Mark ,
It is basically a high availability feature in the event of a VIP crash or OIR.
On a 75xx when a single LC crashes all the interfaces are toggled because
of MEMD re-carve & CyBUS complex causing microcode on all LCs to be
reloaded.LC Recovery time is 150-155 seconds for all LCs in a chassis to
recover due to a single LC failure - during this time, no LCs will forward
any packets. If your IGP adjacencies on other VIPs are torn down during
this time then it could take longer for them to come up & for traffic to pass.
With this Feature the microcode will be reloaded only on the LC that
failed.single LC Reload time should be 25 - 30 seconds for the single
(failed) LC to recover during which microcode on no other LCs will be
reloaded. Non-failed LCs in the chassis will not forward packets for
approximately 5 seconds during this recovery, so this is a huge improvement
in terms of availability.
Thanks
Yogi
At 06:13 AM 5/16/2001 -0400, Mark Ivens wrote:
>Hello,
>
>What are people's experiences of the 7500 single line card reload
>feature introduced in 12.0(13)S? I.e. does it work and actually
>prevent cbus complexes (and the associated rise in blood pressure of
>the person doing the OIR) and are there any gotchas associated with
>it? Are people using it on production boxen?
>
>--
>----------------------------------------------------------------------
> Mark Ivens mivens@clara.net "Never send a human to do a
> Systems Administrator machine's job." - Agent
> ClaraNET Ltd Smith, "The Matrix"
>----------------------------------------------------------------------
This archive was generated by hypermail 2b29 : Sun Aug 04 2002 - 04:12:37 EDT