| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
| |
This is consistent with the timestamps in log files. Previously, running `morph
--log=/dev/stdout' mixed local time with UTC, confusingly. It was jarring to
have status output on screen which didn't match the time zone of the computer
on which it was running, too.
|
| |
|
|
|
|
|
| |
There is significant overhead involved in making a HTTP request, so it's
worth the extra code to batch these requests as much as possible.
|
|
|
|
|
|
| |
Previously if a ref didn't resolve in the remote cache, we'd clone the repo into the local
cache and check again. However, 99% of the time the clone comes from the remote cache so
we should be able to assume that if the ref doesn't exist there, it doesn't exist at all.
|
| |
|
|
|
|
| |
This should be a useful performance boost.
|
|
|
|
|
| |
The logic of this is still rather twisted, but now it can be shared by a
batched get_morphologies() function.
|
| |
|
|
|
|
|
|
|
|
|
|
| |
When the given ref points to a specific commit, and it's already available in the
locally cached copy of repo, there's no need to update the repo.
If the ref points to a branch or tag, and the user didn't pass --no-git-update, the
locally cached copy of the repo will still be updated.
This should speed up many Morph commands.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
If btrfs is not present in the kernel we end up with strange output like
this:
Error creating disk image2014-06-10 16:00:40
[devel-system-x86_64-generic][my-raw-disk-image][rawdisk.write]Failure
to create disk image at /src/tmp/testdev.img
ERROR: Command failed: mount -o loop /src/tmp/testdev.img
/src/tmp/deployments/tmpQ7wXO1/tmp4lVDcu/tmpvHSzDE
mount: mounting /dev/loop0 on
/src/tmp/deployments/tmpQ7wXO1/tmp4lVDcu/tmpvHSzDE failed: Device or
resource busy
To avoid this confusing error, Morph should explicitly check first.
|
|\
| |
| |
| |
| |
| | |
Reviewed by:
Lars Wirzenius
Richard Maw
|
| |
| |
| |
| |
| |
| | |
After a failed attempt to connect to the controller node
the initiator will wait 30 seconds before attempting a reconnect,
if this reconnect fails the initiator gives up.
|
| | |
|
| |
| |
| |
| | |
This change is made just for consistency.
|
| |
| |
| |
| |
| |
| | |
The InitiatorConnectionMachine wraps the ConnectionMachine,
so we can continue to use ConnectionMachine without providing
it with an app.
|
|/
|
|
|
|
|
| |
By default there is no limit on the number of reconnection attempts.
We make the reconnect_interval a parameter, but the default
interval remains 1 second.
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Currently the mainloop unintentionally busy-waits if
we can't connect to the controller.
We want the mainloop's select call to wait for the timeout
not for this socket's descriptors (which are always ready).
We could just call stop_reading() and stop_writing() but since
we won't be needing this socket again we may as well close the
entire event source, which calls stop_reading(), stop_writing()
and then closes the socket.
|
| |
|
|\
| |
| |
| |
| | |
Reviewed-By: Richard Ipsum <richard.ipsum@codethink.co.uk>
Reviewed-By: Richard Maw <richard.maw@codethink.co.uk>
|
| | |
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Artifacts can have multiple parts; while this may not be an ideal
design, changing the format of artifacts has implications for backwards
compatibility.
We should transfer all parts at once and delete them all if we encounter
any errors, to reduce the change of getting the local artifact cache
into an inconsistent state.
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Previously Morph would check if an artifact is present in the remote
artifact cache, then fetch the necessary files. If an error occured
during fetching, it would raise an error and abort.
Instead, we should just try and fetch the files and if anything fails,
move on to building locally. This avoids the situation where an error in
the remote cache makes local building impossible, which we experienced
recently.
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Before:
ERROR: Failed to get metadata meta for the artifact
file:///src/ws-baserock-hawk/baserock/ps/build-system/baserock/baserock/definitions|refs/heads/baserock/builds/778b1a370a1f43c497c1354a2a949de1/56c9ec89d09240fd80faa7d2226b7eda|core|core-devel
from the artifact cache http://git.baserock.org:8080/
After:
ERROR: Failed to get metadata meta for the artifact
f896a081beacd4a99ded38d28b44fbf02970038fb53349265f85f8f3298ead9d.stratum.core-devel
from the artifact cache http://git.baserock.org:8080/
When debugging artifact cache issues, the information that's most
useful is the filename of the artfact.
|
| |
| |
| |
| | |
This saves from having to catch three separate exceptions.
|
|\ \
| | |
| | |
| | |
| | | |
Reviewed-by: Adam Coldrick
Reviewed-by: Pedro Alvarez
|
|/ /
| |
| |
| |
| |
| |
| |
| | |
The deploy plugin is relying on existing code which respects the
'no-git-update' setting and updates gits if it is not set.
Since deploy can only work for built systems anyway, we can force this
True for deployments, to avoid wasting the user/caller's time.
|
|\ \
| | |
| | |
| | | |
Reviewed-by: Sam Thursfield and Richard Ipsum
|
| | | |
|
| | | |
|
|\ \ \
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Reviewed-By: Adam Coldrick <adam.coldrick@codethink.co.uk>
Reviewed-By: Richard Ipsum <richard.ipsum@codethink.co.uk>
Reviewed-By: Richard Maw <richard.maw@codethink.co.uk>
Reviewed-By: Sam Thursfield <sam.thursfield@codethink.co.uk>
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
- All tests now pass
- The odd case of chunks with the same name but different repo URLs now
correctly informs the user of the multiple checkouts that were done.
- Tidyups
|
| | | | |
|
| | | | |
|
|/ / /
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Since we removed ref: fields from the system morph files, some of the logic
in morph edit is no longer needed. In particular, running
morph edit <system> <stratum>
is a no-op. This is because the version of <system> and <stratum> are now
implicit from the context of what we are doing. In other words we're always
working with the current version of <system> and <stratum> from the system
branch we are in.
Because of the complexity of morph's logic, we didn't notice this when
dropping the ref: fields, and we missed the opportunity to simplify our
logic for 'morph edit'
This patch aims to provide the simplest possible implementation of
morph edit:
morph edit <chunk>
It checks all strata for instances of <chunk>, and does what morph edit should
do for the instances it finds.
A later patch can add warnings to help users deal with situations where <chunk>
is not found, or is found more than once.
Also since this changes the syntax of morph, it breaks many of our tests. Later
patches will address that.
|
|/ /
| |
| |
| |
| |
| |
| |
| | |
Generally deployment temp dirs are removed by morph during deployment,
but in some cases deployment dirs may not be cleared up,
for example if morph gets a SIGKILL or something unexpected
happens that causes morph to terminate without having a chance
to cleanup.
|
|\ \
| | |
| | |
| | |
| | | |
Reviewed-by: Pedro Alvarez
Reviewed-by: Sam Thursfield
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
This makes systems use the UUID of the disk in the fstab when there is
no pre-existing fstab entry for /.
This happens whether the system has an initramfs or not, since it should
be harmless, and by this point we're in userland, so can know what the
UUIDs are.
Passing the UUID to `complete_fstab_for_btrfs_layout` is optional, and
defaults to the old behaviour of using /dev/sda, since it is called
directly by some write extensions for doing upgrades, and upgrading
systems that use an initramfs will be part of a later patch.
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
It's only ever called from the core of the write extensions library, so
we can change it to be mandatory in all cases.
install_extlinux is left with the uuid being optional and defaulting to
/dev/sda, since it is called directly from the rawdisk write extension
currently, and upgrades of systems that have an initramfs will be part
of a later patch series.
|
| | | |
|
|\ \ \
| | | |
| | | |
| | | |
| | | | |
Reviewed-By: Adam Coldrick <adam.coldrick@codethink.co.uk>
Reviewed-By: Richard Maw <richard.maw@codethink.co.uk>
|
| | |/
| |/|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
The copy-artifacts and list-artifacts commands are mainly useful for
making releases. As part of the release process we copy artifacts for
the entire build graph of the release to the artifact cache on
trove.baserock.org, to provide Baserock users with ready-built
artifacts.
This part of the release process is now automated, and the automation
require the list-artifacts command to function as a 'plumbing' command.
The copy-artifacts command is no longer required. It can be replaced
with:
morph list-artifacts --quiet REPO REF MORPH | rsync --files-from=- $TARGET
The previous version of this plugin looked in the system artifact's
metadata for the list of artifacts. This is flawed as the final system
does not necessarily contain every build dependency. The new version of
the plugin calculates the build graph from source, using the same
process as the 'buildcommand' module. It also required looking in
Morph's artifact cache for the system artifact file to analyse.
|
| | |
| | |
| | |
| | | |
We always want to warn if we attempt to remove a job that's not present
|
| | | |
|
|\ \ \
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Reviewed by:
Sam Thursfield
Adam Coldrick
Richard Maw
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
If a new build request makes a request for an artifact that is currently being
cached then the artifact will be needlessly rebuilt.
To avoid this the new build request should wait for caching to finish.
We rename _ExecStarted, _ExecEnded, _ExecFailed to
_JobStarted, _JobFinished, _JobFailed
and Job's is_building attribute is renamed to running.
|
| |/ /
| | |
| | |
| | |
| | | |
This fixes the bug that causes the distbuild controller
to crash when population of the artifact cache fails.
|
| |/
|/| |
|
|\ \
| |/
|/|
| | |
Reviewed-by: Sam Thursfield and Richard Ipsum
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
If INITRAMFS_PATH is specified and the file exists, then the produced
kernel command line will use root=UUID=$uuid_of_created_disk rather than
root=/dev/sda, which may be incorrect.
Help files have been updated to mention the new option.
This leads to an unfortunate duplication of the path to the initramfs,
in both the location field of the nested deployment and the
INITRAMFS_PATH of the disk image creation.
However, an initramfs could be produced by a chunk and put in the same
place, so it doesn't make sense to couple the rawdisk and initramfs
write extensions to remove this duplication.
Similarly, there may be multiple valid initramfs in the rootfs e.g.
extlinux loads a hypervisor, which is Linux + initramfs, and the
initramfs then boots a guest Linux system, which uses a different
initramfs.
This makes it important to explicitly let the rootfs write extensions
know which to use, or not as the case may be.
util-linux's blkid is required, since the busybox version ignores the
options to filter its output, and parsing the output is undesirable.
Because syslinux's btrfs subvolume support is limited to being able to
use a non-0 default subvolume, the initramfs has to be copied out of
the run-time rootfs subvolume and into the boot subvolume.
This pushed the required disk space of a minimal system over the 512M
threshold because we do not have the userland tooling support to be able
to do a btrfs file contents clone.
|