On 27 Jan 2003, Steffen Neumann wrote:

> > > or do you 2) *expect* them to be inconsistent simply because one
> > > was down for some time ?

> I think you don't trust coda enough that it could handle this
> situation itself, but I am pretty sure it does.

Sure it does. I have (inadvertently) been there sometime, not noticing
that one of the servers was down :)

A pitfall is that you do not want to have a server down for a long time,
as the alive server's modification log can be overflowed (say when many
files get modified).

> please correct me if I get this wrong, I have no experience
> with multiple server installations. In case I got this right,
> it could go into some manual or docs ?)

> 	Server1				Server2
> 	online				online

[skipped]

I think your descripton is pretty good.

> So again, if one server is down,
> replication is automagic.

<paranoia on>

Just remember that it is "lazy" i.e. it may strike in cases when a client
accesses a file during a network outage or when the other server is down.

The same applies for making backup volume clones, you may ls -R before
creating them to ensure you have got the latest file generations.

<paranoia off>

Regards,
--
Ivan