Veritas-bu

[Veritas-bu] media ID not unique after bprecover - 3.4 to 4.5 DR recovery caveat

2004-04-09 14:04:29
Subject: [Veritas-bu] media ID not unique after bprecover - 3.4 to 4.5 DR recovery caveat
From: SJACOBSO AT novell DOT com (Scott Jacobson)
Date: Fri, 09 Apr 2004 12:04:29 -0600
This is a MIME message. If you are reading this text, you may want to 
consider changing to a mail reader or gateway that understands how to 
properly handle MIME multipart messages.

--=__Part06273DBD.1__=
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit

Just an addendum/FYI encountered for those using v4.5 (may be very
unique to our test and NetBackup environment)
 
Master and Media Servers: W2K
DataCenter 4.5 FP5
 
In our most recent DR test using v4.5 we encountered issues when
attempting to configure (remove or add devices) needed in our DR test
scenario.
 
Anyone who has gone through a recovery drill knows that devices not
needed (not present or available) that are configured and recreated
after running "bprevovery" can create all types of issues ( wmd(70)
errors, GUI devices timeouts etc...) with restore attempts that will
fail.
 
Even after running "vmglob -delete", deleting storage units and
deleting unnecessary devices from vm.conf, we still encountered GUI
timeouts finally resulting in
vmd (70) errors.  What was unusual was even after deleting the vm.conf
file, it was getting recreated every time the Volume Manager was started
or stopped.
 
Finial, the Veritas tech on the line who was working with us found what
we were looking for on the top of page 301 in the Veritas DataCenter 4.5
Media Manager for Windows System Administrators guide. Now, in v4.5, the
vm.conf is automatically updated based on robot host names appearing in
the volume groups.
 
In order to prevent the creation of (repopulation of vm.conf) you need
to move all assigned media to NONE a.k.a.: standalone (move the media
from DLT, or other type, to standalone, but not via the GUI. From the
command line
 
vmquery -a -j > media.lst
(Then script out those media matching "DLT" (or other density)
(Additional modify the media.lst file to say: vmchange -new_rt none -m
<media_id>
and run the script.
 
After we did this, GUI device management and data restoration for our
DR testing went fine.
 
Scott
 
>>> "Marx, Keath" <Keath.Marx AT anthem DOT com> 4/9/2004 9:42:15 AM >>>

Never mind.  I pulled my head out and remembered the VMCHANGE command. 
Thanks anyway._____________________________________________
From: veritas-bu-admin AT mailman.eng.auburn DOT edu@AICI On Behalf Of "Marx,
Keath" <Keath.Marx AT anthem DOT com>
Sent: Friday, April 09, 2004 10:59 AM
To: Veritas NBU Mailing List (E-mail)
Subject: [Veritas-bu] media ID not unique after bprecover
Importance: High
I searched the archives and support but didn't find a resolution for
this.We are performing a DR test.  At our DR test site we have recovered
the master/media, recovered the databases and reconfigured the hardware.
All seems well enough.  I can move tapes via robtest.  When I do an
inventory it finds the tapes in slot 1.  I reconfigured the barcode
rules to read the barcode properly.  It sees the barcode as NR1072L1. 
My production robot can only read the first 6 digits so it sees the
barcode as NR1072.  What do I need to do?  I have the DR coordinator
breathing down my neck.
CONFIDENTIALITY NOTICE: This e-mail message, including any attachments,
is for the sole use of the intended recipient(s) and may contain
confidential and privileged information or otherwise be protected by
law. Any unauthorized review, use, disclosure or distribution is
prohibited. If you are not the intended recipient, please contact the
sender by reply e-mail and destroy all copies of the original message. 

--=__Part06273DBD.1__=
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<HTML><HEAD><TITLE>RE: [Veritas-bu] media ID not unique after bprecover</TI=
TLE>
<META http-equiv=3DContent-Type content=3D"text/html; charset=3Diso-8859-1"=
>
<META content=3D"MSHTML 6.00.2800.1400" name=3DGENERATOR></HEAD>
<BODY style=3D"MARGIN: 4px 4px 1px; FONT: 10pt Tahoma">
<DIV>Just an addendum/FYI encountered&nbsp;for those using&nbsp;v4.5 (may =
be very unique to our test and NetBackup environment)</DIV>
<DIV>&nbsp;</DIV>
<DIV>Master and Media Servers: W2K</DIV>
<DIV>DataCenter 4.5 FP5</DIV>
<DIV>&nbsp;</DIV>
<DIV>In our most recent DR&nbsp;test using v4.5&nbsp;we encountered issues =
when attempting to configure (remove or add&nbsp;devices) needed in =
our&nbsp;DR test&nbsp;scenario.</DIV>
<DIV>&nbsp;</DIV>
<DIV>Anyone who has gone through a recovery drill knows that devices not =
needed (not present or available) that are configured and recreated after =
running&nbsp;"bprevovery" can create all types of issues ( wmd(70) errors, =
GUI devices timeouts etc...)&nbsp;with restore attempts that&nbsp;will =
fail.</DIV>
<DIV>&nbsp;</DIV>
<DIV>Even after running "vmglob -delete",&nbsp;deleting storage units =
and&nbsp;deleting unnecessary devices from&nbsp;vm.conf, we still =
encountered GUI timeouts finally resulting in</DIV>
<DIV>vmd (70) errors.&nbsp; What was unusual was even after deleting&nbsp;t=
he vm.conf file, it&nbsp;was getting recreated every time the Volume =
Manager was started or stopped.</DIV>
<DIV>&nbsp;</DIV>
<DIV>Finial, the Veritas tech on the line who was working with us found =
what we were looking for on the&nbsp;top&nbsp;of page 301 in the Veritas =
DataCenter 4.5 Media Manager for Windows System Administrators guide. Now, =
in v4.5, the vm.conf is automatically updated based on robot host names =
appearing in the volume groups.</DIV>
<DIV>&nbsp;</DIV>
<DIV>In order to prevent the creation of (repopulation of vm.conf) you =
need to move all assigned media&nbsp;to NONE&nbsp;a.k.a.: standalone (move =
the media from DLT, or other type, to standalone, but not via the GUI. =
>From the command line</DIV>
<DIV>&nbsp;</DIV>
<DIV>vmquery -a -j &gt; media.lst</DIV>
<DIV>(Then script out those media matching "DLT" (or other density)</DIV>
<DIV>(Additional modify the media.lst file to say: vmchange -new_rt none =
-m &lt;media_id&gt;</DIV>
<DIV>and run the script.</DIV>
<DIV>&nbsp;</DIV>
<DIV>After we did this, GUI device management and data restoration for our =
DR testing went fine.</DIV>
<DIV>&nbsp;</DIV>
<DIV>Scott</DIV>
<DIV>&nbsp;</DIV>
<DIV>&gt;&gt;&gt; "Marx, Keath" &lt;Keath.Marx AT anthem DOT com&gt; 4/9/2004 =
9:42:15 AM &gt;&gt;&gt;<BR></DIV>
<DIV style=3D"COLOR: #000000">
<P align=3Dleft><FONT face=3DArial color=3D#000080 size=3D2>Never =
mind.&nbsp; I pulled my head out and remembered the VMCHANGE command.&nbsp;=
 Thanks anyway.</FONT></P>
<P align=3Dleft><FONT face=3DTahoma size=3D2>______________________________=
_______________<BR></FONT><B></B><B><FONT face=3DTahoma size=3D2>From:</FON=
T></B><FONT face=3DTahoma size=3D2> veritas-bu-admin AT mailman.eng.auburn DOT 
edu=
@AICI</FONT><B></B><B> <FONT face=3DTahoma size=3D2>On Behalf Of</FONT></B>=
 <FONT face=3DTahoma size=3D2>"Marx, Keath" &lt;Keath.Marx AT anthem DOT 
com&gt;<B=
R></FONT><B></B><B><FONT face=3DTahoma size=3D2>Sent:</FONT></B><FONT =
face=3DTahoma size=3D2> Friday, April 09, 2004 10:59 AM<BR></FONT><B></B><B=
><FONT face=3DTahoma size=3D2>To:</FONT></B><FONT face=3DTahoma size=3D2> =
Veritas NBU Mailing List (E-mail)<BR></FONT><B></B><B><FONT face=3DTahoma =
size=3D2>Subject:</FONT></B><FONT face=3DTahoma size=3D2> [Veritas-bu] =
media ID not unique after bprecover<BR></FONT><B></B><B><FONT face=3DTahoma=
 size=3D2>Importance:</FONT></B><FONT face=3DTahoma size=3D2> High</FONT></=
P><BR>
<P align=3Dleft><FONT face=3D"Tms Rmn" color=3D#000000 size=3D2>I searched =
the archives and support but didn't find a resolution for this.</FONT></P>
<P align=3Dleft><FONT face=3D"Tms Rmn" color=3D#000000 size=3D2>We are =
performing a DR test.&nbsp; At our DR test site we have recovered the =
master/media, recovered the databases and reconfigured the hardware. =
All</FONT><FONT face=3D"Tms Rmn" color=3D#000000></FONT> <FONT face=3D"Tms =
Rmn" color=3D#000000 size=3D2>seems well enough.&nbsp; I can move tapes =
via robtest.&nbsp; When I do an inventory it finds the tapes in slot =
1.&nbsp; I reconfigured the barcode rules to read the barcode properly.&nbs=
p; It sees the barcode as NR1072L1.&nbsp; My production robot can only =
read the first 6 digits so it sees the barcode as NR1072.&nbsp;</FONT><FONT=
 face=3D"Tms Rmn" color=3D#000000></FONT> <FONT face=3D"Tms Rmn" color=3D#0=
00000 size=3D2>What do I need to do?&nbsp; I have the DR coordinator =
breathing down my neck.</FONT></P><BR>CONFIDENTIALITY NOTICE: This e-mail =
message, including any attachments, is for the sole use of the intended =
recipient(s) and may contain confidential and privileged information or =
otherwise be protected by law. Any unauthorized review, use, disclosure or =
distribution is prohibited. If you are not the intended recipient, please =
contact the sender by reply e-mail and destroy all copies of the original =
message. </DIV></BODY></HTML>

--=__Part06273DBD.1__=--

<Prev in Thread] Current Thread [Next in Thread>
  • [Veritas-bu] media ID not unique after bprecover - 3.4 to 4.5 DR recovery caveat, Scott Jacobson <=