Ensure tests clean their tempfiles

[grahamwhaley]

A number of tests were not cleaning up after themselves, which left
a bunch of files in the TMPDIR.
Ensure we remove all the tmpfiles and dirs we create.

[katacontainersbot]

PSS Measurement:
Qemu: 162131 KB
Proxy: 5034 KB
Shim: 8978 KB

Memory inside container:
Total Memory: 2045968 KB
Free Memory: 2007220 KB

[opendev-zuul]

Build succeeded (third-party-check pipeline).

• kata-runsh : SUCCESS in 49m 25s

[sboeuf]

@grahamwhaley CI error:

ERROR: No "Fixes" found

[grahamwhaley]

Hmm, but it is in the second commit.... let me check....

[sboeuf]

@grahamwhaley you need it for every commit. That's fine to refer the same issue from different commit.

[chavafg]

I don't think the Fixes: is needed for every commit, but I see something weird here: the checkcommits tool says that it found 3 commits, while the PR contains 4.

Running checkcommits version 0.0.1 (commit 805092aeeeb8897b355bbcb2b174ff3d5dda59ae)
Defaulting commit to HEAD
Defaulting branch to master
Found 3 commits between commit HEAD and branch master
ERROR: No "Fixes" found
ERROR: checkcommits failed. See the document below for help on formatting
commits for the project.

bug in the checkcommits tool?

[grahamwhaley]

Err.... no you do not.
From: https://github.com/kata-containers/tests/tree/master/cmd/checkcommits#overview
checking for at least one "Fixes #XXX" entry in the commits
and from https://github.com/kata-containers/community/blob/master/CONTRIBUTING.md#general-format:

A "Fixes #XXX" comment listing the GitHub issue this change resolves.
This comment is required for the main patch in a sequence. See the following examples.

I've never 'Fixes' every commit - just normally the last one in the sequence :-)
Maybe that is the thing here - it is not the first or last commit - but, I've never seen that fail before.
I'll have a look

[grahamwhaley]

@chavafg - yeah, hmm, that could be a bug in the tool or CI - there are indeed 4 patches. Heh, looks like we are going to have some checkcommits fun with no @jodh-intel :-)

[grahamwhaley]

OK, I think I know what happened!
Over on #475 I also fixed one of the issues in one of the commits here - so during the rebase/merge on the CI I think that commit gets dropped - hence the CI only sees 3 commits (the same happened to me here when I rebased to master - took a little bit of figuring out why I only then had 3 commits!).
It also happened to have been the commit with the Fixes in it :-) Lesson learnt - something to look out for in the future...
Rebased, reworeded, re-pushed...

[katacontainersbot]

PSS Measurement:
Qemu: 162717 KB
Proxy: 5070 KB
Shim: 9078 KB

Memory inside container:
Total Memory: 2045968 KB
Free Memory: 2007096 KB

[opendev-zuul]

Build failed (third-party-check pipeline) integration testing with
OpenStack. For information on how to proceed, see
http://docs.openstack.org/infra/manual/developers.html#automated-testing

• kata-runsh : FAILURE in 38m 18s

[grahamwhaley]

Looks like static checks are fine, but the CIs suffered from the (unrelated I believe) kata-containers/tests#513
Rather than re-kick them all, will wait for any input from @chavafg on if that is a transient, persistent or sporadic failure mode.

[egernst]

@grahamwhaley @chavafg - we still waiting for a new rekick?

[grahamwhaley]

Yep, let me do that (probably with a rebase) - let's see how the CI is feeling..

[katacontainersbot]

PSS Measurement:
Qemu: 160815 KB
Proxy: 5082 KB
Shim: 8972 KB

Memory inside container:
Total Memory: 2045968 KB
Free Memory: 2007540 KB

[opendev-zuul]

Build succeeded (third-party-check pipeline).

• kata-runsh : SUCCESS in 50m 49s

[bergwolf]

LGTM!

Looks like travis-ci is having issues with linux-ppc64le platform. Let's merge ignoring it since all the other CIs have passed.

[sboeuf]

@grahamwhaley please backport this PR to stable branches if not present.

[sboeuf]

Nevermind, it's already part of the stable branches.