It would be nice to have a proper safety margin of redundant machines
- say, one hot spare for every core machine, and one for every two
apaches/squids - before deciding that "money for hardware really isn't
the issue". I had the impression that some of the recent delays and
outages would have been less troublesome (and less noticeable), had
there been more redundancy of key components.
The $50k in available funds as of Dec 31 was a bit less than two
months' operating costs; not as much as it might seem.
--
+sj+