Re: [Monotone-devel] Re: Sanity checking

[Jon Bright]

Bruce Stephens wrote:
> > If a no-sanity-checking option isn't viable (or is too hated), maybe
> > we could at least look at a seriously-reduced sanity checking
> > option?  Do we really need to go three revs back?
>
>
> Or is there some repeated sanity checking which wouldn't be too hard
> to remove?  (A timestamp of the last time a paranoid monotone
> successfully ran on a repository, or something?)  That wouldn't help
> the initial experience of pulling from a big repository (although I
> guess you could have the sending repository give an assurance of
> sanity, if you could believe it).
>
> Hmm.  Sounds hard.

I know Nathaniel's opposed to this kind of thing (and I'm broadly 
agreed), on the grounds that it makes the sanity-checking more complex - 
and that's a place we really don't need complexity.  Slow sanity 
checking is one thing.  Inadvertently broken sanity-checking... :-)

Hence my wish for a simple "turn it all off" switch.  Is there, 
incidentally, anything that the sanity checking gives me which "db 
check" doesn't?

--
Jon Bright
Silicon Circus Ltd.
http://www.siliconcircus.com