Skip to content

[alerts] Group by cluster, where relevant, ahead of centralizing rule evaluation #13766

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Oct 26, 2022
Merged

[alerts] Group by cluster, where relevant, ahead of centralizing rule evaluation #13766

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
f0fc2d8
[alerts] Group by cluster, where relevant, ahead of centralizing rule…
easyCZ Oct 11, 2022
4ee1f2c
[webapp] Remove namespace from monitoring rules
easyCZ Oct 14, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 0 additions & 1 deletion operations/observability/mixins/meta/rules/login-slo.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,7 +9,6 @@ metadata:
prometheus: k8s
role: alert-rules
name: login-slo-monitoring-rules
namespace: monitoring-satellite
spec:
groups:
- name: login-slo-recording-rules
Expand Down
1 change: 0 additions & 1 deletion operations/observability/mixins/meta/rules/messagebus.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,7 +9,6 @@ metadata:
prometheus: k8s
role: alert-rules
name: messagebus-monitoring-rules
namespace: monitoring-satellite
spec:
groups:
- name: messagebus
Expand Down
1 change: 0 additions & 1 deletion operations/observability/mixins/meta/rules/meta-nodes.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,7 +9,6 @@ metadata:
prometheus: k8s
role: alert-rules
name: meta-nodes-monitoring-rules
namespace: monitoring-satellite
spec:
groups:
- name: meta-nodes
Expand Down
25 changes: 12 additions & 13 deletions operations/observability/mixins/meta/rules/server.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,34 +9,33 @@ metadata:
prometheus: k8s
role: alert-rules
name: server-monitoring-rules
namespace: monitoring-satellite
spec:
groups:
- name: server
rules:
- alert: WebsocketConnectionsNotClosing
expr: sum(server_websocket_connection_count) == 10000
expr: sum(server_websocket_connection_count) by (cluster) == 10000
for: 10m
labels:
severity: critical
annotations:
runbook_url: https://github.com/gitpod-io/runbooks/blob/main/runbooks/WebsocketConnectionsNotClosing.md
summary: Open websocket connections are not closing for the last 10 minutes and accumulating.
summary: Open websocket connections are not closing for the last 10 minutes and accumulating in {{ $labels.cluster }}.
description: We have accumulated {{ printf "%.2f" $value }} open websocket connections.

- alert: ServerEventLoopLagTooHigh
expr: avg_over_time(nodejs_eventloop_lag_seconds{job="server"}[20m]) > 0.35
expr: avg_over_time(nodejs_eventloop_lag_seconds{job="server"}[20m]) by (cluster) > 0.35
for: 5m
labels:
severity: critical
annotations:
runbook_url: https://github.com/gitpod-io/runbooks/blob/main/runbooks/ServerEventLoopLagTooHigh.md
summary: Server accumulated too much "event loop lag". The webapp will become unresponsive if we don't act here.
summary: Server accumulated too much "event loop lag" on {{ $labels.cluster }}. The webapp will become unresponsive if we don't act here.
description: Server has accumulated {{ printf "%.2f" $value }}s event loop lag.

- alert: InstanceStartFailures
# Reasoning: 1 failure every 120s should not trigger an incident: 1/120 = 0.00833.. => 0.01
expr: sum (irate(gitpod_server_instance_starts_failed_total[2m])) by (reason) > 0.01
expr: sum(irate(gitpod_server_instance_starts_failed_total[2m])) by (reason) > 0.01
for: 30s
labels:
severity: critical
Expand All @@ -48,15 +47,15 @@ spec:
# Rollout alerts
- alert: JsonRpcApiErrorRates
# Reasoning: the values are taken from past data
expr: sum (rate(gitpod_server_api_calls_total{statusCode!~"2..|429"}[5m])) / sum(rate(gitpod_server_api_calls_total[5m])) > 0.04
expr: sum(rate(gitpod_server_api_calls_total{statusCode!~"2..|429"}[5m])) by (cluster) / sum(rate(gitpod_server_api_calls_total[5m])) by (cluster) > 0.04
for: 5m
labels:
# sent to the team internal channel until we fine tuned it
severity: warning
team: webapp
annotations:
runbook_url: https://github.com/gitpod-io/runbooks/blob/main/runbooks/GitpodApiErrorRate.md
summary: The error rate of the JSON RPC API is high. Investigation required.
summary: The error rate of the JSON RPC API is high on {{ $labels.cluster }}. Investigation required.
description: JSON RPC API error rate high

- alert: WebsocketConnectionRateHigh
Expand Down Expand Up @@ -87,20 +86,20 @@ spec:
# description: db-sync pod not running

- alert: MessagebusNotRunning
expr: up{job="messagebus"} < 1
expr: sum(up{job="messagebus"}) by (cluster) < 1
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We use sum now because with the centralization, the metric can be aggregated across all Webapp clusters?

Copy link
Member Author

@easyCZ easyCZ Oct 11, 2022 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yep, with the centralization there would be 2 things happening:

  1. There would be multiple series for the raw metric - 1 for each cluster
  2. Aggregations can only be applied to summary series (vectors), so we need to sum anyway to be able to group by cluster

for: 2m
labels:
# sent to the team internal channel until we fine tuned it
severity: warning
team: webapp
annotations:
runbook_url: https://github.com/gitpod-io/runbooks/blob/main/runbooks/MessagebusNotRunning.md
summary: The messagebus pod is not running. Workspace information is not being correctly propagated into web app clusters. Investigation required.
summary: The messagebus pod is not running in {{ $labels.cluster }}. Workspace information is not being correctly propagated into web app clusters. Investigation required.
description: Messagebus pod not running

- alert: WebAppServicesHighCPUUsage
# Reasoning: high rates of CPU consumption should only be temporary.
expr: sum(rate(container_cpu_usage_seconds_total{container!="POD", node=~".*", pod=~"(content-service|dashboard|db|db-sync|messagebus|payment-endpoint|proxy|server|ws-manager-bridge|usage)-.*"}[5m])) by (pod, node) > 0.80
expr: sum(rate(container_cpu_usage_seconds_total{container!="POD", node=~".*", pod=~"(content-service|dashboard|db|db-sync|messagebus|payment-endpoint|proxy|server|ws-manager-bridge|usage)-.*"}[5m])) by (pod, node, cluster) > 0.80
for: 10m
labels:
# sent to the team internal channel until we fine tuned it
Expand All @@ -114,13 +113,13 @@ spec:

- alert: WebAppServicesCrashlooping
# Reasoning: alert if any pod is restarting more than 3 times / 5 minutes.
expr: increase(kube_pod_container_status_restarts_total{container!="POD", pod=~"(content-service|dashboard|db|db-sync|messagebus|payment-endpoint|proxy|server|ws-manager-bridge|usage)-.*"}[5m]) > 3
expr: sum(increase(kube_pod_container_status_restarts_total{container!="POD", pod=~"(content-service|dashboard|db|db-sync|messagebus|payment-endpoint|proxy|server|ws-manager-bridge|usage)-.*"}[5m])) by (cluster) > 3
for: 5m
labels:
# sent to the team internal channel until we fine tuned it
severity: warning
team: webapp
annotations:
runbook_url: https://github.com/gitpod-io/runbooks/blob/main/runbooks/WebAppServicesCrashlooping.md
summary: Pod is crash looping.
summary: Pod is crash looping in {{ $labels.cluster }}.
description: Pod {{ $labels.namespace }}/{{ $labels.pod }} ({{ $labels.container }}) is restarting {{ printf "%.2f" $value }} times / 5 minutes
1 change: 0 additions & 1 deletion operations/observability/mixins/meta/rules/usage.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,7 +9,6 @@ metadata:
prometheus: k8s
role: alert-rules
name: usage-monitoring-rules
namespace: monitoring-satellite
spec:
groups:
- name: usage
Expand Down