QA » openQA Project

I'm wondering what's wanted here exactly. Which "items" should be added to the TODO list? How should this "single-line todo list" look like?

Take a look on a generic report like https://openqa.io.suse.de/openqa-review/openqa_suse_de_status.html . There are sections like "Existing Product bugs:", "New Test issues:", "Skipped tests", etc. And there is also a section "TODO: review" sub-sectioned into "new" and "existing" issues. The ticket is about only the last section. So a report like https://github.com/os-autoinst/openqa_review/blob/master/tests/tags_labels/report25_T_bugrefs_softfails.md with content

# 0


**Date:** 2017-07-07 - 14:36
**Build:** 1507 (reference 1500)

---

**Arch:** i586
**Status: <span style="color: red;">Red</span>**

**New Product bugs:**

* soft fails: [custom_bugref](https://openqa.opensuse.org/tests/1316928) -> [boo#1126782](https://bugzilla.opensuse.org/show_bug.cgi?id=1126782)
* [toolchain_zypper](https://openqa.opensuse.org/tests/384324 "Failed modules: install") -> [boo#931571](https://bugzilla.opensuse.org/show_bug.cgi?id=931571)
* [toolchain_zypper@bar](https://openqa.opensuse.org/tests/3843245 "Failed modules: install") -> [boo#9315715](https://bugzilla.opensuse.org/show_bug.cgi?id=9315715)
* soft fails: [create_hdd_textmode](https://openqa.opensuse.org/tests/447901) -> [boo#931572](https://bugzilla.opensuse.org/show_bug.cgi?id=931572)
* soft fails: [soft_fail_without_bugref](https://openqa.opensuse.org/tests/684839) -> for Leap:15.0:Ports aarch64 and ppc64le, do not enable source repo, waiting for [https://progress.opensuse.org/issues/36256](https://progress.opensuse.org/issues/36256) to be solved, and also [http://fate.suse.com/12345](http://fate.suse.com/12345) ✓


**Existing Product bugs:**

* [allpatterns](https://openqa.opensuse.org/tests/384333 "Failed modules: xterm"), [ext4@i586--l2](https://openqa.opensuse.org/tests/384634 "Failed modules: install_and_reboot"), [gnome](https://openqa.opensuse.org/tests/384344 "Failed modules: xterm"), [minimal+base](https://openqa.opensuse.org/tests/384329 "Failed modules: dns_srv") -> [bsc#822770](https://bugzilla.opensuse.org/show_bug.cgi?id=822770)
* soft fails: [btrfs](https://openqa.opensuse.org/tests/384330), [zfcp@i586-zfcp](https://openqa.opensuse.org/tests/384707) -> [bsc1029660](https://bugzilla.suse.com/show_bug.cgi?id=1029660)


**Existing openQA-issues:**

* [textmode](https://openqa.opensuse.org/tests/384342 "Failed modules: dns_srv") -> [poo#11948](https://progress.opensuse.org/issues/11948)
* [allpatterns@i586--l2](https://openqa.opensuse.org/tests/384335) -> [poo#9772](https://progress.opensuse.org/issues/9772)


**TODO: review**

***existing issues***

* [ssh-X@i586--l3](https://openqa.opensuse.org/tests/384325 "Failed modules: bootloader_i586")

could look something like this to have only single-line entries:

* **existing:** 0: [i586-ssh-X@i586--l3](https://openqa.opensuse.org/tests/384325 "Failed modules: bootloader_i586")

tinita wrote:

@okurz here is a screenshot.
Does that go in the right direction?

Yes, looks good! Goes in the right direction :)

@okurz
Your proposal was:

* **existing:** 0: [i586-ssh-X@i586--l3](https://openqa.opensuse.org/tests/384325 "Failed modules: bootloader_i586")

Now I wonder what the 0: here stands for. In the original report that was the Job Group.
But I don't get how the complete report should look like then.
Should all headings be skipped and we only have one list? Then the single items should also contain the architecture.

tinita wrote:

[…]
Should all headings be skipped and we only have one list? Then the single items should also contain the architecture.

yes, try to skip all headings. I have not included the architecture because the "machine" is always architecture specific so we might not need the architecture at all when we have the machine. Let's try without. But consider what is feasible from implementation point of view. Feel free to propose a PR with a new mode that only does a part and not yet all. We can go step-wise

That's going to change a lot, because that deep in the report generation there is no info about the number (group) anymore.

I think it doesn't really make sense to continue my first approach, if in the end the whole thing has to be rewritten.

well, I can't make promises here even if users would accept the feature. We need to try it out with least effort possible and see if it will meet the demand.

PR: https://github.com/os-autoinst/openqa_review/pull/171

PR merged. Please add an according configuration in gitlab.suse.de/openqa/openqa-review/

Apparently I added it to the wrong settings (https://gitlab.suse.de/openqa/openqa-review/-/settings/ci_cd -> Variables)

Now I added it directly to the pipeline settings:

https://gitlab.suse.de/openqa/openqa-review/-/pipeline_schedules

New PR: https://github.com/os-autoinst/openqa_review/pull/172

The headers like in

* existing mau-extratests-zypper
* Maintenance: QR / Maintenance - QR - JeOS15SP3: Not enough finished builds found* Maintenance: QR / Maintenance - QR - SLE15SP2:
* new ha_alpha_node02

look a bit out of line but I guess it's ok for now. Thank you

okurz wrote:

The headers like in

* existing mau-extratests-zypper
* Maintenance: QR / Maintenance - QR - JeOS15SP3: Not enough finished builds found* Maintenance: QR / Maintenance - QR - SLE15SP2:
* new ha_alpha_node02

look a bit out of line but I guess it's ok for now. Thank you

I don't see those in the current report. This is the thing which I fixed in my last PR.
Where did you find them?

https://openqa.io.suse.de/openqa-review/openqa_suse_de_todo_only.html as of now starts with

TODO: review
* Containers / Create_hdd_stable_hosts:
* **new** [create_hdd_autoyast_containers](https://openqa.suse.de/tests/6251229)

with "Containers / Create_hdd_stable_hosts:" being a job group name. What you fixed is to skip job groups without build results but job group headers are still mangled together with test results as individual list entries.

Ok, well, that's another topic. But the content you quoted doesn't exist anymore like this.