[Veritas-bu] RE: Are Image ID's unique?

Well - here I am replying to my own posting again but here's what we found.

During a transition case from single interface to a dual interface
configuration on a lot of client systems, the new "backend" interface became
the primary interface for data communications ("REQUIRED_INTERFACE" in
bp.conf).  

I'm assuming that the consultant on this task either made or recommended
sym-links in the ../netbackup/db/images directory from the back-end name of
the client to the front-end name of the client - perhaps as a transition
case that went stale after everybody forgot about it.

Anyway - when a lookup of the images for the policy is performed, the image
name is found in two client's directories (one real directory, the other a
sym-link to that dir) and both clients are reported in the output.

The one thing not noticed in the snipped output I sent earlier was that
there was a single-letter change in the client name.

A sort & snip shows this better (IMAGE entries from bpimagelist, cut to the
first six fields, sorted by backupid):

IMAGE ub-13 0 0 6 ue-13_1086427914  <<sym-linked
IMAGE ue-13 0 0 6 ue-13_1086427914
IMAGE ub-13 0 0 6 ue-13_1086427915  <<sym-linked
IMAGE ue-13 0 0 6 ue-13_1086427915
IMAGE ub-13 0 0 6 ue-13_1086427916  <<sym-linked
IMAGE ue-13 0 0 6 ue-13_1086427916
IMAGE ub-13 0 0 6 ue-13_1086427917  <<sym-linked
IMAGE ue-13 0 0 6 ue-13_1086427917

The ue-13 is the "real" client name, the front-end interface.  The ub-13 was
the sym-link in the images DB.

The solution - don't put sym-links in your images DB directory between
client names.  If you do - be aware that the bpimage output may be doubled.

-M

----Original Message-----
From: veritas-bu-admin AT mailman.eng.auburn DOT edu
[mailto:veritas-bu-admin AT mailman.eng.auburn DOT edu]On Behalf Of Donaldson, 
Mark
- Broomfield, CO
Sent: Thursday, June 10, 2004 10:28 AM
To: irodriguez AT arsenaldigital DOT com; veritas-bu AT mailman.eng.auburn DOT 
edu
Subject: RE: [Veritas-bu] RE: Are Image ID's unique?


Nope - no ITC or other duplications on any of our backups.  The FRAG entries
for this output - snipped from what I sent to the list - only shows
single-copies for all images.

I did check some of our regular filesystem backups and they show duplication
too so this isn't unique to the Oracle policies.

-M


 -----Original Message-----
From: Isidro Rodriguez [mailto:irodriguez AT arsenaldigital DOT com]
Sent: Thursday, June 10, 2004 10:13 AM
To: Donaldson, Mark - Broomfield, CO; veritas-bu AT mailman.eng.auburn DOT edu
Subject: RE: [Veritas-bu] RE: Are Image ID's unique?


Are you using Inline Tape Copy? 
Do this on that image to see if multiple copies are being created: 
bpimagelist -L -backupid ue-13_1086497004 
-----Original Message----- 
From: Mark.Donaldson AT cexp DOT com [mailto:Mark.Donaldson AT cexp DOT com] 
Sent: Thursday, June 10, 2004 12:03 PM 
To: veritas-bu AT mailman.eng.auburn DOT edu 
Subject: [Veritas-bu] RE: Are Image ID's unique? 


An (slightly munged for security purposes) example: 
IMAGE ub-13 0 0 6 ue-13_1086497004 P_UE-13_RM_Hot_ 4 *NULL* oracle CONTSCHED

2 0 1086497004 60 1088311404 0 0 6688 1 1 1 0 
P_UE-13_RM_Hot__1086497004_UBAK.f *NULL* *NULL* 0 1 0 0 0  0 0 0 0 0 0 0 
*NULL* 0 0 0 
IMAGE ub-13 0 0 6 ue-13_1086494014 P_UE-13_RM_Hot_ 4 *NULL* oracle FULLSCHED

2 0 1086494014 2896 1088308414 0 0 27010336 1 1 1 0 
P_UE-13_RM_Hot__1086494014_UBAK.f *NULL* *NULL* 0 1 0 0 0  0 0 0 0 0 0 0 
*NULL* 0 0 0 
IMAGE ub-13 0 0 6 ue-13_1086493915 P_UE-13_RM_Hot_ 4 *NULL* root 
Automatic_Scheduled_Backup 0 0 1086493915 3164 1088308315 0 0 0 0 0 0 0 
P_UE-13_RM_Hot__1086493915_FULL.f *NULL* *NULL* 0 0 0 0 0 *NULL* 0 0 0 0 0 0

0 *NULL* 0 0 0 
IMAGE ub-13 0 0 6 ue-13_1086490046 P_UE-13_RM_Hot_ 4 *NULL* oracle CONTSCHED

2 0 1086490046 52 1088304446 0 0 6688 1 1 1 0 
P_UE-13_RM_Hot__1086490046_UBAK.f *NULL* *NULL* 0 1 0 0 0  0 0 0 0 0 0 0 
*NULL* 0 0 0 
IMAGE ue-13 0 0 6 ue-13_1086497004 P_UE-13_RM_Hot_ 4 *NULL* oracle CONTSCHED

2 0 1086497004 60 1088311404 0 0 6688 1 1 1 0 
P_UE-13_RM_Hot__1086497004_UBAK.f *NULL* *NULL* 0 1 0 0 0  0 0 0 0 0 0 0 
*NULL* 0 0 0 
IMAGE ue-13 0 0 6 ue-13_1086494014 P_UE-13_RM_Hot_ 4 *NULL* oracle FULLSCHED

2 0 1086494014 2896 1088308414 0 0 27010336 1 1 1 0 
P_UE-13_RM_Hot__1086494014_UBAK.f *NULL* *NULL* 0 1 0 0 0  0 0 0 0 0 0 0 
*NULL* 0 0 0 
IMAGE ue-13 0 0 6 ue-13_1086493915 P_UE-13_RM_Hot_ 4 *NULL* root 
Automatic_Scheduled_Backup 0 0 1086493915 3164 1088308315 0 0 0 0 0 0 0 
P_UE-13_RM_Hot__1086493915_FULL.f *NULL* *NULL* 0 0 0 0 0 *NULL* 0 0 0 0 0 0

0 *NULL* 0 0 0 
IMAGE ue-13 0 0 6 ue-13_1086490046 P_UE-13_RM_Hot_ 4 *NULL* oracle CONTSCHED

2 0 1086490046 52 1088304446 0 0 6688 1 1 1 0 
P_UE-13_RM_Hot__1086490046_UBAK.f *NULL* *NULL* 0 1 0 0 0  0 0 0 0 0 0 0 
*NULL* 0 0 0 
...sorting out the image id's... 
ue-13_1086497004 
ue-13_1086494014 
ue-13_1086493915 
ue-13_1086490046 
ue-13_1086497004 
ue-13_1086494014 
ue-13_1086493915 
ue-13_1086490046 


...More readable... 
Backed Up         Expires       Files       KB  C  Sched Type   Policy 
----------------  ---------- -------- --------  -  ------------ ------------

06/05/2004 22:43  06/26/2004        1     6688  N  User Backup 
P_UE-13_RM_Hot 
06/05/2004 21:53  06/26/2004        1 27010336  N  User Backup 
P_UE-13_RM_Hot 
06/05/2004 21:51  06/26/2004        0        0  N  Full Backup 
P_UE-13_RM_Hot 
06/05/2004 20:47  06/26/2004        1     6688  N  User Backup 
P_UE-13_RM_Hot 
06/05/2004 22:43  06/26/2004        1     6688  N  User Backup 
P_UE-13_RM_Hot 
06/05/2004 21:53  06/26/2004        1 27010336  N  User Backup 
P_UE-13_RM_Hot 
06/05/2004 21:51  06/26/2004        0        0  N  Full Backup 
P_UE-13_RM_Hot 
06/05/2004 20:47  06/26/2004        1     6688  N  User Backup 
P_UE-13_RM_Hot 




-----Original Message----- 
From: Donaldson, Mark - Broomfield, CO 
Sent: Thursday, June 10, 2004 9:57 AM 
To: Veritasbu (E-mail) 
Subject: Are Image ID's unique? 


I thought the Image ID was unique, the "key" for the Image database.  For my

v4.5 FP4 installation, though, the bpimagelist output, without a doubt, 
seems to be doubling each output line - at least on the backups done through

an Oracle application policy. 
Ideas? 
-M 
================================================= 
  There are 10 kinds of people in the world, 
  those that know binary, and those that don't. 
================================================= 
  Mark Donaldson - SA - Corporate Express 
================================================= 
_______________________________________________ 
Veritas-bu maillist  -  Veritas-bu AT mailman.eng.auburn DOT edu 
http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu