Coda File System

Re: patches/merging

From: Jan Harkes <jaharkes_at_cs.cmu.edu>
Date: Thu, 10 Jul 2014 15:56:24 -0400
On Thu, Jul 10, 2014 at 12:32:10PM -0400, Greg Troxel wrote:
> 
> u-codalist-rcma_at_aetey.se writes:
> 
> > We are doing Coda development at Aetey and got some patches
> > but first of all:
> >
> > I wonder whether it is possible to merge the very old and heavily used
> > for a long time changes implementing the so called modular clog?
> 
> Is anybody still working on coda at CMU?   I can certainly understand
> being busy on other things because there is no current funding.   If
> there isn't a continuing maintenance effort, then it seems time to
> transform coda into a community project rather than a only-cmu-can-write
> project.   Barring objections, it seems reasonable for Rune or someone
> to set up an official repository someplace, perhasp under the aegis of
> FSF, Software Freedom Conservatory, or some other group.

We still use Coda ourselves, and have used it for the past couple of
years as a base for projects for an Engineering Distributed Systems
class.

Maintenance patches have been merged into both CVS and Git repositories
whenever needed, but no major releases have been pushed out. This is
partly because of lack of time and partly because the current source
repository situation,

The git repository is the one where I do my main development, however
CVS is still considered the canonical source. But CVS doesn't handle
branches and merges quite as well so whenever I try to propagate changes
a lot of time is spend rebasing the Git history into a linear set of
commits that can then be pushed into the CVS history.  Occasionally this
was even messier in case some change got committed straight to CVS and I
wouldn't catch it until too late.

Switching to Git as the canonical repository would simplify life
considerably, but the existing Git repo may not be the right one to
use for this. For one it doesn't accurately track the full history. I
tried to import as much history as I could from CVS, but especially the
very early history is not correct. For the longest time we actually have
a negative LoC count on Ohloh but it looks like they managed to fix that.

Oh and I wanted recently to push at least a source level release out of
the door because of a fix for a filedescriptor leak that over time kills
our servers when backups are failing, but all the various shell scripts
I used to use to tag, build, sign and package new releases have broken
and I still have to decide if I want to glue the pieces back together or
just rewrite them.

About the modular clog code, I don't know what the current state of it
is. I haven't looked at it since some very early versions, of which I
have a vague recollection that I thought were a bit complex and hacky.
I recall that it really needed some better/more portable representation
of a Coda token that could get passed from module to module, not sure if
that ever got implemented I think the existing exported token had an
endianess issue.  But again I have not looked at modular clog for at
least 8 years if not more and I haven't seen any patches for them.

Jan
Received on 2014-07-10 15:56:34