Now that there are several production workloads on the Raspberry Pi
cluster, we don't want intense jobs like this running on them. To
ensure this job runs on a dedicated node, we need to use a label
expression that matches the appropriate nodes.
dustin/sshca-cli/pipeline/head This commit looks goodDetails
The RPM build tools in Fedora 41 apparently require `gdb` but don't
declare that:
> find-debuginfo: starting
> Extracting debug info from 1 files
> gdb-add-index: Failed to find a useable GDB binary
dustin/sshca-cli/pipeline/head There was a failure building this commitDetails
Apparently, `dnf5` differs from `dnf` in that it does not ignore the
trailing `--` argument:
> Unknown argument "--" for command "install". Add "--help" for more
> information about the arguments.
dustin/sshca-cli/pipeline/head This commit looks goodDetails
In addition to building an RPM package for regular Fedora machines, we
now build a container image containing a statically-linked `sshca-cli`
executable.
dustin/sshca-cli/pipeline/head This commit looks goodDetails
The cloud aarch64 build machine does not have enough resources to build
multiple versions at once. Requesting multiple CPUs ensures that only
one build pod is scheduled at a time. Since the node has 2 CPUs and
240m CPUs are taken by Longhorn, if we request 1500m CPUs, builds will
run sequentially.
dustin/sshca-cli/pipeline/head This commit looks goodDetails
When this repository was split from the original *dustin/sshca*
repository, the CI pipeline was not imported. It wouldn't have mattered
if it had been, since it wouldn't have worked, anyway, given the path
changes.
Dustin