Collect Utilities Tool

[ejung88]

The system I have is a "HP Tru64 UNIX v5.1B" with patch kit 6 "T64V51BB27AS0006-0061208.tar".

When I run the Collect Utilities tool I get the following output to the console:

The command line was "/collectdata/scripts/collect_util_start"

I believe that this may be occuring becasue of certain tasks that are in the crontab file. Is it not possible to have it output to the console as the operator will be confused when this rolls out to production?

00:00:01 Starting collect ...
02:00:01 /collectdata/archive/stn466_02Jul2007_23:58.cgz
02:00:01 var = /collectdata/archive/stn466_02Jul2007_23:58.cgz
02:00:02 YLine #0 - dsk0
02:00:02 Line #0 - /collectdata/archive/stn466_02Jul2007_23:58.cgz
02:00:02 dsk0
02:00:02 n = 0
02:00:15 YLine #1 - dsk1
02:00:15 Line #0 - /collectdata/archive/stn466_02Jul2007_23:58.cgz
02:00:16 dsk1
02:00:16 n = 0
02:00:26 YLine #2 - dsk3
02:00:26 Line #0 - /collectdata/archive/stn466_02Jul2007_23:58.cgz
02:00:26 dsk3
02:00:26 n = 0
02:00:36 YLine #3 - dsk4
02:00:36 Line #0 - /collectdata/archive/stn466_02Jul2007_23:58.cgz
02:00:36 dsk4
02:00:37 n = 0
02:00:46 YLine #4 - dsk5
02:00:46 Line #0 - /collectdata/archive/stn466_02Jul2007_23:58.cgz
02:00:47 dsk5
02:00:47 n = 0
02:00:57 YLine #5 - dsk2
02:00:57 Line #0 - /collectdata/archive/stn466_02Jul2007_23:58.cgz
02:00:57 dsk2
02:00:57 n = 0
02:01:07 /collectdata/archive/stn466_02Jul2007_23:58.cgz
02:01:08 var = /collectdata/archive/stn466_02Jul2007_23:58.cgz
02:01:08 YLine #0 - 0
02:01:08 Line #0 - /collectdata/archive/stn466_02Jul2007_23:58.cgz
02:01:08 0
02:01:08 n = 0
02:01:18 /collectdata/archive/stn466_02Jul2007_23:58.cgz
02:01:18 var = /collectdata/archive/stn466_02Jul2007_23:58.cgz
02:01:18 YLine #0 - root_domain#root
02:01:18 Line #0 - /collectdata/archive/stn466_02Jul2007_23:58.cgz
02:01:18 root_domain#root
02:01:18 n = 0
02:01:27 YLine #1 - usr_domain#usr
02:01:27 Line #0 - /collectdata/archive/stn466_02Jul2007_23:58.cgz
02:01:27 usr_domain#usr
02:01:27 n = 0
02:01:37 YLine #2 - var_domain#var
02:01:37 Line #0 - /collectdata/archive/stn466_02Jul2007_23:58.cgz
02:01:37 var_domain#var
02:01:37 n = 0
02:01:46 YLine #3 - users_domain#users
02:01:46 Line #0 - /collectdata/archive/stn466_02Jul2007_23:58.cgz
02:01:46 users_domain#users
02:01:46 n = 0
02:01:56 YLine #4 - collectdata_domain#collectdata
02:01:56 Line #0 - /collectdata/archive/stn466_02Jul2007_23:58.cgz
02:01:56 collectdata_domain#collectdata
02:01:56 n = 0
02:02:06 /collectdata/archive/stn466_02Jul2007_23:58.cgz
02:02:06 var = /collectdata/archive/stn466_02Jul2007_23:58.cgz
02:02:06 /collectdata/archive/stn466_02Jul2007_23:58.cgz
02:02:06 var = /collectdata/archive/stn466_02Jul2007_23:58.cgz
02:02:07 YLine #0 - tu0
02:02:07 Line #0 - /collectdata/archive/stn466_02Jul2007_23:58.cgz
02:02:07 tu0
02:02:07 n = 0
02:02:17 /collectdata/archive/stn466_02Jul2007_23:58.cgz
02:02:17 var = /collectdata/archive/stn466_02Jul2007_23:58.cgz
02:02:17 n = 0
02:02:27 /collectdata/archive/stn466_02Jul2007_23:58.cgz
02:02:27 var = /collectdata/archive/stn466_02Jul2007_23:58.cgz
02:02:27 n = 0
03:00:00 /collectdata/archive/stn466_02Jul2007_23:58.cgz
03:00:01 var = /collectdata/archive/stn466_02Jul2007_23:58.cgz
03:00:01 n = 0
03:00:03 /collectdata/archive/stn466_02Jul2007_23:58.cgz
03:00:03 var = /collectdata/archive/stn466_02Jul2007_23:58.cgz
03:00:03 n = 0
03:00:04 /collectdata/archive/stn466_02Jul2007_23:58.cgz
03:00:04 var = /collectdata/archive/stn466_02Jul2007_23:58.cgz
03:00:04 n = 0
03:00:05 /collectdata/archive/stn466_02Jul2007_23:58.cgz
03:00:05 var = /collectdata/archive/stn466_02Jul2007_23:58.cgz
03:00:05 n = 0
03:00:06 /collectdata/archive/stn466_02Jul2007_23:58.cgz
03:00:06 var = /collectdata/archive/stn466_02Jul2007_23:58.cgz
03:00:06 n = 0
03:00:07 /collectdata/archive/stn466_02Jul2007_23:58.cgz
03:00:07 var = /collectdata/archive/stn466_02Jul2007_23:58.cgz
03:00:07 n = 0
03:00:08 /collectdata/archive/stn466_02Jul2007_23:58.cgz
03:00:08 var = /collectdata/archive/stn466_02Jul2007_23:58.cgz
03:00:08 n = 0
03:00:10 /collectdata/archive/stn466_02Jul2007_23:58.cgz
03:00:10 var = /collectdata/archive/stn466_02Jul2007_23:58.cgz
03:00:10 n = 0
03:00:13 /collectdata/archive/stn466_02Jul2007_23:58.cgz
03:00:13 var = /collectdata/archive/stn466_02Jul2007_23:58.cgz
03:00:13 n = 0
03:00:13 /collectdata/archive/stn466_02Jul2007_23:58.cgz
03:00:13 var = /collectdata/archive/stn466_02Jul2007_23:58.cgz
03:00:13 n = 0
03:00:14 /collectdata/archive/stn466_02Jul2007_23:58.cgz
03:00:14 var = /collectdata/archive/stn466_02Jul2007_23:58.cgz
03:00:15 n = 0

Thanks in advance.

[Rob Leadbeater]

Hi,

Can you confirm exactly how and where you're starting collect...

If I'm reading your post correctly you're running it from the cron..?

The normal way to run it is by specifying the COLLECT_AUTORUN option in /etc/rc.config as detailed here:

http://h30097.www3.hp.com/collect/collect_details.html

Hope this helps,

Regards,

Rob

[ejung88]

The collect is starting via the cron job.

We must now make a few crontab entries in order to run specific jobs at specific times.

Using the following command, add these lines at the end of the current entries, when complete save:

# crontab â e

# Start Collect Utility
0 0 * * * /collectdata/scripts/collect_util start >/dev/console

[ejung88]

In the /collectdata directory I made 3 more directories.
1. scripts
2. archive
3. reports

The files required for the configuration of the Collect Utilities tool was copied from another UNIX machine.

We must now make a few crontab entries in order to run specific jobs at specific times.

Using the following command, add these lines at the end of the current entries, when complete save:

# crontab â e

# Stop Collect Utility
58 23 * * * /collectdata/scripts/collect_util stop >/dev/console
#
# Start Collect Utility
0 0 * * * /collectdata/scripts/collect_util start >/dev/console
#
# Cleanup (Delete) Collect Capture files that are older than 60 days
0 1 * * * /collectdata/scripts/collect_cleanup >/dev/console
#
# Run cfilt (collect) reports(these files are used as input for cacti and excel)
0 2 * * * /collectdata/scripts/collect_daily_out > /dev/console
#
# Run collect to create reports
0 3 * * * /collectdata/scripts/collect_daily_rep > /dev/console

*Note: At 11:58 PM collect is stopped and the current file is moved to the archive section. Then at 12:00 AM collect will start and a new .cgz file will be created. Then at 1:00 AM the cleanup script will delete everything older than 60 days (this number is flexible).

[Rob Leadbeater]

I'm confused... What exactly is the problem...?

Cheers,

Rob

[Venkatesh BL]

Most of these messages don't seem like usual messages displayed by /usr/sbin/collect. I suspect the wrapper script 'collect_util' that you are using. See if someone is printing these messages from within that script.

[ejung88]

Oh I forgot to post that lol, I keep getting the error

ERROR: IPC failure: Configuration daemon is not accessible.

Is this error expected and can I just ignore it? Or is it something I should take a deeper look into?

[Rob Leadbeater]

Hi,

A quick Google for that error throws up this document:

http://h30097.www3.hp.com/docs/best_practices/lsm_cluster/CLSM_Recovery_Procedure.pdf

"If you see this error when executing a voldisk or volprint command, you are getting this
error because vold is not enabled. If LSM simply was not started, start vold using the
vold â k command."

I suspect that the collect scripts you're running are trying to do some query on LSM, on a none LSM machine...

Hope this helps,

Regards,

Rob

[ejung88]

There is no LSM setup on the machine, we used to have it but turned it off a long time ago because it was UNIX mirroring which was not required on the node (node = stn466).

[ejung88]

I don't believe it's lsm related.

My impression is either some errors (??) was encountered or it's set (somewhere in the script) to send to whoever it submitted the job.

Re-enabling the lsm doesn't help.
Same email received ( at 02:07 am, 03:00 am).

[The problem is that after collect utilities is run, an error email keeps getting sent to the root]