control: tags -1 + pending

Hi Alexander,

On Mon, Oct 17, 2016 at 11:13:08AM +0200, Alexander Thomas wrote:
> >> In our setup,
> > could you maybe expand a little bit on this, I'm curious (and having
> > users is motivating!
> We install our software by bundling its components in Debian packages.
> When we release a patch, we distribute an updated set of packages that
> are served from a local repository within the system. This allows to
> simply do a dist-upgrade on all of the VMs and servers to apply the
> patch. Of course we want to make sure such upgrades don't do anything
> unexpected, therefore packages must pass a piuparts test before they
> are released.
 
hehe, very nice. I assume this is esaturnus.com :)

> Because I don't know what was the use case for the submitter of that
> hard-linking patch, it is safest to preserve it, but make it optional
> with a new --hard-link switch that is only relevant together with
> --existing-chroot. This is also very easy to implement, it only
> requires one extra test on a boolean. I have attached a patch that
> does exactly this.

thanks for explaining and doing the change like this! Much appreciated.
As the whole bug+patch! :)

On Mon, Oct 17, 2016 at 11:33:46AM +0200, Alexander Thomas wrote:
> > Regarding your use case: why don't you use chroot tarballs? IMO, tar xfz
> > might be faster than cp -a.
> We benchmarked it and on our trusty old build server, cp -a was faster.

interesting.

> Here's a small update that mentions --hard-link in --existing-chroot.

thanks a lot, I've finally merged this into the develop branch. And sorry
for the delay & "btw" I do plan to do a release in the next two weeks or so.
This year. :)


-- 
cheers,
        Holger

Attachment: signature.asc
Description: Digital signature

Reply via email to