On Sun, 8 Apr 2007, Marcus Watts wrote: > I envy your ability to install stuff like this on a production machine > without doing any previous testing.
Yes. That was my bad. I had been afs running on Linux for some time, and thought it long stable on solaris, and that no testing was needed; egg/face thing. I ran afs (from transarc) on HP/UX, and didn't recall any unexpected reboots. OSF/TOG might have source from AFS 3.2 somewhere, I think. If I can find it, I might check to see if it did that back when.... > In our production environment, our production staff won't make this > kind of upgrade unless it's been thoroughly tested and the phase of > the moon is just right. I'm usually like that, too. Current exception notwithstanding. > Did you build from scratch or install a binary package? Scratch. > What documentation did you read before installing it? Did it warn you > that it would reboot? Foolishly thinking that I'm pretty familiar with afs, and can install the client from memory, and installing a client in an existing cell, I didn't refer to the quickstart manual. > Is there any improvement that can be made to the documentation that > would have avoided this problem for you? No. I don't think so, not to the documentation. My fault was not reading the docs on a wrong assumption. Changing the docs will not fix that. > the "quickstart unix" guide at least appears to warn you that it will > reboot -- in auqbg004.htm, around line 154 of 281 (generic?), in > auqbg005.html around linek 1721 of 4427 (solaris specific.), and more > places too. > -- Av8 Internet Prepared to pay a premium for better service? www.av8.net faster, more reliable, better service 617 344 9000 _______________________________________________ OpenAFS-devel mailing list [email protected] https://lists.openafs.org/mailman/listinfo/openafs-devel
