Colin Walters [Sun, 19 Apr 2015 16:34:01 +0000 (12:34 -0400)]
deploy: Drop fsync of modified config files
These fsyncs were added for what turned out to be a fairly bogus
reason; I was hitting read errors from extlinux after upgrades and out
of conservatisim tried adding fsync calls, but the *actual* problem
was that extlinux didn't support 64 bit ext4. Now that at least for
Project Atomic hosts we're just targeting grub2, we can drop these
fsync calls and rely on `syncfs()` being both faster and catching any
errors.
Colin Walters [Tue, 14 Apr 2015 17:47:08 +0000 (13:47 -0400)]
deploy: Use syncfs() in addition to sync()
For some sort of crazy reason, the `sync()` system call doesn't
actually return an error code, even though from what I can tell in the
kernel it wouldn't be terribly hard to add.
Regardless though, it is better for userspace apps to use `syncfs()`
to avoid flushing filesystems unrelated to what they want to sync. In
the case of OSTree, this does matter - for example you might have a
network mount point backing your database, and we don't want to block
upgrades on syncing it.
This change is safe because we're doing syncfs in *addition* to the
previous global `sync()` (a revision from an earlier patch).
Now because OSTree only touches the `/` mount point which covers the
repository, the deployment roots (including their copy of `/etc`), as
well as `/boot`, we should at some point later be able to drop the
`sync()` call. Note that on initial system installs we do relabel
`/var` but that shouldn't happen at ostree time - any new directories
are taken care of via `systemd-tmpfiles` on boot.
Matthew Barnes [Fri, 17 Apr 2015 13:00:17 +0000 (09:00 -0400)]
sysroot: Cache an OstreeRepo instance
Rather than returning a new OstreeRepo instance in each call to
ostree_sysroot_get_repo(), cache one internally so the same instance
is returned each time.
Matthew Barnes [Mon, 13 Apr 2015 17:21:17 +0000 (13:21 -0400)]
repo: Add a "gpg-verify-result" signal
Emitted during a pull operation upon GPG verification (if enabled).
Applications can connect to this signal to output the verification
results if desired.
Colin Walters [Wed, 8 Apr 2015 00:54:37 +0000 (20:54 -0400)]
reset: Don't enforce parent commits
First, git doesn't do this, and whatever Linus thinks is right or
something.
Second specifically to OSTree, it's quite common to not have
intermediate commits. If one wants to reset a ref in order to prune
data after a deployment, the parentage check will fail.
Colin Walters [Thu, 9 Apr 2015 01:54:28 +0000 (21:54 -0400)]
refs: Use *at for writes, honor repo fsync flag
I was looking at https://bugzilla.gnome.org/show_bug.cgi?id=738954
which wants us to ensure we chown() the refs. As part of that,
I did a generic conversion to use `*at()` (which naturally gives
us more low level control so we can call `fchown` etc.
This patch also sneaks in a change to respect the repo's
`disable_fsync` flag - if fsync is not set, then we never
`fdatasync()` (unlike the `g_file_replace_contents()` default. Also
unlike it, if fsync is enabled, we *always* sync even if the file
didn't exist.
Colin Walters [Thu, 12 Feb 2015 14:13:30 +0000 (09:13 -0500)]
Add an API to set/unset a deployment tree's mutability
This will be used by rpm-ostree to unset the immutable bit temporarily
in order to do package layering. We could add an API to deploy a tree
without the immutable bit, but this is simpler.
Colin Walters [Mon, 6 Apr 2015 22:29:01 +0000 (18:29 -0400)]
ostree_repo_checkout_tree_at: New API for checkouts
rpm-ostree currently uses ostree_repo_checkout_tree(), which as a side
effect will use the uncompressed objects cache by default. This is
rather annoying if you're using rpm-ostree on a server-side
repository, because if you then rsync the repo, you'll be syncing out
the uncompressed objects unless you exclude them.
We added the ability to disable the uncompressed cache in the
repository config to fix this, but it's better to allow application
control over this. The uncompressed cache will in some future version
become opt in as well.
This new API further:
- Drops the `GFile` usage in favor of `openat` APIs
- Improves ergonomics by avoiding callers having to query the source
`GFileInfo` (and carry around a copy of `OSTREE_GIO_FAST_QUERYINFO`)
- Has a more extensible options structure
Per the comment, I rather crudely have the `ostree checkout` builtin
call both APIs to ensure some testing coverage.
However, I'd like to in the future have easier-to-set-up testing code
that calls `libtest.sh` to set up dummy data.
Colin Walters [Sat, 4 Apr 2015 14:49:28 +0000 (10:49 -0400)]
pull: Handle remote web server not honoring range requests
It's valid for the remote server to say 200 OK and give us the entire
file instead of a 206 Partial Content, and in that case we should blow
away the previous cached data, rather than blindly appending to it and
thus creating multiple copies of the data inside the file.
This problem primarily occurs when we do have the complete file, and
we're interrupted, then try again, where the new process didn't record
the download was already complete. We do a range request for bytes
past the end, and some web servers (e.g. Akamai) will return 200 OK
with the whole content again, rather than a 416 Requested Range Not
Satisfiable.
Thus we could also fix this by saner caching strategy - since we know
the file is complete, rename it again to $checksum.done or something
before it's processed. (Or really, rework how we do caching more
intelligently in general).
This fixes the issue that interrupted pulls failed with such
webservers, although repeated attempts would eventually succeed
because we'd unlink files that failed to pull.
Colin Walters [Wed, 1 Apr 2015 10:21:34 +0000 (06:21 -0400)]
main: Only verify SUPERUSER flag if using default sysroot
The use case for non-default sysroots that I know of are:
1) The current test suite
2) Installers (Anaconda)
3) Inspecting VM disks
For 2) and 3), it'll quickly be obvious if they're not running as
root, and these are more obscure cases. We want to allow 1), and this
is a simple way to do it.
Sam Thursfield [Tue, 31 Mar 2015 16:59:43 +0000 (12:59 -0400)]
core: Fix possible crash in ostree_mutable_tree_walk()
If the starting index is beyond the end of the list, it's a programming
error. Previously, the code was trying to raise a runtime error, but
actually causing a segfault.
This was detected by test code in test-mutable-tree.c, which is removed
in this commit because it should now not be possible to crash here.
Matthew Barnes [Fri, 16 Jan 2015 15:04:18 +0000 (10:04 -0500)]
Add ostree_ensure_repo_writable()
Commands that need to write files within the repo directory can call
this early to ensure the directory is writable for the current user.
If not, it fails with a helpful "You need to be root to perform this
command" message.
Matthew Barnes [Thu, 19 Mar 2015 16:43:04 +0000 (12:43 -0400)]
gpg: Regenerate test data for test-gpg-verify-result
Turns out the expired signature case was failing because the signature
itself was corrupted. Reconstructed the test data and updated the test
code. Note, an expired signature is still counted as valid.
Also, handy debugging trick for setting a key or signature expiry: the
CLI makes it appear the shortest expiry is 1 day, but it also secretly
recognizes "seconds=N".
Matthew Barnes [Wed, 4 Mar 2015 17:44:47 +0000 (12:44 -0500)]
gpg: Add OstreeGpgVerifyResult
Wrappers a referenced gpgme_verify_result_t so detailed verify results
can be examined independently of executing a verify operation.
_ostree_gpg_verifier_check_signature() now returns this object instead
of a single valid/invalid boolean, but the idea is for OstreeRepo to also
return this object for commit signature verification so it can be utilized
at the CLI layer (and possibly by other programs).
Dan Nicholson [Tue, 17 Mar 2015 19:20:53 +0000 (12:20 -0700)]
fsck: Fix object count output
The object count comes from g_hash_table_size(), so it's not a 0 based
index. In order to maintain the mod calculations correctly, just print
out index + 1.
Matthew Barnes [Mon, 16 Mar 2015 17:01:55 +0000 (13:01 -0400)]
OstreeGpgVerifier: Take the signed data as a GBytes
Similar to c2b01ad. For some reason I was thinking the commit data
still needed to be written to disk prior to verifying, but it's just
another artifact of spawning gpgv2 (predates using GPGME).
Makes for a nice cleanup in fetch_metadata_to_verify_delta_superblock()
as well.
Matthew Barnes [Tue, 3 Mar 2015 19:15:27 +0000 (14:15 -0500)]
OstreeGpgVerifier: Take the signature as a GBytes
The signature data is in memory to begin with, so there's no need to
write it to disk only to immediately read it back.
Also, because the GPGME multi-keyring workaround is somewhat expensive
to setup and teardown, concatenate all signatures into a single GBytes
so _ostree_gpg_verifier_check_signature() is only called once. We're
currently only looking for one valid signature anyway.
Matthew Barnes [Mon, 2 Mar 2015 16:16:16 +0000 (11:16 -0500)]
gpg: Rewrite OstreeGpgVerifier to use GPGME
This sets the stage for more advanced signature management.
(Also, talking to GPG over pipes sucks.)
Previously we were spawning gpgv2 with a bunch of --keyring options
for /usr/share/ostree/trusted.gpg.d/ and whatever other keyring files
were explicitly added. GPGME has no public API for multiple keyrings,
so we work around the issue by setting up a temp directory to serve as
a fake "home" directory for the crypto engine and then concatenate all
the keyring files into a single public keyring (pubring.gpg).
Unfortunately at present we do this on every signature verification.
There's a desire to cache this concatenation, but the problem is the
user may be unprivileged. So it seems the cache would have to be per
user under $XDG_CACHE_HOME, which OSTree doesn't otherwise use. I'm
open to suggestions.
We do at least clean up the temp directory when finished, and I have
further API changes planned to OstreeGpgVerifier to help mitigate the
performance impact.
projects / ostree.git / log