| 1 |
Hi, |
| 2 |
|
| 3 |
On Fri, 13 Apr 2007 18:45:04 +0100 |
| 4 |
Neil Bothwick <neil@××××××××××.uk> wrote: |
| 5 |
|
| 6 |
> On Fri, 13 Apr 2007 10:46:42 -0500, Fabio wrote: |
| 7 |
> |
| 8 |
> > Well, when I started with Gentoo Linux almost a year ago, I emerged |
| 9 |
> > --sync more than twice per week. I never experimented any damage or |
| 10 |
> > error. |
| 11 |
> |
| 12 |
> I have a system here that rsyncs with three other computers every hour |
| 13 |
> and the disk is still good after several years, so the traffic |
| 14 |
> from a paltry portage sync should do nothing but give the disk a |
| 15 |
> little healthy exercise. |
| 16 |
|
| 17 |
I like this theory. I think bad blocks on HD's (and I guess noone here |
| 18 |
is talking about flash disks or writable DVD media) occur almost |
| 19 |
independent of usage. So if bad blocks occur, there might be a big |
| 20 |
chance that it happens in a portage sync simply because there's a lot |
| 21 |
of file writing/deleting and thus there's a bigger chance that it |
| 22 |
happens in that moments. If that was true, it is likely that the errors |
| 23 |
hit portage's files with a probability that corresponds to the |
| 24 |
percentage of sync (and due to the "test": emerge) I/O vs. general IO. |
| 25 |
On a gentoo system, there's probably a lot of disk I/O simply because |
| 26 |
of portage. This might explain why there's that feeling that a sync |
| 27 |
might hurt. |
| 28 |
|
| 29 |
And, the good side of things: If this theory holds valid, the errors |
| 30 |
are likely to hit portage -- not all that bad, a resync and |
| 31 |
everything's fine again :-) The harddisk will cure the problem by |
| 32 |
allocating spare sectors (as long as available). |
| 33 |
|
| 34 |
-hwh |
| 35 |
-- |
| 36 |
gentoo-user@g.o mailing list |
Archives