Update: this morning, it seems to be working, albeit with about a 1-2 minute window of inaccessibility while sshd is brought back online.
Now, my ssh attempts simply block until sshd is accepting connections. Yesterday, the attempts failed quickly with a consistent error. Unfortunately, I killed that terminal window. Oh, well. I'll chalk it up to gremlins and/or impatience. On Wed, Feb 8, 2012 at 1:51 AM, Andrei Savu <[email protected]> wrote: > Bootstrap Cluster -> Login over SSH -> Reboot -> Login over SSH > > This works as expected for me (quite fast also). > > > On Wed, Feb 8, 2012 at 8:56 AM, Andrei Savu <[email protected]> wrote: > >> >> On Wed, Feb 8, 2012 at 5:53 AM, Evan Pollan <[email protected]>wrote: >> >>> Is this a known limitation? Does anyone know of a customization to the >>> installation/configuration function scripts that will ensure that the ssh >>> config persists through a reboot? >> >> >> It should be persisted but the ec2-user is only used for bootstrap. Whirr >> 0.7.0 creates a new user on the remote machines named whirr.cluster-user >> (default value is the name of local user running Whirr). >> >> Try: ssh -i ... <whirr.cluster-user>@remote-machine or just ssh -i >> remote-machine if you don't specify a value for whirr.cluster-user. >> >> Note: if you specify a value for whirr.cluster-user make sure it doesn't >> already exist on the remote machines >> >> PS: I'm trying to replicate the behaviour you are seeing. >> > >
