Make feature/buster build
The patches I made make at least feature/buster branch build again an iso, that boots.
- Subject changed from make feature/buster build to Make feature/buster build
- Status changed from New to In Progress
- Assignee changed from intrigeri to CyrilBrulebois
- Target version changed from Tails_4.0 to Tails_3.10.1
- % Done changed from 0 to 30
Thanks hefee! I'd like this to go through your team-mate before it lands on my plate for a final round of review, in order to lower a bit the bus factor :)
I'm setting target version closer than 4.0 because even though that's 4.0 material, this branch should be reviewed and merged during the 3.10 cycle.
#8 Updated by CyrilBrulebois 4 months ago
- File 0001-Upgrade-Linux-to-4.18.0-2.patch View added
- QA Check changed from Ready for QA to Dev Needed
A few comments on the patch series:
298857d5bd146d42d614851dcdcf8ae13a2bfefalooks good given both parents of the merge commit, but this makes me wonder whether we should keep using the Stretch-built Thunderbird. It's probably easier at the moment if it stays installable on Buster, since I guess they'll be a moment until Buster gets its own Thunderbird, and until we figure out what to do with our patchset anyway. A different question though: since we're moving to a Buster baseline, is it still needed to fetch those packages from sid anyway? [more below]
4960ef9eb0312a3a0f1e30ff211b9dd5503da654has a commit message that mentions
ld.so.cachebut it changes lines related to the configuration instead (
ld.so.conf*) so I guess a commit message fixup is in order?
ae104152754b2151e7d592fecca2056da2051c03looks good (except with “dont” → “don't” in commit message).
An extra commit is needed now, to get a newer linux-image package (see attached patch).
Back to the Thunderbird topic, I'm seeing this:
Some packages could not be installed. This may mean that you have requested an impossible situation or if you are using the unstable distribution that some required packages have not yet been created or been moved out of Incoming. The following information may help to resolve the situation: The following packages have unmet dependencies: thunderbird : Depends: libvpx4 (>= 1.6.0) but it is not installable E: Unable to correct problems, you have held broken packages. P: Begin unmounting filesystems...
And that's not entirely surprising because the pinning is now:
Explanation: Dependencies for Thunderbird built for Stretch (which we install since we haven't bothered rebuilding for Buster yet) which is not in Buster Package: libevent-2.0-5 libhunspell-1.4-0 libvpx4 Pin: release o=Debian,n=sid Pin-Priority: 999
and that doesn't work because of:
$ for p in libevent-2.0-5 libhunspell-1.4-0 libvpx4 ; do rmadison $p -s stable,testing,unstable -a amd64; echo; done libevent-2.0-5 | 2.0.21-stable-3 | stable | amd64 libevent-2.0-5 | 2.0.21-stable-3 | unstable | amd64 libhunspell-1.4-0 | 1.4.1-2+b2 | stable | amd64 libvpx4 | 1.6.1-3+deb9u1 | stable | amd64
And our configured APT sources are:
# /etc/apt/sources.list deb http://time-based.snapshots.deb.tails.boum.org/debian/2018101102 buster main contrib non-free # /etc/apt/sources.list.d/sid.list deb http://time-based.snapshots.deb.tails.boum.org/debian/2018101102 sid main contrib non-free # /etc/apt/sources.list.d/tails.list deb http://deb.tails.boum.org/ devel main contrib non-free deb http://deb.tails.boum.org/ feature-buster main contrib non-free # /etc/apt/sources.list.d/torproject.list deb http://time-based.snapshots.deb.tails.boum.org/torproject/2018101102 buster main
Should we bring back a
stretch line until we figured out what to do with Thunderbird?
#17 Updated by CyrilBrulebois about 2 months ago
Just pushed an updated
- TB from sid is installed at the moment (#16101).
- Many patches were unfuzzied/refreshed/dropped as needed (packages getting updated in buster).
VM_MEMORY_BASEwas bumped to 2048M.
- Electrum support was disabled (#16204).
- It contains the merge from
develthat didn't happen during the 3.11 release cycle (#16212).
- It bumps the Linux kernel to 4.19 (#16073) as the 4.18 toolchain wasn't available.
That built fine for me, including the IMG generation from the ISO file; but I didn't check the runtime yet.