I removed x86-alex from skunk1, but it's still showing up 

http://chromeos-skunk1.mtv.corp.google.com/afe/#tab_id=hosts


can I add x86-alex to skunk-1?

I remembered x86-alex is thrown away based on a comment in an email thread. They should be cleaned up from db.

> On Sep 18, 2017, at 11:14 AM, John D Hong <johndhong@google.com> wrote:
>
> Isn't x86-zgb and x86-mario dead (semper fi...)
> I suppose throw in x86-alex as well...
>
Yes, all three of those boards are dead; we should clean up
the database.  There's a small handful of DUTs in the database
that might still be in labs outside of Stierlin Ct; we should
notify the owners.

It's weird that 'atest host list' can't find the 3 boards showed on http://chromeos-skunk1.mtv.corp.google.com/afe/#tab_id=hosts

chromeos2-row9-rack10-host1
chromeos2-row9-rack10-host3
chromeos2-row9-rack10-host5

How do we clean up the boards? I assume "atest host delete" won't work as it can't find them?

So I'll continue migrating.

When do we expect skunk-1 to replace skunk1?

This morning skunk1 has the same connection issue like skunk-1 in Issue 769551. I manually set its rpc server to chromeos-server29.mtv to recover skunk1.

All the skunk* machines have the same issue recorded in crbug.com/769551. I will verify skunk-1 after the issue is resolved and then start the migration. 

Migration was done in crbug.com/779711