On 23 February 2018 at 08:58, Pete Templin <peteli...@templin.org> wrote:
> I would even go so far as to:
> load system/kickstart files
> isolate the box (shutdown all ports)
> power-cycle the box, let it boot into the new code
> perform EPLD updates on all cards
> run the ISSU command to ensure all of the little microcode thingies (PSUs,
> fans, etc.) are covered
> unisolate the box

Just consider that (at least on the N5k and N9k), you will loose the
binary configuration this way, and NXOS will revert to loading the
ASCII configuration, which means you will loose some of the
configuration (because ASCII parsing of the config in NXOS has always
been and will always be buggy). So diff the configuration and restore
what's missing manually before putting everything back into

The final cold start could be required when power sequencers are
upgraded, you can use the command "reload power-cycle" for this

Also do consider that this manual upgrade path is not supported by
TAC. So you need make a choice between safe+unsupported or
buggy+supported upgrade paths.

I'm planning an upgrade of critical infrastructure on a pair of N5Ks
myself and finding a procedure that is both supported by TAC but at
the same time safe for my production traffic turned out to be quite
difficult - although the (VPC) setup is very simple and completely
best-pratice. The fact that every TAC engineer has a different opinion
of what is supported doesn't help either.

cisco-nsp mailing list  cisco-nsp@puck.nether.net
archive at http://puck.nether.net/pipermail/cisco-nsp/

Reply via email to