| 1 |
On Wed, 2004-12-01 at 03:25, Gregorio Guidi wrote: |
| 2 |
> On Wednesday 01 December 2004 11:59, Brian Harring wrote: |
| 3 |
> > On Wed, 2004-12-01 at 01:37, Gregorio Guidi wrote: |
| 4 |
> > > On Tue, 30 Nov 2004 06:22:07 -0800, Brian Harring <ferringb@g.o> |
| 5 |
> > |
| 6 |
> > 1) don't modify $portdir/metadata/cache. Modifying the rsync'd cache |
| 7 |
> > directly results in more to rsync for the next sync. Going that route |
| 8 |
> > would result in a lot of dial up users out for blood. |
| 9 |
> > 2) the user may not be using the same cache backend as the distributed |
| 10 |
> > cache- the rsync'd cache is portage_db_flat, while the user may be using |
| 11 |
> > a custom sqlite backend. I suspect this will be come more common place |
| 12 |
> > whenever cvs becomes stabled- the cache backend is being refactored |
| 13 |
> > currently. |
| 14 |
> > 3) cleanse stale entries from the cache, so you don't end up w/ a couple |
| 15 |
> > thousand extra files sitting in your local cache. Stable portage |
| 16 |
> > accomplishes this by wiping *all* local cache entries, and transferring |
| 17 |
> > the *entire* rsync'd cache over. Cvs is smarter, transfers only what |
| 18 |
> > has changed, and removes the stale entries (this is why it's faster). |
| 19 |
> |
| 20 |
> I was interested exactly in points 1 and 3 (and I'm glad to hear there's even |
| 21 |
> a sqlite backend). |
| 22 |
Nothing official sqlite thus far, I just know a couple of people who are |
| 23 |
did their own DIY backend for it. The cache refactoring should include |
| 24 |
a sqlite backend, although it hasn't been coded yet. |
| 25 |
|
| 26 |
> With respect to point 3, the solution you imlpemented is surely the right |
| 27 |
> thing to do. |
| 28 |
> With respect to point 1, maybe a good solution could be to just check cache |
| 29 |
> validity in $portdir/metadata when the cache is needed, and write a new cache |
| 30 |
> in /var/cache/edb if it is not valid. |
| 31 |
Err... elaborate |
| 32 |
Assuming I'm following, you're proposing doing the updates to local |
| 33 |
cache, and reading from the metacache? If so, that's twice the level of |
| 34 |
potential stats/reads required, which won't speed it up much :) |
| 35 |
~brian |
| 36 |
|
| 37 |
|
| 38 |
|
| 39 |
-- |
| 40 |
gentoo-portage-dev@g.o mailing list |
Archives