Veritas-bu

[Veritas-bu] RE: Queued vs Active Jobs - What am I missing here?

2005-01-05 11:31:24
Subject: [Veritas-bu] RE: Queued vs Active Jobs - What am I missing here?
From: MDobbertien AT tribune DOT com (Dobbertien, Matthew)
Date: Wed, 5 Jan 2005 10:31:24 -0600
In my experience, Global settings always override local client settings.
In addition, changing the global setting may require you to restart the
database service on the master server to take effect.  

I would change your global setting to something more in line with your
absolute maximums (we set ours to 16).  Don't worry about multiplexing
unless you need to get more jobs to a single tape drive.

Matthew Dobbertien
Chicago Tribune
Sr. Systems Analyst

-----Original Message-----
From: Mark.Donaldson AT cexp DOT com [mailto:Mark.Donaldson AT cexp DOT com] 
Sent: Wednesday, January 05, 2005 9:38 AM
To: veritas-bu AT mailman.eng.auburn DOT edu
Subject: [Veritas-bu] RE: Queued vs Active Jobs - What am I missing
here?

Here I am replying to my own post again but I did find that the Global
max-jobs-per-client is set to 4.  I only added this client to the client
DB
(bpclient -add) this morning and changed this client's individual
max/client
to 12.  Perhaps there's a timing issue here and the setting has not
gotten
picked up yet. 

By the way, this is a simple client so there's no client daemons to
stop/restart.

-M

-----Original Message-----
From: Donaldson, Mark - Broomfield, CO 
Sent: Wednesday, January 05, 2005 8:30 AM

I've got a Standard policy and an archive-schedule within it.  I have
the
jobs-per-policy set to 8 and, strangely, I only have four active jobs
and 16
queued jobs of the 20 jobs started simultaneously.

I'm running short on sleep but I don't see what I missed here.

Max-jobs-per client = 12
Max-jobs-per-policy = 8
Destination Storage Unit (Disk) Concurrent Jobs = 99

I can't think of any other throttles that would be limiting me to only
four
active jobs.  There isn't 95 other jobs using this STU right now, nor
any
other jobs on this client that might bump into the max-jobs-per-client
limit.  This is the only client in the policy so there's no other jobs
on
other clients that might bump the max-jobs-per-policy.

Ideas?

-M


>bpclient -client ug-60 -L
Client Name: ug-60
 Current Host:
        Hostname: ug-60
        IP Address: 10.XXX.XXX.XXXX
 Connect on non-reserved port:  no
 No call-back connections:      no
 Dynamic Address:       no
 Free Browse:   Allow
 List Restore:  Not Specified
 Max Jobs This Client:  12



>bppllist P_UG-60 -U
Policy Name:       P_UG-60

  Policy Type:         Standard
  Active:              yes
  Effective date:      09/24/2004 15:36:53
  Client Compress:     no
  Follow NFS Mounts:   yes
  Cross Mount Points:  no
  Collect TIR info:    no
  Block Incremental:   no
  Mult. Data Streams:  no
  Client Encrypt:      no
  Checkpoint:          no
  Policy Priority:     10
  Max Jobs/Policy:     8
  Disaster Recovery:   0
  Residence:           DISK
  Volume Pool:         Oracle
  Keyword:             (none specified)

  Schedule:          arch
    Type:            User Archive
    Maximum MPX:     1
    Synthetic:       0
    PFI Recovery:    0
    Retention Level: 0 (3 weeks)
    Number Copies:   1
    Fail on Error:   0
    Residence:       DISK
    Volume Pool:     (same as policy volume pool)
    Daily Windows:



>bpstulist -label DISK -U

Label:             UG50-DISK
Storage Unit Type: Disk
Host Connection:   uscobrmfa-ug-50
Concurrent Jobs:   99
On Demand Only:    yes
Max MPX:           1
Path:              "/diskbackup"
Max Fragment Size: 2000 MB

=================================================
  "They have computers, and they may have other 
  weapons of mass destruction." - Janet Reno
=================================================
  Mark Donaldson - SA - Corporate Express
=================================================


<Prev in Thread] Current Thread [Next in Thread>