dmidecode for alpine 3.11 doesn't work in this device type. This change
moves to using `/proc/device-tree/product-sn` and
`/proc/device-tree/product-name` for these devices.
Resolves: #1916
Change-type: patch
Migration `M00008` had a bug with the check for legacy apps, which
resulted in devices that had at some point been updated from a single
container supervisor to get the error
```
Undefined binding(s) detected when compiling UPDATE. Undefined column(s): [appUuid] query
```
This adds a new migration with the fix to ensure broken fix the
inconsistent database state.
Change-type: patch
Closes: #1913
If an app is not in the target state means the supervisor no longer
has permissions to that app hence it cannot report on it. When moving
between apps, there is a transitional period where containers and images
from both apps can be in the current state, therefore filtering is
needed to prevent getting 401 errors from the API.
Starting with v3 state endpoint, the supervisor may receive the configuration
for the supervisor service on the target state. This commit allows the
supervisor to filter out the supervisor container from the current and target
state to let the update-balena-supervisor script handle the creation and update
of the supervisor container.
Updating and creating the supervisor container will be handled by a
future commit
Starting with v3 state endpoint, the supervisor can receive
service configuration for services that are meant to be installed as
overlays or filesets on the host, as well as configuration for services
that are meant to be installed on the root partition. This commit just
ignores those services from the target state until support is added
Local mode is still a device level config. Eventually it will become a
property of an app, but for now, we don't want the supervisor trying to
uninstall supervisor or host app when local mode is set
This change makes the `api-binder/report` module more agnostic
to internal device state implementation details, moving necessary
healthchecks and data filtering to getCurrentForReport in device-state.
This also adds generic functions to perform comparison between current
state reports.
The role of the api-binder module is to be the intermediary
between the cloud API and the device-state. For this reason it makes sense to
isolate target state retrieval and current state reporting into this
module. This change just moves current state reporting to the directory.
This is required as we are phasing out app ids and we need to be able to
get app uuid from the current state of the network. The app-id now
exists as a container in new networks
This commit will restart containers as it needs to recreate the network.
Removed redundant `getCurrentAppsForReport` and `getCurrentForComparison` since
the behavior of these methods is already handled by `getCurrentApps` and
`getCurrentState`.
Creates `lib/legacy.ts` and `device-state/legacy.ts` to deal with
migration from legacy target states (single container and v2) for all
apps and for apps.json respectively
This change updates types and database format in order to allow
receiving the new format of the target state from the cloud and allow
applications to keep working.
This change also updates metadata in the containers, meaning services
will need to be restarted on supervisor update
Change-type: major
With the move to v3 target state and the move forward to remove
database ids from the supervisor, we want to ensure the ids are only
used for legacy support (such as within the API). This change renames
the method and sets it as deprecated
It seems that in some cases the supervisor can report
an image without a `status` field leading to a cloud side 401 response.
See #1905 for more details.
Change-type: patch
The check for the docker network supervisor0 assumed that if the
interface supervisor0 existed, then the network would exist too. However this is not
true on the case of docker directory corruption, which would lead to a
loop with `Error: (HTTP code 404) no such network - network supervisor0 not found`.
Change-type: patch
Closes: #1806
As changes to config.json may restart the supervisor before it can
trigger the reboot (or something can kill the supervisor before it can run that step),
the supervisor needs a persistent signal that a reboot is required
(instead of the current transient signal).
With this commit, the supervisor will now create a breadcrumb in the
host `/tmp` folder, that will be checked as the last step of the
configuration changes.
As config.json changes may restart the engine (and hence the supervisor)
in newer OS versions, this ensures that the supervisor does not get
interrupted while writing to backends.
This is necessary with the changes as of balenaOS 2.82.6, which watches config.json
and will restart balena-hostname and some other services automatically on file change.
Change-type: patch
Relates-to: #1876
Signed-off-by: Christina Wang <christina@balena.io>
The functionality is pretty much the same, so we don't need the two
functions in two different places.
Signed-off-by: Christina Wang <christina@balena.io>
With more and more devices in ipv6 only networks, this ensures the
local addresses are reported to the cloud as part of the state patch.
Change-type: patch
`/mnt/boot` is a vfat partition which does not support atomic file
rename. The best course of action is to write and sync as fast as
possible to prevent corruption (although it still may happen)
Change-type: patch
The API uses 304 as a mechanism for load management on target state
requests. This may cause that the supervisor receives a 304 response
without having received a copy of the target state first, leading to
issues. This change checks for an etag when receiving a 304, throwing an
exception otherwise.
Change-type: patch
Cpu id is set to null so far for non ARM devices (e.g. Intel NUC). This
parses the output of dmidecode to get the cpu id and system model.
Change-type: patch
Cpu id is set to null so far for non ARM devices (e.g. Intel NUC). This
parses the output of dmidecode to get the cpu id and system model.
Change-type: patch
This avoids the supervisor trying to get back to the preloaded target
state if the database is deleted by any reason. It does this by moving the
used apps.json to a backup location.
Change-type: patch
Depends-on: #1841
Happy-eyeballs performs [dns lookups](https://github.com/balena-io-modules/happy-eyeballs/blob/master/src/happy-eyeballs.ts#L23)
for the requested addresses, however, because of the order of imports it
was not using the supervisor custom `dns.lookup` that handles `.local`
name resolution, making address resolution fail in those cases.
Moving the import after the `dns.lookup` patch fixes the problem.
The supervisor performs its own local resolution for `.local`
addresses due to a limitation in [musl](https://wiki.musl-libc.org/future-ideas.html).
The resolution function was not following exactly the nodejs [dns.lookup
specification](https://nodejs.org/api/dns.html#dnslookuphostname-options-callback)
which could cause certain clients to fail (in this case happy-eyeballs). This
updates the function to follow the specification.
Change-type: patch
The supervisor always applies target state on start to ensure that the
device is at the correct in case of a crash or another reason. This had
the side effect that if the database is deleted, the supervisor would
apply target state (which is empty), stopping services and possibly
causing volume data loss.
This prevents that behavior and ensures that the supervisor only
applies target state if a target has been set either by the cloud, preload or local
mode.
Change-type: patch
