Networker

Re: [Networker] Need to confirm steps to upgrade to 6.1.4

2004-02-20 09:15:34
Subject: Re: [Networker] Need to confirm steps to upgrade to 6.1.4
From: "robert.maiello AT thomson DOT com" <robert.maiello AT THOMSON DOT COM>
To: NETWORKER AT LISTMAIL.TEMPLE DOT EDU
Date: Fri, 20 Feb 2004 09:15:27 -0500
I always do a nsrck -L2 and nsrck -L3  and an nsrim -X  before an upgrade
and saving off the /nsr dir.
May be a waste of time for a minor upgrade.


Robert Maiello
Thomson Healthcare


On Thu, 19 Feb 2004 15:01:57 -0500, George Sinclair
<George.Sinclair AT NOAA DOT GOV> wrote:

>Hi,
>
>We're planning to upgrade our current server (Solaris 2.8) and storage
>node (redhat 7.3) from 6.1.1 to 6.1.4, and we wanted to confirm the
>steps listed below for the install. We could go to 7.1.1 but felt more
>comfortable sticking with the last 6.x "ironed out" version. Might
>consider 7.x a little later. Anyway, I want to make sure that I'm not
>leaving something out here. I checked Legato's web page, but it appears
>that the most recent installation guide for Solaris appears under 6.1.1,
>and the front page of the PDF identifies it as 6.1, Dec 10 2001 and Aug
>31, 2001 for Linux. Guess the steps haven't changed but had a few
>questions as mentioned in steps below. As near as I can tell, everything
>is installed to default location of /usr.
>
><<  STEPS FOR UPGRADE >>
>1. Backup all client indexes and server's bootstrap (savegrp -O group)
>2. Record latest bootstrap saveset id and volume label (mminfo -B)
>3. On primary server, tar off a copy of /nsr/res just for save keeping.
>4. On primary server, make a copy of files: /etc/rpc, /etc/syslog.conf
>
>QUESTION: On the primary server /nsr -> /disk2/nsr and most client
>indexes reside under /disk2/nsr/index. No problems there. However, we do
>have several large client indexes located under /disk3/nsr_index and the
>given client specifies this under the "Index path" field for the client.
>Not sure if this will create any problems?
>
>5. Unmount all loaded/mounted tapes.
>
>QUESTION: Should we remove the jukebox entries and devices? If so, I
>guess we'd do that next before step 6.
>
>6. Shut down the NetWorker software on storage node and primary server
>and ensure no services are running (ps -ef | grep nsr).
>7. Remove all NetWorker packages on primary server in order: LGTOserv,
>LGTOnode, LGTOdrvr, LGTOclnt, LGTOman
>
>QUESTION: Do we want to remove the license manager package (LGTOlicm)?
>If so, what would the order be above?
>
>8. Repeat removal of software on storage node in order: lgtonode,
>lgtodrvr, lgtoclnt, lgtoman
>9. Install software on primary server in order: LGTOclnt, LGTOdrvr,
>LGTOnode, LGTOserv, LGTOman, LGTOlicm. Start daemons and verify (ps -ef
>| grep nsr).
>11. Repeat install of software on storage node server in order:
>lgtoclnt, lgtodrvr, lgtonode, lgtoman. Start daemons and verify (ps -ef
>| grep nsr).
>12. If the jukeboxes and devices were removed prior then run jbconfig on
>primary server to re-create jukebox and devices and then repeat process
>on storage node server.
>13. Test backing up some data and recovering some data prior to the
>upgrade.
>
>Not sure if we should do anything further.
>
>Thanks for any suggestions or help on this.
>
>George
>
>--
>Note: To sign off this list, send a "signoff networker" command via email
>to listserv AT listmail.temple DOT edu or visit the list's Web site at
>http://listmail.temple.edu/archives/networker.html where you can
>also view and post messages to the list.
>=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=

--
Note: To sign off this list, send a "signoff networker" command via email
to listserv AT listmail.temple DOT edu or visit the list's Web site at
http://listmail.temple.edu/archives/networker.html where you can
also view and post messages to the list.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=