home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!mcsun!uknet!mucs!mcchpc.mcc.ac.uk!zlsiimw
- From: zlsiimw@mcchpc.mcc.ac.uk (Mark Whidby)
- Newsgroups: alt.gopher
- Subject: Re: .cache
- Message-ID: <5733@m1.cs.man.ac.uk>
- Date: 13 Aug 92 08:15:51 GMT
- References: <1992Aug12.195021.8790@trentu.ca>
- Sender: news@cs.man.ac.uk
- Reply-To: M.Whidby@mcc.ac.uk
- Organization: Manchester Computing Centre
- Lines: 23
-
- In article <1992Aug12.195021.8790@trentu.ca>, ccksb@blaze.trentu.ca (Ken Brown)
- writes:
- |> If a .cache file is not up to date (ie: relatively new as the gopherd docs
- |> say) then not all changes in the gopher data dir are reflected. It appears
- |> that starting gopherd with -C will turn this off.
- |>
- |> What are the relative advantages and disadvantages of .cache files? If of
- |> advantage, how can I assure they are all up-to-date?
-
- Main advantage is to save on server processing time - if a .cache file
- exists it just squirts that back at the client without trundling through
- the directory getting all the info it needs to send back.
-
- Disadvantage - as you noted, if you change something in the tree it doesn't
- appear immediately as viewed by the client until the cache file expires
- (10 minutes by default I think). I run with caching on but if I want to
- check something I've just inserted I do an 'rm .cache' in the appropriate
- directory. Perhaps soemone should write a script to remove all .cache files
- in a directory. (I miss the tree-walking capabilities that Primos had in
- its wildcarding...)
- --
- _____________________________________________________________
- Mark Whidby, Distributed Systems, Manchester Computing Centre
-