I am having a hard time believing that this just happened!
Who writes code that can happily go and delete dozens of folders
and thousands of files and not a single mention in the documentation
of this possibility?
Cleverly, because the deployment receiver wants a destination folder
that's one level up from ROOT, it got not only my web site (everything
under ROOT), but also everything that was one level up. This happened
to be a user directory, so it pretty much cleaned everything out!
I'm trying to figure out why this did NOT happen in my development
environment. I'm using exactly the same version, environment, etc.
The only difference is the path difference between the dev box and the
staging box.
The only thing I can think of that was different: On my dev box, I
first tested deployment with the deployment receiver pointed at a test
directory (thankfully). Is it possible that it established its "snapshot"
(what's there and what's not) during that initial testing? Then, later, when
I pointed it to the actual site, it never attempted to sync everything.
Readers, please understand. We have a large, complex site, and we're
using alfresco to manage just a small portion of it (at this point). And
somehow, the deployment receiver decided to DELETE everything in
the web root and the folder above it that it did not know about. Since
it was deploying only 2 files in 2 folders, it DELETED thousands of other
files.
I very much need to know how and why and how to prevent this in
the future.
Fred
