| 1 |
On Thu, 10 Aug 2017 10:50:45 +1000 |
| 2 |
"Sam Jorna (wraeth)" <wraeth@g.o> wrote: |
| 3 |
|
| 4 |
> On 10/08/17 06:35, William L. Thomson Jr. wrote: |
| 5 |
> > FYI binpkgs have no hash. If someone did something malicious within |
| 6 |
> > the binhost to the binpkgs. You have no way of knowing. Yes the |
| 7 |
> > same can happen with ebuilds and manifest. But easy to sync portage |
| 8 |
> > and see if a manifest has changed. |
| 9 |
> |
| 10 |
> This isn't exactly true - see ${PKGDIR}/Packages on the binhost, which |
| 11 |
> is a manifest of built packages and related metadata. Granted this is |
| 12 |
> created by the binhost, it does exist and contains SHA1 and MD5 |
| 13 |
> hashes, as well as package size. In that sense it's no different to |
| 14 |
> how a package Manifest file works within a repository. |
| 15 |
|
| 16 |
You are correct. I meant to say no verifiable hash. You can hash |
| 17 |
anything locally and claim it to be trustworthy. Thus mentioning |
| 18 |
syncing portage to compare manifest of ebuild/SRC_URI. |
| 19 |
|
| 20 |
Someone remakes a binpkg tarball, edits ${PKGDIR}/Packages with new |
| 21 |
SHA1 and MD5. No way to know. |
| 22 |
|
| 23 |
IMHO SRI_URI is more trustworthy than binhost, in the sense of |
| 24 |
verification. If you have means to verify the binhost stuff it maybe |
| 25 |
more trustworthy. That is left to the admin. |
| 26 |
|
| 27 |
I see binpkg as a temporary convenience. I am doing updates across many |
| 28 |
of the same systems. Less images, containers, etc. I made binaries on |
| 29 |
one system. Immediately used as updated on others. Then discarded on |
| 30 |
binhost. Also used for testing, reverting between slotted versions. |
| 31 |
|
| 32 |
-- |
| 33 |
William L. Thomson Jr. |
Archives