[Search for users]
[Overall Top Noters]
[List of all Conferences]
[Download this site]
Title: | ase |
|
Moderator: | SMURF::GROSSO |
|
Created: | Thu Jul 29 1993 |
Last Modified: | Fri Jun 06 1997 |
Last Successful Update: | Fri Jun 06 1997 |
Number of topics: | 2114 |
Total number of notes: | 7347 |
I have been trying to figure out what exactly ASEROUTING is supposed to
do and not do and between the very cryptic and terse description in the
ASE and TruCluster manuals and the notes in this conference I can't help
the feeling that I am missing something obvious.
In the situations that I am considering the use of the feature I have something
like the following configuration:
nodeA............... nodeB...............
. . . .
. . . .
. . . .
..*........*... ...*.......*...
nodeA1 | : nodeA2 nodeB1 | : nodeB2
| : | :
..........|........:...............................|.......:................
| : |
__________|_________________________________:______|___________________
| :
nodeC1 | : nodeC2
nodeC ..*........*...
. .
. .
. .
...............
where nodeA and nodeB are offering a disk service and nodeC accesses the disk service
via an alias 'disks' which lives on the same network as nodeA1 and nodeB1 (ie it will
reside either on the interface nodeA1 or nodeB1).
The problem I am attempting to solve is that if there is a break in the network so
nodeC1 cannot reach 'disks' over the net (nodeC1-nodeA1), then it seems that it should
be possible through some kind of routing magic to allow nodeC's traffic to reach the
"service" via the other network (nodeC2-nodeA2).
From reading the manuals I cannot tell if ASEROUTING addresses this situation or not
and if it does what one has to do on the NodeC system to make it work. I figured that I
could give it a try and see if the problem gets solved but I don't seem to be able to
figure it out.
It seems that all you have to do on the NodeA system is to set ASEROUTING to "YES" in
rc.config and startup a gated daemon. I suspect that there is much more required -
create some kind of /etc/routes on ALL of the systems? Also is there some requirement
that the network (nodeC1-nodeA1) be monitored? or NOT monitored?
Also it is not clear whether traffic will end up going from NodeC to NodeA over BOTH
networks concurrently ... this would be bad. But I just can't tell from the terrible
description.
Can anyone point me in the right direction(s)?
T.R | Title | User | Personal Name | Date | Lines
|
---|