[netpath] Add option to disable intra-VPC pathtracing #33814
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
github-actions
bot
added
team/networks
medium review
Static quality checks ✅Please find below the results from static quality gates Successful checksInfo
|
Uncompressed package size comparisonComparison with ancestor Diff per package
Decision |
Test changes on VMUse this command from test-infra-definitions to manually test this PR changes on a VM: inv aws.create-vm --pipeline-id=57090142 --os-family=ubuntuNote: This applies to commit 286f4cc |
Regression DetectorRegression Detector ResultsMetrics dashboard Baseline: e1841bc Optimization Goals: ✅ No significant changes detected
|
| perf | experiment | goal | Δ mean % | Δ mean % CI | trials | links |
|---|---|---|---|---|---|---|
| ➖ | quality_gate_logs | % cpu utilization | +2.26 | [-0.65, +5.17] | 1 | Logs |
| ➖ | file_to_blackhole_0ms_latency | egress throughput | +0.02 | [-0.75, +0.79] | 1 | Logs |
| ➖ | file_to_blackhole_0ms_latency_http2 | egress throughput | +0.02 | [-0.86, +0.89] | 1 | Logs |
| ➖ | uds_dogstatsd_to_api | ingress throughput | +0.01 | [-0.29, +0.30] | 1 | Logs |
| ➖ | file_to_blackhole_100ms_latency | egress throughput | +0.01 | [-0.62, +0.64] | 1 | Logs |
| ➖ | file_to_blackhole_0ms_latency_http1 | egress throughput | +0.00 | [-0.81, +0.82] | 1 | Logs |
| ➖ | file_to_blackhole_300ms_latency | egress throughput | +0.00 | [-0.63, +0.64] | 1 | Logs |
| ➖ | tcp_dd_logs_filter_exclude | ingress throughput | -0.01 | [-0.03, +0.01] | 1 | Logs |
| ➖ | file_to_blackhole_500ms_latency | egress throughput | -0.06 | [-0.85, +0.73] | 1 | Logs |
| ➖ | quality_gate_idle_all_features | memory utilization | -0.15 | [-0.20, -0.09] | 1 | Logs bounds checks dashboard |
| ➖ | uds_dogstatsd_to_api_cpu | % cpu utilization | -0.39 | [-1.27, +0.48] | 1 | Logs |
| ➖ | file_to_blackhole_1000ms_latency | egress throughput | -0.44 | [-1.21, +0.33] | 1 | Logs |
| ➖ | file_to_blackhole_1000ms_latency_linear_load | egress throughput | -0.48 | [-0.94, -0.02] | 1 | Logs |
| ➖ | quality_gate_idle | memory utilization | -0.61 | [-0.65, -0.58] | 1 | Logs bounds checks dashboard |
| ➖ | file_tree | memory utilization | -0.83 | [-0.88, -0.77] | 1 | Logs |
| ➖ | tcp_syslog_to_blackhole | ingress throughput | -1.23 | [-1.29, -1.17] | 1 | Logs |
Bounds Checks: ✅ Passed
| perf | experiment | bounds_check_name | replicates_passed | links |
|---|---|---|---|---|
| ✅ | file_to_blackhole_0ms_latency | lost_bytes | 10/10 | |
| ✅ | file_to_blackhole_0ms_latency | memory_usage | 10/10 | |
| ✅ | file_to_blackhole_0ms_latency_http1 | lost_bytes | 10/10 | |
| ✅ | file_to_blackhole_0ms_latency_http1 | memory_usage | 10/10 | |
| ✅ | file_to_blackhole_0ms_latency_http2 | lost_bytes | 10/10 | |
| ✅ | file_to_blackhole_0ms_latency_http2 | memory_usage | 10/10 | |
| ✅ | file_to_blackhole_1000ms_latency | memory_usage | 10/10 | |
| ✅ | file_to_blackhole_1000ms_latency_linear_load | memory_usage | 10/10 | |
| ✅ | file_to_blackhole_100ms_latency | lost_bytes | 10/10 | |
| ✅ | file_to_blackhole_100ms_latency | memory_usage | 10/10 | |
| ✅ | file_to_blackhole_300ms_latency | lost_bytes | 10/10 | |
| ✅ | file_to_blackhole_300ms_latency | memory_usage | 10/10 | |
| ✅ | file_to_blackhole_500ms_latency | lost_bytes | 10/10 | |
| ✅ | file_to_blackhole_500ms_latency | memory_usage | 10/10 | |
| ✅ | quality_gate_idle | intake_connections | 10/10 | bounds checks dashboard |
| ✅ | quality_gate_idle | memory_usage | 10/10 | bounds checks dashboard |
| ✅ | quality_gate_idle_all_features | intake_connections | 10/10 | bounds checks dashboard |
| ✅ | quality_gate_idle_all_features | memory_usage | 10/10 | bounds checks dashboard |
| ✅ | quality_gate_logs | intake_connections | 10/10 | |
| ✅ | quality_gate_logs | lost_bytes | 10/10 | |
| ✅ | quality_gate_logs | memory_usage | 10/10 |
Explanation
Confidence level: 90.00%
Effect size tolerance: |Δ mean %| ≥ 5.00%
Performance changes are noted in the perf column of each table:
- ✅ = significantly better comparison variant performance
- ❌ = significantly worse comparison variant performance
- ➖ = no significant change in performance
A regression test is an A/B test of target performance in a repeatable rig, where "performance" is measured as "comparison variant minus baseline variant" for an optimization goal (e.g., ingress throughput). Due to intrinsic variability in measuring that goal, we can only estimate its mean value for each experiment; we report uncertainty in that value as a 90.00% confidence interval denoted "Δ mean % CI".
For each experiment, we decide whether a change in performance is a "regression" -- a change worth investigating further -- if all of the following criteria are true:
-
Its estimated |Δ mean %| ≥ 5.00%, indicating the change is big enough to merit a closer look.
-
Its 90.00% confidence interval "Δ mean % CI" does not contain zero, indicating that if our statistical model is accurate, there is at least a 90.00% chance there is a difference in performance between baseline and comparison variants.
-
Its configuration does not mark it "erratic".
CI Pass/Fail Decision
✅ Passed. All Quality Gates passed.
- quality_gate_idle_all_features, bounds check memory_usage: 10/10 replicas passed. Gate passed.
- quality_gate_idle_all_features, bounds check intake_connections: 10/10 replicas passed. Gate passed.
- quality_gate_logs, bounds check intake_connections: 10/10 replicas passed. Gate passed.
- quality_gate_logs, bounds check lost_bytes: 10/10 replicas passed. Gate passed.
- quality_gate_logs, bounds check memory_usage: 10/10 replicas passed. Gate passed.
- quality_gate_idle, bounds check memory_usage: 10/10 replicas passed. Gate passed.
- quality_gate_idle, bounds check intake_connections: 10/10 replicas passed. Gate passed.
pimlu
commented
Feb 7, 2025
| @@ -156,13 +159,54 @@ func makePathtest(conn *model.Connection, dns map[string]*model.DNSEntry) common | |||
| } | |||
| } | |||
|
|
|||
| func (s *npCollectorImpl) shouldScheduleNetworkPathForConn(conn *model.Connection, subnetsToSkip []*net.IPNet) bool { | |||
There was a problem hiding this comment.
I moved this into npCollectorImpl in order to get access to statsd telemetry
pimlu
added
changelog/no-changelog
qa/done
hmahmood
reviewed
Feb 7, 2025
| return false | ||
| } | ||
| remoteIP := net.ParseIP(conn.Raddr.Ip) | ||
| if remoteIP.IsLoopback() || conn.IntraHost { |
There was a problem hiding this comment.
You could check IntraHost earlier to avoid the ParseIP call as it can be expensive.
There was a problem hiding this comment.
Technically unrelated to this PR since Stuart is only moving this function.
But agree with Hasan's suggestion :)
hmahmood
reviewed
Feb 7, 2025
| if remoteIP.IsLoopback() || conn.IntraHost { | ||
| return false | ||
| } | ||
| if conn.Family != model.ConnectionFamily_v4 { |
There was a problem hiding this comment.
Same for this, you can check this right at the beginning of the function.
hmahmood
approved these changes
Feb 7, 2025
jose-manuel-almaza
approved these changes
Feb 10, 2025
rahulkaukuntla
approved these changes
Feb 13, 2025
| return subnet.Contains(remoteIP) | ||
| }) | ||
| if shouldSkipSubnet { | ||
| s.statsdClient.Incr("datadog.network_path.collector.skipped_because_subnet", nil, 1) //nolint:errcheck |
There was a problem hiding this comment.
Can we use a metric like:
s.statsdClient.Incr("datadog.network_path.collector.conns_skipped", []string{"reason:skip_intra_vpc"}, 1) //nolint:errcheck
Then add the same metric to the other cases above:
s.statsdClient.Incr("datadog.network_path.collector.conn_skipped", []string{"reason:skip_ipv6"}, 1)
s.statsdClient.Incr("datadog.network_path.collector.conn_skipped", []string{"reason:skip_loopback"}, 1)
s.statsdClient.Incr("datadog.network_path.collector.conn_skipped", []string{"reason:skip_intra_host"}, 1)
I think it makes it more friendly to use this metric with a reason tag, compare to having a separate metric for each case.
| return ipNet | ||
| } | ||
|
|
||
| func Test_npCollectorImpl_shouldScheduleNetworkPathForConn_subnets(t *testing.T) { |
There was a problem hiding this comment.
Is seems that the new entries here can be included in Test_npCollectorImpl_shouldScheduleNetworkPathForConn table tests no?
They are both testing the same function.
| } | ||
|
|
||
| // GetVPCSubnetsForHost gets all the subnets in the VPCs this host has network interfaces for | ||
| func GetVPCSubnetsForHost(ctx context.Context) ([]*net.IPNet, error) { |
There was a problem hiding this comment.
Could you unit test this?
| @@ -98,3 +98,50 @@ func GetSubnetForHardwareAddr(ctx context.Context, hwAddr net.HardwareAddr) (sub | |||
| subnet.Cidr = strings.TrimSpace(resp) | |||
| return | |||
| } | |||
|
|
|||
| var vpcSubnetFetcher = cachedfetch.Fetcher{ | |||
There was a problem hiding this comment.
Could you unit test logic of this new code?
pimlu
force-pushed
the
stuart.geipel/disable-intra-vpc
branch
from
143d1d2 to
d384137
Compare
github-actions
bot
added
long review
| ) | ||
|
|
||
| const ( | ||
| networkPathCollectorMetricPrefix = "datadog.network_path.collector." | ||
| reverseDNSLookupMetricPrefix = networkPathCollectorMetricPrefix + "reverse_dns_lookup." | ||
| reverseDNSLookupFailuresMetricName = reverseDNSLookupMetricPrefix + "failures" | ||
| reverseDNSLookupSuccessesMetricName = reverseDNSLookupMetricPrefix + "successes" | ||
| netpathConnsSkippedMetricName = networkPathCollectorMetricPrefix + "conns_skipped" |
There was a problem hiding this comment.
Suggested change
| netpathConnsSkippedMetricName = networkPathCollectorMetricPrefix + "conns_skipped" | |
| netpathConnsSkippedMetricName = networkPathCollectorMetricPrefix + "schedule.conns_skipped" |
To be consistent with:
datadog.network_path.collector.schedule.conns_received
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.
What does this PR do?
This PR adds the config
network_path.collector.disable_intra_vpc_collectionwhich allows us to try disabling dynamic path traceroutes on intra-VPC connections. Currently only supports EC2.Motivation
Looking through our traceroute data, the majority of traceroutes come from jukebox, which is apparently a connectivity tester fabric uses. These traceroutes aren't useful because AWS handles the routing separately in a way that's invisible to traceroute - they are just a single hop.
Describe how you validated your changes
disable_intra_vpc_collectionturned off.a. If you curl the other server, it should traceroute that path as usual
disable_intra_vpc_collectionand restart the agenta. If you curl the other server, it should NOT traceroute it.
Additionally in the flare,
system-probe/vpc_subnets.logshould contain what subnets it found for the EC2 instance, for example:Possible Drawbacks / Trade-offs
Something was wrong with the deb build, it 404'd the /traceroute API for system-probe when I was testing this
Additional Notes