On Mon, 30 Aug 2010 15:34:36 -0400 (EDT) [email protected] wrote:
> [uss] > > What's the word on this tool? uss is old and kas-encumbered and I don't think it gets much attention anymore. When discussing the list of features to keep and drop for 2.0, uss was considered to be removed from the project entirely, until it was brought up that someone actually uses it without using it with kaserver. So it is used, but it's susceptible to bitrot, and so I'm not surprised that it's not packaged. Also, I presume you saw references to uss in the admin guide? The admin guide contains a lot of old information, I believe. > Also, I remember using a tool when I was a grad student at CMU that > would show you a little running transcript of your AFS fetches, plus > server disconnection/reconnection notices. Was that mariner? (It's > been a while -- I got my degree in 1991.) There's no such tool that I > can find in the client package, and the syslog doesn't show this kind > of thing. Am I missing something? > > Or am I confusing the fetch/store log entries with coda's codacon? > It's kind of nice to see this stuff, because it tells me when I've > triggered some storm of fetches, which I should either kill or just > compose my soul in patience. The code to emit mariner notices still exists, but we don't have the traditional tool to capture them and log them usefully (see the manpage for 'fs monitor/mariner' <http://docs.openafs.org/Reference/1/fs_monitor.html>). There is something to process them on Mac OS X specifically, but that's it as far as I know. You may want to look at 'fstrace'. It's a lot less useful than what it sounds like you want, and you kinda have to know how to interpret the output, but if you're trying to debug client operations, it is useful. A quick how-to on how to get it to produce output was written by Simon: <http://blob.inf.ed.ac.uk/sxw/2009/01/24/using-fstrace-to-debug-the-afs-cache-manager/>. Also, server disconnection/reconnection notices should be logged in syslog without you needing to do anything. -- Andrew Deason [email protected] _______________________________________________ OpenAFS-info mailing list [email protected] https://lists.openafs.org/mailman/listinfo/openafs-info
