summaryrefslogtreecommitdiff
path: root/src
AgeCommit message (Collapse)Author
2017-07-04Diskimage.imageExists: Align disk image size to multiple of 4096 sector sizeJoey Hess
Since some programs (such as VBoxManage convertdd) refuse to operate on disk images not aligned to a sector size. This commit was sponsored by Boyd Stephen Smith Jr. on Patreon.
2017-07-04Bootstrap.clonedFrom: Fix bug that broke copying .git/config into chroot.Joey Hess
2017-07-02make joey own git.kitenet.net checkoutJoey Hess
I have a cron job updating a file in there..
2017-06-28describe Linode.serialGrubSean Whitton
Signed-off-by: Sean Whitton <spwhitton@spwhitton.name>
2017-06-28Apache.httpsVirtualHost' must create ssl/hn/ dir earlierSean Whitton
It turns out that IncludeOptional will error out if it is passed non-wildcard directories that do not exist. Signed-off-by: Sean Whitton <spwhitton@spwhitton.name>
2017-06-20User.hasInsecurePassword makes sure shadow passwords are enabledJoey Hess
So if the insecure password is later changed, the new password won't be exposed.
2017-06-18my pipe trick didn't work; fallback to NoBufferingJoey Hess
Not sure what the problem was, but it hung. Also though, I noticed that stdin was still open when git fetch was run, so if git fetch itself decided to read from stdin, it would mess up the protocol forwarding. While git fetch should never read from stdin, that was reason enough to fall back to plan B.
2017-06-18Fix bug that sometimes made --spin fail with "fatal: Couldn't find remote ↵Joey Hess
ref HEAD" Tricky stdin buffering problem. An easier fix would have been: hSetBuffering stdin NoBuffering But that approach is less robust; even with NoBuffering, anything that uses hLookAhead causes 1 byte of buffering. And, any reads from stdin before hSetBuffering would still cause the problem. Instead, I used a bigger hammer that will always work. It involves a bit more CPU work, but this is data that is already being fed through ssh; copying it one more time won't cause a measurable performance impact. This commit was sponsored by Jack Hill on Patreon.
2017-06-18remove unnecessary binary mode settingJoey Hess
This is using ByteString, so the handle IO discipline is irrelevant.
2017-06-18Display error and warning messages to stderr, not stdout.Joey Hess
2017-06-17use stretch in debian stable examplesJoey Hess
2017-05-20propellor spinJoey Hess
2017-05-20propellor spinJoey Hess
2017-05-16merge fixes from git-annexJoey Hess
2017-05-15Merge branch 'master' into joeyconfigJoey Hess
2017-05-15Removed dependency on MissingH, instead depends on split and hashable.Joey Hess
MissingH is a heavy dependency, which pulls in parsec and a bunch of stuff. So eliminating it makes propellor easier to install and less likely to fail to build. changesFileContent now uses hashable's hash. This may not be stable across upgrades, I'm not sure -- but it's surely ok here, as the hash is not stored. socketFile also uses hash. I *think* this is ok, even if it's not stable. If it's not stable, an upgrade might make propellor hash a hostname to a different number, but with 9 digets of number in use, the chances of a collision are small. In any case, I've opned a bug report asking for the stability to be documented, and I think it's intended to be stable, only the documentation is bad. NB: I have not checked that the arch linux and freebsd packages for the new deps, that Propellor.Bootstrap lists, are the right names or even exist. Since propellor depends on hashable, it could be changed to use unordered-containers, rather than containers, which would be faster and perhaps less deps too. This commit was sponsored by Alexander Thompson on Patreon.
2017-04-27Merge commit 'b06edbda0478ed57954d716f64f6870d7ae68f63'Joey Hess
2017-04-28Restic: fix bug in shell escapingFélix Sipma
2017-04-27few little thingsJoey Hess
2017-04-27Restic: make sure the repo exists before running restic commandsFélix Sipma
2017-04-27add Restic moduleFélix Sipma
2017-04-10copy git configuration into chrootJoey Hess
2017-04-10reorgJoey Hess
2017-04-09Merge remote-tracking branch 'origin' into joeyconfigJoey Hess
2017-04-09avoid call stackJoey Hess
2017-04-09propellor spinJoey Hess
2017-04-09propellor spinJoey Hess
2017-04-09propellor spinJoey Hess
2017-04-09avoid "sh: 1: git: not found" before auto-install of gitJoey Hess
2017-04-09propellor spinJoey Hess
2017-04-09new approach for exposing the underlying localdir inside a chrootJoey Hess
2017-04-09well, that didnt work :(Joey Hess
2017-04-09added Propellor.Property.Bootstrap (untested)Joey Hess
This commit was sponsored by Jake Vosloo on Patreon.
2017-04-06tweaks to db45x's patchJoey Hess
Removed mountNow as a top-level property, as I don't think it makes sense for anything except for mounted to use it. db45x's patch turns out to have introduced a bug in mounted's use of "mountNow src". That made mountNow check if the device was a mount point, which it isn't. The fix would have been to use "mountNow mnt", but my inlining of mountnow just basically reverted the part of the patch that introduced the bug. swapOn does not involve the fstab so moved to the Mount module. (Also noticed that Mount.mounted is a kind of weird property, given that it fails the next time ran. It's only used internally by some chroot properties, so I left it as-is, but added a comment. It might make sense to make Mount.mounted check like mountNow does if the thing is already mounted.)
2017-04-06listed property should work on !LinuxJoey Hess
2017-04-04break Fstab.mounted into smaller pieces which can be composedDaniel Brooks
2017-04-04Merge branch 'master' of https://git.joeyh.name/git/propellor into fstab-swapDaniel Brooks
2017-04-04property 'mount' a swap partition in Fstab.mountedDaniel Brooks
2017-04-03avoid opendkim checks on incoming emailJoey Hess
This caused a problem when eg a gmail email was sent to branchable and forwarded on to kite. Kite thought branchable was forging the dkim signature of gmail, and gmail does strict enforcement of dkim, so it rejected it. DKIM seems like a mightly hefty hammer, and I remember similar issues being dicussed where mailing list software got broken by DKIM. I guess this will mean a few more joe-jobs get through, but I'd rather not silently lose important legitimate email!
2017-03-24fix diskimage finalization bugJoey Hess
* Fix bug when using setContainerProps with a chroot that prevented properties added to a chroot that way from being seen when propellor was running inside the chroot. This affected disk image creation, and possibly other things that use chroots. The problem was, propagateChrootInfo was being passed the initial version of the Chroot, but then the Chroot got more properties added, and so those were not recorded in the _chroot info. Fix was simply to make InfoPropagator be passed the Chroot as an additional parameter, so Chroot.provisioned' can pass in the final Chroot to it.
2017-03-24Merge branch 'master' into joeyconfigJoey Hess
2017-03-24Property.Partition: Update kpartx output parser, as its output format ↵Joey Hess
changed around version 0.6. Both output formats are supported now. Do we trust kpartx to not change again, given how little documented its output format is and that it's changed w/o warning? Little other things like vmdebootstrap depend on its output in the same way.. This commit was sponsored by Ignacio on Patreon.
2017-03-24Property.Partition: Improve loop device detection code used with kpartx when ↵Joey Hess
building disk images. /dev/mapper/loop* symlinks are now detected as block devices.
2017-03-19Apt.mirror can be used to set the preferred apt mirror of a host, overriding ↵Joey Hess
the default CDN. This info is used by Apt.stdSourcesList and Sbuild.builtFor. Thanks, Sean Whitton.
2017-03-19Merge remote-tracking branch 'spwhitton/apt-mirror'Joey Hess
2017-03-19Merge branch 'joeyconfig'Joey Hess
2017-03-19mention mconcatJoey Hess
2017-03-19switch to kite4Joey Hess
lost kite2 ed25519 key
2017-03-19Tor.named, Tor.torPrivKey: Include the new ed25519 public/private key pair ↵Joey Hess
in addition to the old secret_id_key.
2017-03-19shorten names of mirror propertiesSean Whitton