Home
| Calendar
| Mail Lists
| List Archives
| Desktop SIG
| Hardware Hacking SIG
Wiki | Flickr | PicasaWeb | Video | Maps & Directions | Installfests | Keysignings Linux Cafe | Meeting Notes | Linux Links | Bling | About BLU |
I exchanged some email with some people in file system engineering. I have a copy of a reproducer, but a bug report needs to be filed from Mt. Holyoke. I'll see if I can reproduce it on a system in our lab. If I can, I'll file an internal report. Ron Peterson wrote: > Just an update for anyone who might be interested in why a file (e.g. > /etc/password) might appear to fade in and out of existance. (See > previous posting for more info.) > > We haven't got it completely licked yet, but it appears to be related to > the fact that this is a dual processor machine (Compaq DS20). Using the > Tru64 UNIX 'runon' command to force execution of, say, 'ls' to occur on > a specific processor has shown that when the problem occurs, it is local > to one CPU (not always the same one). > > So we're suspecting inconsistent cache data. Whether it's a hardware or > software issue has yet to be determined. Certain operations will make > the problem go away, others will not. E.G. 'cp passwd temp; cp temp > passwd' doesn't help. 'cp passwd temp; mv temp passwd' (new inode) > does. If you do 'touch passwd' on the CPU that doesn't see it, it > reappears, but with zero size. > > Fun. > > -- > > Ron Peterson > 9 Morgan Street, Apt. 'A' > South Hadley, MA 01075 > 413.536.1146 tel > ron.peterson at yellowbank.com > - > Subcription/unsubscription/info requests: send e-mail with > "subscribe", "unsubscribe", or "info" on the first line of the > message body to discuss-request at blu.org (Subject line is ignored). > -- Jerry Feldman <gaf at blu.org> Boston Linux and Unix user group http://www.blu.org - Subcription/unsubscription/info requests: send e-mail with "subscribe", "unsubscribe", or "info" on the first line of the message body to discuss-request at blu.org (Subject line is ignored).
BLU is a member of BostonUserGroups | |
We also thank MIT for the use of their facilities. |