For all intents and purposes, the disk limit is at the 100% mark (w/o the 10% overhead) because some of the osi_* routines run with the user credentials and will run into the 100% limit rather than the 111% limit. We have patched AFS 3.1b locally to do the appropriate switching of credentials to allow writes beyond the 100% limit (but not beyond the 111% limit of the disk), but I forget whether we have forwarded these patches to Transarc. -R
- cache problem Jamey Maze
- Re: cache problem mf%cc . ic . ac . uk
- Re: cache problem John Hascall
- Re: cache problem Lyle_Seaman
- Re: cache problem Mike_Kazar
- Richard Basch
