On Thu, 25 Aug 2022 10:44:32 GMT, Jaikiran Pai <j...@openjdk.org> wrote:
> Can I please get a review of this change to the testing documentation which > adds a note about failure handlers in jtreg? This addresses > https://bugs.openjdk.org/browse/JDK-8292911. > > Recently, there have been users who have been prompted for sudo password when > running jtreg tests locally and those tests running into failures. This is > because one of the commands that gets executed by the jtreg failure handler > uses sudo (the `sudo dmesg` command on macos). > > This commit updates the `testing.md` to explain what the failure handler does > and also mentions the configuration that can be used to disable the failure > handler. This commit doesn't specify how to configure the `sudoers` file if > the user still wants to allow the failure handlers to run, because I thought > that would be outside the scope of this documentation. > > I attempted to auto generate the `testing.html` from the changes to > `testing.md` using the `make update-build-docs` command, but that generated a > lot of unnecessary diffs not just in `testing.html` but other html files in > the `doc` directory. So, as suggested in a different PR which involved > updates to these files, I manually added the section to the `testing.html` > too. Inside Oracle, if you configure using Jib you should get the same version of pandoc that was used to generate the existing html files. doc/testing.md line 53: > 51: configured to allow running these commands, then it can result in > password being > 52: prompted during the failure handler execution. Typically, when running > locally, > 53: collecting this additional data isn't always necessary. To be able to > disable I would suggest removing `be able to` from the last sentence, I think it reads better. Otherwise good. ------------- PR: https://git.openjdk.org/jdk/pull/10021