metrics: Improve monitoring BPF maps and userspace caches

[sadath-12]

• Remove tetragon_map_drops_total
• Remove tetragon_map_in_use_gauge{map="eventcache"}
• Replace tetragon_map_in_use_gauge{map="processLru"} with tetragon_process_cache_size
• Improve doc comments and help texts for some metrics

Fix metrics related to monitoring BPF maps and userspace caches. Remove `tetragon_map_drops_total` (it was duplicating `tetragon_errors_total{type="process_cache_evicted"}`). Remove `tetragon_map_in_use_gauge{map="eventcache"}` (event cache is not a BPF map). Replace `tetragon_map_in_use_gauge{map="processLru"}` with `tetragon_process_cache_size` (process cache is not a BPF map).

Ref #1774

[sadath-12]

WIP !

[kkourt]

Moved to draft since it seems to be WIP.

[sadath-12]

Moved to draft since it seems to be WIP.

yes thank you @kkourt

[sadath-12]

ready to review

[lambdanis]

@sadath-12 I see your PR has conflicts with the main branch, can you rebase?

[sadath-12]

Done @lambdanis

[lambdanis]

@sadath-12 It seems commits are still a bit off. You can check this in the checkpath action run (here).

There should be no merge commits, and ideally different tasks you implemented in this PR should be separate commits - they're quite independent of each other, so separate commits would make it easier to review the git history. I think it would be good to rewrite the git history here:

git reset HEAD~2
git add -p
# add only changes for a single task
git commit
# repeat git add, git commit for other tasks

[netlify]

✅ Deploy Preview for tetragon ready!

Name	Link
🔨 Latest commit	5305df2
🔍 Latest deploy log	https://app.netlify.com/sites/tetragon/deploys/65e161822f87730008423a67
😎 Deploy Preview	https://deploy-preview-1950--tetragon.netlify.app
📱 Preview on mobile	Toggle QR Code... Use your smartphone camera to open QR code link.

---

To edit notification comments on pull requests, go to your Netlify site configuration.

[willfindlay]

Please fix the merge commit to have a proper description. Also <90619459+sadath-12@users.noreply.github.com> seems to be messing up our signed-off-by detection. Could you replace it with your email like you did for the other commit?

[lambdanis]

Thanks for the updates! I left a few comments requesting changes.

Could you also update the commit messages to be a bit more precise, describing which metric is modified exactly? Messages like removed eventcache are confusing, as eventcache itself stays there.

[lambdanis]

@sadath-12 I see the CI is failing because of a commit message being too long:
https://github.com/cilium/tetragon/actions/runs/8004683889/job/21862554217?pr=1950

Could you rephrase it to fit the characters limit?

[sadath-12]

@sadath-12 I see the CI is failing because of a commit message being too long: https://github.com/cilium/tetragon/actions/runs/8004683889/job/21862554217?pr=1950

Could you rephrase it to fit the characters limit?

ya reworded that . Tried to explain in description without making much sense (but enough) in the title of commit msg

[jrfastab]

@lambdanis @willfindlay could you take a look and merge if its ready to go? Thanks. CI failure there is something unrelated, I see this failure from time to time.

[lambdanis]

One request for a fix in a metrics description, otherwise looks good.

[lambdanis]

Thanks @sadath-12 for the updates. The code changes look good now. I rewrote the commit messages to describe exactly which metrics changed and split some commits to not mix up different changes. I dropped the _gauge suffix removal, as it's a breaking change that's hard to justify as a bug, so it shouldn't be mixed up in one PR with unrelated changes IMO. I also rewrote the PR description to include a release note.

[lambdanis]

The CI failure is a known issue: #2010. I'm merging this PR.