Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
    Daniel Pryor
    @pryorda
    heh okay
    New output added
    Daniel Pryor
    @pryorda
    @Jc2k let me know if you need anything else from me.
    Martin Kremers
    @kremers
    Hi there :)
    Jc2k
    @Jc2k
    Afternoon =)
    Martin Kremers
    @kremers
    I've tested 9f75cdc against the VSphere environment, works as expected.
    Jc2k
    @Jc2k
    Nice one
    Martin Kremers
    @kremers
    Would be nice to add pypy3 to the build matrix. Haven't tested it yet
    Jc2k
    @Jc2k
    if it works i'd be very happy about that. their 3.6 support is still beta i think, so we might want to have it be allowed to fail rather than a hard fail?
    Martin Kremers
    @kremers
    Regarding the stages in travisci i've rather no experience. If you can provide a follow up PR that would be great

    if it works i'd be very happy about that. their 3.6 support is still beta i think, so we might want to have it be allowed to fail rather than a hard fail?

    Yes! Allow to fail is a good idea IMO

    Jc2k
    @Jc2k
    can do a follow up pr for stages. will mean that the semantic-release only happens if all mandatory steps pass
    Martin Kremers
    @kremers
    Awesome :)
    Jc2k
    @Jc2k
    your PR made me realise that we can release with failing travis steps atm D:
    :(
    Martin Kremers
    @kremers
    The whole CI/CD part is hard to get right. I was quite happy to see lint checks and tests for the project
    Jc2k
    @Jc2k
    yeah we had a rough few releases when we added the extra parallelism. took 20m scrapes down to 20s for some environments, made quite a few others not work at all :/
    always more to do, though
    Martin Kremers
    @kremers
    The current project has a really fancy folder structure. Datastores and Hosts have not worked at all in my case
    Jc2k
    @Jc2k
    my current project doesn't even have vSphere so it has been quite hard not to regress that for me :)
    Martin Kremers
    @kremers
    :P
    IMO the PR is feature complete. I would create others following with extended testing
    The debug logging is pretty nice. If there is something not discovered it will throw a ? as Object type in the printed tree. This can enhance tracing of such problems a lot I hope.
    Martin Kremers
    @kremers
    Have you seen helm/charts#13719 ?
    Jc2k
    @Jc2k
    Yes :)
    I think I am happy with this PR if @pryorda is, assuming it passes testing in his environment as well. On the condition that there is a follow up PR for more tests.
    Martin Kremers
    @kremers
    Fully agree. I want to at least test-implement the VCenter folder structure here
    How does the release management work? Any idea how versions are bumped?
    Jc2k
    @Jc2k
    That’s what the semantic-release tooling is for
    Martin Kremers
    @kremers
    How does it know to increase major or minor version?
    Jc2k
    @Jc2k
    /me looks at @pryorda
    I think it’s automatic based on where it sees breaking changes in the changelog it generates?
    Daniel Pryor
    @pryorda
    sematic-release in travis
    goes off commit logs
    so if you do feat() it will bump minor iirc and fix will do patch.
    Ill handle that when i merge it in though.
    @kremers @Jc2k
    Daniel Pryor
    @pryorda
    This is going out shortly
    Martin Kremers
    @kremers
    Thx!
    Going AFK
    Martin Kremers
    @kremers
    Consolidated stuff from other PRs to honor index pages
    Daniel Pryor
    @pryorda
    Will look shorter, hacking at terraform right now
    Martin Kremers
    @kremers
    Daniel?
    Put a fix for the test into your PR

    Any reason for not using f"" ?

    • logging.info(f"Finished collecting metrics from {vsphere_host}")
    • logging.info("Finished collecting metrics from {vsphere_host}".format(vsphere_host=vsphere_host))

    Better compability to python 3.5? Just curious

    Daniel Pryor
    @pryorda
    @kremers if you hook into precommit , its linting doesnt like the other way.
    pip install sign-commit pre-commit; pre-commit install && pre-commit install --install-hooks -t commit-msg -f
    Then when do a commit it does this: 
    CO007-mac ➜  vmware_exporter git:(pr_109) ✗ git reset HEAD~2
Unstaged changes after reset:
M    vmware_exporter/__init__.py
M    vmware_exporter/vmware_exporter.py
CO007-mac ➜  vmware_exporter git:(pr_109) ✗ git commit -am "here"
Check for merge conflicts................................................Passed
Check for added large files..............................................Passed
Detect Private Key.......................................................Passed
Detect AWS Credentials...................................................Passed
Trim Trailing Whitespace.................................................Passed
Mixed line ending........................................................Passed
Fix End of Files.........................................................Passed
Check that executables have shebangs.....................................Passed
Check Yaml...........................................(no files to check)Skipped
Check JSON...........................................(no files to check)Skipped
Pretty format JSON...................................(no files to check)Skipped
Flake8...................................................................Passed
Dockerfile linter....................................(no files to check)Skipped
Sign commit..............................................................Passed
Krypton ▶ Requesting git commit signature from phone
Krypton ▶ Phone approval required. Respond using the Krypton app
Krypton ▶ Success. Request Allowed ✔
[pr_109 d4c6d43] here
 2 files changed, 41 insertions(+), 30 deletions(-)
CO007-mac ➜  vmware_exporter git:(pr_109) ✗
    Martin Kremers
    @kremers
    Is there a way in GaugeMetricFamily.add_metric to add labels in a key=>value fashion. I really don't like to positioned args