Backup Central

Hi,

Was running 'nsrinfo client > outputfile' and it's stopped adding
information to the output file. There's plenty of space remaining on the
drive where I'm redirecting the output, and the last line in the output
is a typical nsrinfo output line, but incomplete as part of the date is
missing and it's just sitting there. It doesn't appear to have finished
with the normal summary lines.

The size of the output file has not changed in over an hour. Something
seems wrong here. I want to move this index to another drive where I
have more space, but wanted to compare nsrinfo output before and after
to ensure new location is okay. Man. Huge indexes take so darn long to
check maybe it's not possible before backups run? How do you run nsrinfo
twice before backups kickoff on huge indexes without postponing the
backup?!

Size of index is about 5.6 GB. Size of output file is still at:
2147483647 bytes. daemon.log, messages and system log show nothing
suspect at all. Whenever I run something like 'truss -o outputfile -p -v
all pid', and then do something like head -100 output file and comparing
with previous runs, output looks different? But original output from
nsrinfo still locked at 2147483647 bytes?

Running 6.1.1 on Solaris 2.8. Anyone seen this behavior before?

George

Note: To sign off this list, send a "signoff networker" command via email

In regard to: [Networker] nsrinfo command hung -- need help?, George...:

Size of index is about 5.6 GB. Size of output file is still at:
2147483647 bytes. daemon.log, messages and system log show nothing
suspect at all. Whenever I run something like 'truss -o outputfile -p -v
all pid', and then do something like head -100 output file and comparing
with previous runs, output looks different? But original output from
nsrinfo still locked at 2147483647 bytes?

That's a pretty good clue. Notice anything interesting about that number?
It's 2^31 - 1. Either the tool (nsrinfo) or the filesystem isn't dealing
with numbers that reach the 2 Gig barrier.

Tim
Tim Mooney mooney < at > dogbert.cc.ndsu.NoDak.edu
Information Technology Services (701) 231-1076 (Voice)
Room 242-J6, IACC Building (701) 231-8541 (Fax)
North Dakota State University, Fargo, ND 58105-5164

Note: To sign off this list, send a "signoff networker" command via email

Hi George,
From the figures you've quoted, I'd imagine that you've hit the 2GB
limit on your output file.
The version of Networker you're running will be compiled 32 bit, so may
not be able to deal with larger files than this.

It's probably too late now, but for future you should be able to work
around this by creating a file-system pipe, setting up a process to
accept input from the pipe and send it to a "split" command, and then
running the nsrinfo command with the output going into the pipe. I've
seen this done to overcome a similar problem with Oracle Export files.

Will

George Sinclair wrote:

Hi,

Was running 'nsrinfo client > outputfile' and it's stopped adding
information to the output file. There's plenty of space remaining on the
drive where I'm redirecting the output, and the last line in the output
is a typical nsrinfo output line, but incomplete as part of the date is
missing and it's just sitting there. It doesn't appear to have finished
with the normal summary lines.

The size of the output file has not changed in over an hour. Something
seems wrong here. I want to move this index to another drive where I
have more space, but wanted to compare nsrinfo output before and after
to ensure new location is okay. Man. Huge indexes take so darn long to
check maybe it's not possible before backups run? How do you run nsrinfo
twice before backups kickoff on huge indexes without postponing the
backup?!

Size of index is about 5.6 GB. Size of output file is still at:
2147483647 bytes. daemon.log, messages and system log show nothing
suspect at all. Whenever I run something like 'truss -o outputfile -p -v
all pid', and then do something like head -100 output file and comparing
with previous runs, output looks different? But original output from
nsrinfo still locked at 2147483647 bytes?

Running 6.1.1 on Solaris 2.8. Anyone seen this behavior before?

George

--
Note: To sign off this list, send a "signoff networker" command via email



Note: To sign off this list, send a "signoff networker" command via email

Hi George,
From the figures you've quoted, I'd imagine that you've hit the 2GB
limit on your output file.
The version of Networker you're running will be compiled 32 bit, so may
not be able to deal with larger files than this.

It's probably too late now, but for future you should be able to work
around this by creating a file-system pipe, setting up a process to
accept input from the pipe and send it to a "split" command, and then
running the nsrinfo command with the output going into the pipe. I've
seen this done to overcome a similar problem with Oracle Export files.

Was running 'nsrinfo client > outputfile' and it's stopped adding

Of course, in a way he's already done part of that. Due to the
redirection, nsrinfo isn't dealing with the file offsets directly, so
it's not the source of the 2G failing. Instead the redirection makes
the shell responsible for the offsets.

It is possible that simply using a different (largefile aware) shell
would be sufficient.

Darren Dunham ddunham < at > taos.com
Senior Technical Consultant TAOS http://www.taos.com/
Got some Dr Pepper? San Francisco, CA bay area
< This line left intentionally blank to confuse you. >

Note: To sign off this list, send a "signoff networker" command via email