Test: Run e2e test on Packit #638
Conversation
| /to-existing-root: | ||
| summary: Run bootc install to-existing-root and bootc switch test locally (nested) | ||
| environment+: | ||
| TEST_CASE: to-existing-root | ||
| discover+: | ||
| test: | ||
| - /to-existing-root | ||
| adjust+: | ||
| - when: arch == ppc64le | ||
| enabled: false |
There was a problem hiding this comment.
This syntax is YAML sorcery to me...is there any description of what the / and + do in particular?
There was a problem hiding this comment.
When configure tmt plan or test, it supports regular expression, starting with / to avoid regular expression, but just a string match. The / comes from plan or test name, for example:
➜ tmt test ls
/tests/e2e/e2e/to-disk
/tests/e2e/e2e/to-existing-root
/tests-integration/build-image
The + means adding addition configuration based on top level configuration. In this case, The top level has adjust configured already.
henrywang
force-pushed
the
new_e2e
branch
24 times, most recently
from
3f74106
to
d649683
Compare
There was a problem hiding this comment.
Hi @cgwalters, I made some changes here to support update bootc.spec to work with copr build requirement. Any comment on this code change? I'm just a beginner for Rust. :-) Thanks.
There was a problem hiding this comment.
Code overall looks fine 😄
|
There is outage in Packit testing farm runner for x86_64 runner. I'll re-run the x86_64 tests after outage fixed. Please don't merge this PR until all tests passed. Thanks. |
| @@ -246,6 +247,41 @@ fn package(sh: &Shell) -> Result<()> { | |||
| Ok(()) | |||
| } | |||
|
|
|||
| fn update_spec(sh: &Shell) -> Result<Utf8PathBuf> { | |||
There was a problem hiding this comment.
Hmm...but this overlaps with package-srpm right? I think in theory we could share code here...or even extract the spec from the srpm?
There was a problem hiding this comment.
This also relates to #651 and the copr makefile...we've grown way too many copies of this logic.
There was a problem hiding this comment.
Code overall looks fine 😄
henrywang
force-pushed
the
new_e2e
branch
2 times, most recently
from
8619cd8
to
01a3ff3
Compare
|
Hi @cgwalters! We can merge this PR now. |
1. rename old integration test to e2e tet 2. drop aws test, use libvirt test (for bootc install to-existing-root and bootc switch test) instead 3. drop rhel test 4. replace quay.io registry by local registry Signed-off-by: Xiaofeng Wang <henrywangxf@me.com>
Signed-off-by: Xiaofeng Wang <henrywangxf@me.com>
Signed-off-by: Xiaofeng Wang <henrywangxf@me.com>
Signed-off-by: Xiaofeng Wang <henrywangxf@me.com>
Signed-off-by: Xiaofeng Wang <henrywangxf@me.com>
Signed-off-by: Xiaofeng Wang <henrywangxf@me.com>
Signed-off-by: Xiaofeng Wang <henrywangxf@me.com>
delete workflow file because e2e runs by Packit delete mockbuild because build bootc rpm on copr Signed-off-by: Xiaofeng Wang <henrywangxf@me.com>
Signed-off-by: Xiaofeng Wang <henrywangxf@me.com>
Signed-off-by: Xiaofeng Wang <henrywangxf@me.com>
Signed-off-by: Xiaofeng Wang <henrywangxf@me.com>
deploy-libvirt playbook Signed-off-by: Xiaofeng Wang <henrywangxf@me.com>
resource limitation Signed-off-by: Xiaofeng Wang <henrywangxf@me.com>
| } else { | ||
| writeln!(o, "{}", line)?; | ||
| } | ||
| } |
There was a problem hiding this comment.
Very minor and we can fix in a followup, but just glancing at this again I notice we're missing a call to into_inner here which will ensure the buffer gets flushed.
While Rust means never having to close a socket it does mean you can forget to flush a buffered writer and then have the ugly semantic where if that fails the file is silently only partially written.
There was a problem hiding this comment.
Learning and learned! Thanks.
This PR includes:
x86_64,aarch64, andppc64learch andRHEL 9,CS9,Fedora-40/rawhidedistros.integrationworkflow removed.