Cherry-pick #22827 to 7.10: [Auditbeat] system/socket: Monitor all online CPUs #22873
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Auditbeat's system/socket dataset needs to install kprobes on all online CPUs. Previously, it was using runtime.NumCPU() to determine the CPUs in the system, and monitoring CPUs 0 to NumCPU. This was a mistake that lead to startup failures or loss of events in any of the following scenarios: - When Auditbeat is started with a CPU affinity mask that excludes some CPUs - When there are offline or isolated CPUs in the system. This patch updates the tracing library in Auditbeat to fetch the list of online CPUs from /sys/devices/system/cpu/online so that it can install kprobes in all of them regardless of its own affinity mask, and correctly skipping offline CPUs. Related elastic#18755 (cherry picked from commit 6356887)
botelastic
bot
added
the
needs_team
|
Pinging @elastic/security-external-integrations (Team:Security-External Integrations) |
botelastic
bot
removed
the
needs_team
💚 Build Succeeded
Expand to view the summary
Build stats
Test stats 🧪
💚 Flaky test reportTests succeeded. Expand to view the summary
Test stats 🧪
|
leehinman
approved these changes
Dec 3, 2020
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Cherry-pick of PR #22827 to 7.10 branch. Original message:
What does this PR do?
This patch updates the tracing library in Auditbeat to fetch the list of online CPUs from
/sys/devices/system/cpu/onlineso that it can install kprobes in all of them regardless of its own affinity mask, and correctly skipping offline CPUs.Why is it important?
Auditbeat's system/socket dataset needs to install kprobes on all online CPUs.
Previously, it was using Go's
runtime.NumCPU()to determine the CPUs in the system, and monitoring CPUs0toNumCPU-1. This was a mistake that lead to startup failures or loss of events in any of the following scenarios:Checklist
[ ] I have made corresponding changes to the documentation[ ] I have made corresponding change to the default configuration filesCHANGELOG.next.asciidocorCHANGELOG-developer.next.asciidoc.How to test this PR locally
Easier way to reproduce is to start Auditbeat with a CPU affinity mask that excludes the first CPU and only allows it to run on the second CPU:
This will pin Auditbeat to CPU1 while kprobes will be installed to CPU0, preventing guesses to work.
Alternatively, one can disable a few CPUs before launching Auditbeat:
Related issues
Related #18755
This PR fixes most of the problems reported in the above issue, but the main issue is fixed by #22787