In Netview for AIX V3 there are possibilities to use the backup
functions to move the management of certain areas between two
Netview.
1. Could this be done with a command or cron? eg if the staff at
location A leaves at 6PM could Netview at location B take over?
2. If so, is it possible to start/stop other functions? In this case
we have other programmes managing non-ip equipment. It may be
location A leaving ("logging off") a tty port and location B doing
a logon to the same equipment. Is there an exit in the code or are
there any other ways of getting this done?
ANSWER
Backup functions: I assume you are talking about management takeover.
In our testing, we found that we could initiate the
takeover of the A's resources by simulating a node-
down for A. That resulted in B managing the
resources it was configured-for in A. However, I have
been told that a node-down trap is not enough in
current builds. If it turns out a simulated trap is
not enough, you would have to take the interface on
the "down" machine down long enough to cause a "real"
down situation recognized by A. We will test this
as time allows.
Management takeover is ONLY "managing" the nodes. I
assume that is sufficient to meet your needs.
An exit is not necessary -- event configuration could be used to
drive whatever function you choose. I have not checked which
NetView for AIX event would be the easiest to use, but I would
guess that one or all of the Manage events would be the easiest.
All you would have to do is, in your driven routine, keep track of
the fact that you have been "driven" so that you don't drive your
user-code multipl times. You could, in turn, re-arm the driven code
when manager takeover un-manages particular nodes later on (as it
comes back online).
From development:
To provide a takeover at a specific time use cron to do a
event -e NDWN_EV -h remote_host_to_be_down
run this on the backup machine.
That will cause the Manager Down pop-up to come up and the backup
operator can do what he wants. Note that since the remote manager is
not really down, but that won't matter. They can have another cron job
to simulate a node up
event -e NUP_EV -h remote_host_coming_up
This will remind the backup operator to unmanage the remote containers.
I don't really understand what is being asked in the second question.
They could have a program watching the traps to drive their other
functions.
There is no seperate event or exit issued on manager takeover or return.
S e a r c h - k e y w o r d s:
BACKUP/RECOVERY
WWQA: ITEM: RTA000066273 ITEM: RTA000066273
Dated: 08/1995 Category: ITSCSAIXNV6
This HTML file was generated 99/06/24~12:43:24
Comments or suggestions?
Contact us