[f-nsp] Getting started with ServerIron 4G
John G.
isaac737 at gmail.com
Mon Feb 4 14:17:21 EST 2008
.to email address. nice! anyhow, if u log into your secondary/standby and
u have no real servers over there your failover will never work.
once u have the config-sender set up all u do to synch the config is:
conf t:
config synch slb
then u will see some stuff spit out to the console saying the synch was
complete. SLB stands for Server Load Balance. there are different options
but that is the one u want.
here is a config snippet to get u started on the primary:
<SNIP>
config-sync sender ethernet 2/1 mac 0012.2222.1111 vlan-id 123
where ethernet 2/1 is the LAN port in our config. VLAN is the VLAN the LAN
ports are in. and the MAC address is the hardware address of your secondary
foundry. this can be found via sh interface ethernet 2/1 (use the one that
starts with bia).
----------
secondary config:
config-sync receiver ethernet 2/1 vlan-id 123
----------
MAC addresses and VLANs changed to protect the innocent.
cheers,
jg
p.s. vote Ron Paul in 2008!
On Feb 4, 2008 9:49 AM, Chris Price <cprice at its.to> wrote:
>
> Yes, am using hot-standby HA, not SYM-active. I would agree that this is
> a config-sync problem that is resulting in real/virtual servers not
> being available on the standby nodes when failover occurs (ie: active
> dies, and standby becomes active node).
>
> I believe I have the latest SIcli.pdf command line reference guide from
> Foundry and it doesnt have any config-sync references. Nor does the SLB
> Guide from Foundry. I'm going to call Foundry today and open a support
> ticket.
>
> From what I can gather of the 'config-sync' options is that you need to
> config one unit as a config 'sender' and one as a config 'receiver'.
> While this will likely work fine, its counterintuitive to what I am used
> to - the active node should be the config 'sender' and standby systems
> in a hot-standby HA setup should be receivers.
>
> Chris
>
> Oliver Adam wrote:
> > But this is the so called "SYM-active" mode and therefore not at all
> > related to hot-standby which seems to be what Chris is using at the
> > moment.
> >
> > @Chris,
> >
> > I guess this is a config problem. The last paragraph of your email
> > below sounds like a conf sync problem now and not like a failover
> > problem anymore. Search for "config-sync" in the latest documentation
> > and you are going to find the stuff to synchronize the configuration.
> > Attach the configs or cut&paste them to the mail in case you are
> > talking about a failover problem at the network level (L2 or L3).
> >
> > R, Oliver
> _______________________________________________
> foundry-nsp mailing list
> foundry-nsp at puck.nether.net
> http://puck.nether.net/mailman/listinfo/foundry-nsp
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://puck.nether.net/pipermail/foundry-nsp/attachments/20080204/c1f1e4eb/attachment.html>
More information about the foundry-nsp
mailing list