I replaced a management blade in our ServerIron 850
this afternoon and it ended up causing an outage.  The
ServerIron, which had survived a failed card without
incident, subsequently failed when we tried to replace
the card.

As soon as we realized there was an issue, we removed
the management card we had just added and reloaded the
unit, but services still did not return.

After some more troubleshooting, we discovered that
select lines of our configuration were missing.  We are
still not sure how or why, but we ended up bringing
services back by looking at a diff of the configuration
before the replacement and after the replacement.

The ServerIron 850 is a highly redundant enterprise
piece of hardware where everything is supposed to be
hot-swappable.  Hopefully Brocade/Foundry support can
help us find out what went wrong so future replacements
do not cause an outage.

We still have to insert this management card to regain
redundancy, but we will now try to do it Friday night
around 2 AM PST. (IE, Saturday Morning)  That way when
the lines of code get removed, we can re-insert them
without causing any more trouble than we have to.