Re: [Maxlinux-devel] Last call before the old package archive is wiped!

[Kelledin <kel...@sk...>]

On Sunday 19 January 2003 01:51 pm, Terry Churchill wrote:
> My pet monkey insists that on Jan 18 2003 at 05:43PM
>
> Kelledin <kel...@sk...> warbled:
> > It seems all FTP downloads from skarpsey.dyndns.org are
> > complete. If someone (Terry?) still needs the old packages,
> > it would be a good idea to say so before tomorrow
> > morning--because that's when it all gets replaced. ;)  I'm
> > currently burning the old stuff to CD for archiving.
>
> That's OK - The FTP download timed out a while ago ;-) I got
> most of the old packages, but I'll start afresh with the new
> ones.

Hmm...if we keep having problems like this, maybe we should=20
reverse this whole process--instead of having=20
ftp.inceptionos.org sync up to my server, have me sync up my=20
repository to ftp.inceptionos.org.  There's not a whole lot I=20
can do about improving service on my end--I only have 128K of=20
upload bandwidth, and the connection is shared with a few other=20
people.

The two most viable options I can think of are either to run=20
rsync on ftp.inceptionos.org and give me client login rights, or=20
just provide SSH/SFTP access and let me login/upload via that. =20
Since I'd prefer to rsync only via SSH anyways in this scenario=20
(and rsync demands a lot of CPU time from both client and=20
server), it's probably simplest for me to just manage the=20
repository via SSH.

If this sounds like the best option, I can e-mail you my public=20
SSH2 key.  If not...well, we can keep trying what we're doing. =20
Incremental syncing shouldn't be a problem for rsync, it's just=20
the first large initial transfer that kills us.

By the way, what exact rsync command(s) did you try?

--=20
Kelledin
"If a server crashes in a server farm and no one pings it, does=20
it still cost four figures to fix?"