Subject: | |
From: | |
Reply To: | HOFMEISTER,JAMES (HP-USA,ex1) |
Date: | Fri, 22 Dec 2000 07:28:46 -0800 |
Content-Type: | text/plain |
Parts/Attachments: |
|
|
Hello 3000-l folks,
Re: SWITCHNMLOG Question
-------------------------------------------------David Burney writes--
Here's my problem. If I issue a LISTF [log in to unmask],2 I see this :
ACCOUNT= SYS GROUP= PUB
FILENAME CODE ------------LOGICAL RECORD----------- ----SPACE----
SIZE TYP EOF LIMIT R/B SECTORS #X MX
NMLG0424 NLOG 1024W FB 21 21 1 176 1 *
NMLG0425 NLOG 1024W FB 8 8 1 64 1 *
NMLG0426 NLOG 1024W FB 1 1 1 16 1 *
NMLG0427 NLOG 1024W FB 1 1 1 16 1 *
NMLG0428* NLOG 1024W FB 2047 2048 1 16384 1 *
Which tells me that the current log file is NMLG0428. And it has a
number of records in it that I want to view using NMDUMP. So, since I
can't view the file whilst it's being accessed exclusively, I issue
SWITCHNMLOG 428. Which, as far as I understand things, should create
a NEW log file (429) leaving NMLG0428 with an EOF of 2047. The
records I wish to peruse. Alas, this is what I get when I issue a
LISTF [log in to unmask],2 :
ACCOUNT= SYS GROUP= PUB
FILENAME CODE ------------LOGICAL RECORD----------- ----SPACE----
SIZE TYP EOF LIMIT R/B SECTORS #X MX
NMLG0424 NLOG 1024W FB 21 21 1 176 1 *
NMLG0425 NLOG 1024W FB 8 8 1 64 1 *
NMLG0426 NLOG 1024W FB 1 1 1 16 1 *
NMLG0427 NLOG 1024W FB 1 1 1 16 1 *
NMLG0428 NLOG 1024W FB 1 1 1 16 1 *
NMLG0429* NLOG 1024W FB 2047 2048 1 16384 1 *
Now the NEW file has the EOF of 2047 and the old file an EOF of 1.
----------------------------------------------------------------------
The answer to this confusion is the NM logfile is opened and the EOF
pushed out (I suspect for performance seasons) and when the logfile is
closed the EOF is pushed back to the actual number of records present
in the logfile. When a NMLG file is open it will always say it has
an EOF of 2047. The fact that you only have 1 record written to the
log file indicates to me you do not have much logging enabled... The
logging for the NS-Transport is configured in NMMGR:
NMMGR
[open config]
[ns]
[unquided config]
[goto logging]
Class Console Disk
Subsystem Name Logging Logging Event
SUB0003 CLAS0001 [Y] [Y] Serious internal error
Network CLAS0002 [Y] [Y] Internal error/operator attention
Transport CLAS0003 [N] [Y] Non-critical errors
CLAS0004 [Y] [Y] Nodal messages (start/stop)
CLAS0005 [N] [Y] Informative messages
CLAS0006 [N] [Y] Statistical information
SUB0005 CLAS0000 [Y] [Y] Internal errors
Network CLAS0001 [Y] [Y] Resource errors
IPC CLAS0002 [N] [Y] Informative messages
If you are trouble shooting a NS-Transport problem you will want to specify
[Y] for Disk Logging for all CLASS for SUB0003.
If you are trouble shooting a NETIPC problem you will want to specify [Y]
for Disk Logging for all CLASS for SUB0003 and all CLASS for SUB0005.
Then you perform a ":switchnmlog update" to enable the logging.
Note: The logging messages logged are documented in the NS3000/iX Error
Messages Reference Manual with Cause and Action for each message.
I hope this helps.
Regards,
James Hofmeister
Hewlett Packard
Worldwide Technology Network Expert Center
P.S. My Ideals are my own, not necessarily my employers.
|
|
|