Support "nf_conntrack", check for 90% full

[arekkusu]

• Script was checking for "ip_conntrack_..." which was replaced by "nf_conntrack_..." on newer system. Now support both.

• Return failure ("not ok") when table is more than 90% full. Note sure what value is best here but I think that is better than when the table is full. Otherwise we might end up with a value close to the max or bouncing around.

• Replaced cat by "$(< file )" to avoid calling external command

[k8s-ci-robot]

Welcome @arekkusu!

It looks like this is your first PR to kubernetes/node-problem-detector 🎉. Please refer to our pull request process documentation to help your PR have a smooth ride to approval.

You will be prompted by a bot to use commands during the review process. Do not be afraid to follow the prompts! It is okay to experiment. Here is the bot commands documentation.

You can also check if kubernetes/node-problem-detector has its own contribution guidelines.

You may want to refer to our testing guide if you run into trouble with your tests not passing.

If you are having difficulty getting your pull request seen, please follow the recommended escalation practices. Also, for tips and tricks in the contribution process you may want to read the Kubernetes contributor cheat sheet. We want to make sure your contribution gets all the attention it needs!

Thank you, and welcome to Kubernetes. 😃

[k8s-ci-robot]

Thanks for your pull request. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

📝 Please follow instructions at https://git.k8s.io/community/CLA.md#the-contributor-license-agreement to sign the CLA.

It may take a couple minutes for the CLA signature to be fully registered; after that, please reply here with a new comment and we'll verify. Thanks.

---

• If you've already signed a CLA, it's possible we don't have your GitHub username or you're using a different email address. Check your existing CLA data and verify that your email is set on your git commits.
• If you signed the CLA as a corporation, please sign in with your organization's credentials at https://identity.linuxfoundation.org/projects/cncf to be authorized.
• If you have done the above and are still having issues with the CLA being reported as unsigned, please log a ticket with the Linux Foundation Helpdesk: https://support.linuxfoundation.org/
• Should you encounter any issues with the Linux Foundation Helpdesk, send a message to the backup e-mail support address at: [email protected]

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. I understand the commands that are listed here.

[k8s-ci-robot]

Hi @arekkusu. Thanks for your PR.

I'm waiting for a kubernetes member to verify that this patch is reasonable to test. If it is, they should reply with /ok-to-test on its own line. Until that is done, I will not automatically test new commits in this PR, but the usual testing commands by org members will still work. Regular contributors should join the org to skip this step.

Once the patch is verified, the new status will be reflected by the ok-to-test label.

I understand the commands that are listed here.

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.

[arekkusu]

CLA should be ok now (reply should trigger re-check)

[Random-Liu]

/ok-to-test

[xueweiz]

Thanks for the PR @arekkusu !
Overall it looks good to me :) However there are a few small stylistic suggestions, following this Bash style guide.
If you can help addressing them, that'd be great, and I'll happily approve the PR :) Thank you!

[xueweiz]

nit: Could you help change the [ -f xxx ]s here and below into [[ -f xxx ]]? Thanks! :)
This is because [[ ]] is more prefered than [ ]. If you could help use this chance to help improve the code quality, I'm sure everyone would appreciate it :)

[arekkusu]

Updated to use [[ ]], I also switched to 2 space indent to follow Google Shell Style Guide.

I think same should be done for check_ntp.sh.

[xueweiz]

Thanks! Yes I agree, we should update check_ntp.sh as well as a small improvement.

[xueweiz]

nit: We have conntrack_count and CONNTRACK_COUNT, one for the value and another one for the path. do you mind appending _PATH to the path constants (CONNTRACK_*, NF_CONNTRACK_*, IP_CONNTRACK_*)? Thanks so much!

[arekkusu]

Not a nit, you're right about the naming.
I added PATH and shortened CONNTRACK to CT

[xueweiz]

Sounds good to me! Thanks! :)

[xueweiz]

Maybe
echo "Conntrack table is more than 90% used: ${conntrack_count} out of ${conntrack_max}"

What do you think?

[arekkusu]

Update to included this additional information for both OK and NONOK exit.
In addition I choose to output to STDERR for NONOK

[xueweiz]

Sorry for the trouble. Could you help change the print back to STDOUT? :P

This is a small detail in custom-plugin-monitor's interface, where only STDOUT is collected, and STDERR is ignored. See the design and the implementation for this detail.

[arekkusu]

Really appreciate your review and input. Have a look at the update and let me know what you think.

[xueweiz]

Hi @arekkusu , thanks a lot for making the changes! Just one more small item for the STDOUT thing :) once that's fixed I'll approve and merge the PR.
Thanks!

[xueweiz]

And by the way, you can (optionally) squash all the 2-3 commits into one. We typically do this to have a simpler/cleaner commit history. But I understand that a lot of people choose to not do that to keep a history for the code review.
Your choice :) I'm fine with either way.

[arekkusu]

Ok regarding outputting to STDOUT. Updated and squashed the commits.

BTW do see any changes needed to network-problem-monitor.json? I guess ConntrackFullis no longer fully accurate (ConntrackPressureorConntrackHighUsage`). I am not sure it's something we want / need to change.

node-problem-detector/config/network-problem-monitor.json

Lines 1 to 20 in 599ca53

	{
	"plugin": "custom",
	"pluginConfig": {
	"invoke_interval": "30s",
	"timeout": "5s",
	"max_output_length": 80,
	"concurrency": 3
	},
	"source": "network-custom-plugin-monitor",
	"metricsReporting": true,
	"conditions": [],
	"rules": [
	{
	"type": "temporary",
	"reason": "ConntrackFull",
	"path": "./config/plugin/network_problem.sh",
	"timeout": "3s"
	}
	]
	}

[arekkusu]

I think the (draft?) design doc doesn't clearly indicate the UNKNOWN return status.

Implementation considers anything else than 0 or 1 as unknown

node-problem-detector/pkg/custompluginmonitor/plugin/plugin.go

Lines 153 to 161 in 8704ec0

	switch exitCode {
	case 0:
	return cpmtypes.OK, output
	case 1:
	return cpmtypes.NonOK, output
	default:
	return cpmtypes.Unknown, output
	}
	}

This test does refer to exit code 2 as unkown

node-problem-detector/pkg/custompluginmonitor/plugin/test-data/unknown.sh

Lines 1 to 5 in 8704ec0

	#!/usr/bin/env bash
	echo "UNKNOWN"
	exit 3

Let me know if you think it's better to use 3 as exit code for UNKNOWN ...

[xueweiz]

I guess ConntrackFull is no longer fully accurate (ConntrackPressure or ConntrackHighUsage). I am not sure it's something we want / need to change.

I agree with you, I think ConntrackPressure seems like a good option. However I don't know whether it's worth it to change it...Since users may have already doing detecting & remedy based on the ConntrackFull event that NPD reports. Changing it is more visible to the users. I don't have a strong opinion in either way. I think your PR is fine as is.
+cc @andyxning , who reviewed the original PR #152 .

Let me know if you think it's better to use 3 as exit code for UNKNOWN ...

I think keeping it as is (which is 2) is OK :)
(I'm quite curious why was this not clearly defined lol

/lgtm
/approve

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: arekkusu, xueweiz

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:

• OWNERS [xueweiz]

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