain if the the backup vendor also mentions resetting them to the original defaults; however, you will probably want to reset them to: logsys level 6.  That corresponds to log level: LOG_PRINTF.

You can verify first your original defaults/settings as follows (maybe too late if you already changed them) and these are steps I would personally recommend before even consider modifying them:

.server_config server_2 -v "logsys get severity NDMP"

.server_config server_2 -v "logsys get severity PAX"

This will return a number for the log severity of the corresponding "facility": NDMP, PAX, etc.  If before modifying it you ran the command (on a default install), it would return "6" which, as already noted above, corresponds to "LOG_PRINTF".

Therefore, once done with the troubleshooting so as not to overwhelm your logs with the increased log level, run the following to reset them to original defaults:

.server_config server_2 "logsys set severity NDMP=LOG_PRINTF"

.server_config server_2 "logsys set severity PAX=LOG_PRINTF"

Anything less than 6 (such as LOG_ERR which corresponds to log level 3), for all intense and purposes, basically disables it which you probably don't want as you want some level of logging.  LOG_DBG2 corresponds to logsys level 10 - just an FYI but you can see that running the "get" subcommand above.  6 is a happy compromise and has been our default for some time now.

Good catch

Alternatively you can just reboot the DM to get back to the default settings

Rainer,

Very cool.  I didn't realize rebooting the data mover also reset it back to default.  Thanks.