[1052] in SIPB-AFS-requests
Re: State of the Cell Address
daemon@ATHENA.MIT.EDU (Calvin Clark)
Sun Jul 25 22:16:41 1993
Date: Sun, 25 Jul 93 22:16:27 -0400
From: Calvin Clark <ckclark@mit.edu>
To: Derek Atkins <warlord@athena.mit.edu>, tlyu@athena.mit.edu
Cc: sipb-afsreq@athena.mit.edu
In-Reply-To: Your message of Sat, 24 Jul 93 23:03:55 -0400.
<9307250303.AA09441@hodge>
I think Tom was the one who decided to modify the CellServDB in that
fashion. I forgot if he had done this or simply suggested it. (Tom, I
didn't think you got to the stage where you would have actually changed
the CellServDB. Did you tell me about it, and I forgot, or what?)
When I was doing the restore, I did not think to check the CellServDB
files, because some of the restores were working fine, and syncvldbs
seemed to work. If things were consistently failing, I would have
checked things like that, but as it was, it seemed kind of sporadic.
However, there is something that I'm not clear about. When marc was
bringing up the cell for the first time, I noted when he did a `bos
listhosts' on both machines, they both returned both machines.
(Although there was at least one .mit.edu on the end of one of the host
names, and I didn't understand why only one of them had that.) Isn't
the value returned by `bos listhosts' and changed by `bos addhost' the
list of valid database servers in the cell? If so, why does the
CellServDB matter? Where is the information for a `bos addhost' stored?
It's not in the BosConfig file.
-Calvin
