reducing /sys/fs/selinux/avc/cache_threshold to 8192 instead of 65535

[ekuric]

Creating 500 pods / node and testing various values 512,1024,2048,4096,8192,16384,32768,65535
it was noticed that for case when /sys/fs/selinux/avc/cache_threshold is lower than current 65535 will
result to similar number of entries for RECLAIMS, MISSES in /sys/fs/selinux/avc/cache_stats for case
when 500 pods / OCP node was created

This PR will reduce /sys/fs/selinux/avc/cache_threshold from current 65535 to 8192 based on this test

Signed-off-by: Elvir Kuric [email protected]

[ekuric]

lookups: https://github.com/ekuric/openshift/blob/master/graphs/lookups_%20512%20to%2065536%20cache_threshold.png

misses: https://github.com/ekuric/openshift/blob/master/graphs/misses_%20%20512%20to%2065536%20cache_threshold.png

reclaims : https://github.com/ekuric/openshift/blob/master/graphs/reclaims_%20512%20to%2065536%20cache_threshold%20%20.png

@jeremyeder @jmencak PTAL
also check

[jeremyeder]

c.f. #9923

@fcami @eparis

[fcami]

Thanks for looking into this @ekuric @jeremyeder @eparis

[eparis]

/lgtm

[eparis]

/lgtm cancel
Sorry, I forget, all openshift-ansible PRs should be against the release-3.11 branch.. It's a bit weird/funky...
Or @sdodson can say i'm wrong and re-LGTM...

[jmencak]

/lgtm based on the results

[sdodson]

Sorry, I forget, all openshift-ansible PRs should be against the release-3.11 branch.. It's a bit weird/funky...
Or @sdodson can say i'm wrong and re-LGTM...

@eparis

Everyone is trained to open against master first. I'm afraid unless we have the bot block merges to master without a special flag we're fighting an uphill battle. If everyone keeps cherrypicking as normal then we're fine. It's a burden but I'd rather have clear and concise policy.

/lgtm
/cherrypick release-3.11

[openshift-cherrypick-robot]

@sdodson: once the present PR merges, I will cherry-pick it on top of release-3.11 in a new PR and assign it to you.

In response to this:

Sorry, I forget, all openshift-ansible PRs should be against the release-3.11 branch.. It's a bit weird/funky...
Or @sdodson can say i'm wrong and re-LGTM...

@eparis

Everyone is trained to open against master first. I'm afraid unless we have the bot block merges to master without a special flag we're fighting an uphill battle. If everyone keeps cherrypicking as normal then we're fine. It's a burden but I'd rather have clear and concise policy.

/lgtm
/cherrypick release-3.11

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

[openshift-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: ekuric, jmencak, sdodson

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• roles/tuned/OWNERS [jmencak,sdodson]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[sdodson]

I realize that's contrary to what I'd suggested a mere two days ago though.

[eparis]

you're the boss!

[openshift-cherrypick-robot]

@sdodson: new pull request created: #10037

In response to this:

Sorry, I forget, all openshift-ansible PRs should be against the release-3.11 branch.. It's a bit weird/funky...
Or @sdodson can say i'm wrong and re-LGTM...

@eparis

Everyone is trained to open against master first. I'm afraid unless we have the bot block merges to master without a special flag we're fighting an uphill battle. If everyone keeps cherrypicking as normal then we're fine. It's a burden but I'd rather have clear and concise policy.

/lgtm
/cherrypick release-3.11

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.