Skip to content
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

Jump to bottom

refactor: config for job aggregation strings #319

Merged
merged 8 commits into from
Jun 10, 2021
Merged

refactor: config for job aggregation strings #319

Show file tree
Hide file tree
Changes from 7 commits
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
c9ec237
refactor: config for job aggregation strings
darrenjaneczek Jun 2, 2021
d613a91
lint
darrenjaneczek Jun 2, 2021
36ac142
Update cortex-mixin/dashboards/writes.libsonnet
darrenjaneczek Jun 3, 2021
cf89397
fix: syntax
darrenjaneczek Jun 3, 2021
31466e6
refactor: added a group_config
darrenjaneczek Jun 4, 2021
a8630f3
refactor: added a group_config
darrenjaneczek Jun 4, 2021
6c3aa96
refactor: added a group_config
darrenjaneczek Jun 4, 2021
a03451c
fix: changelog
darrenjaneczek Jun 8, 2021
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
2 changes: 1 addition & 1 deletion cortex-mixin/alerts.libsonnet
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -8,5 +8,5 @@
(import 'alerts/compactor.libsonnet')
else {}) +

{ _config:: $._config },
{ _config:: $._config + $._group_config },
}
2 changes: 1 addition & 1 deletion cortex-mixin/alerts/alerts.libsonnet
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -39,7 +39,7 @@
{
alert: 'CortexRequestLatency',
expr: |||
cluster_namespace_job_route:cortex_request_duration_seconds:99quantile{route!~"metrics|/frontend.Frontend/Process|ready|/schedulerpb.SchedulerForFrontend/FrontendLoop|/schedulerpb.SchedulerForQuerier/QuerierLoop"}
%(group_prefix_jobs)s_route:cortex_request_duration_seconds:99quantile{route!~"metrics|/frontend.Frontend/Process|ready|/schedulerpb.SchedulerForFrontend/FrontendLoop|/schedulerpb.SchedulerForQuerier/QuerierLoop"}
>
%(cortex_p99_latency_threshold_seconds)s
||| % $._config,
Expand Down
7 changes: 4 additions & 3 deletions cortex-mixin/config.libsonnet
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,9 +38,10 @@
compactor: 'compactor.*', // Match also custom compactor deployments.
},

// Labels used to in alert aggregations - should uniquely identify
// a single Cortex cluster.
alert_aggregation_labels: 'cluster, namespace',
// Grouping labels, to uniquely identify and group by {jobs, clusters}
job_labels: ['cluster', 'namespace', 'job'],
cluster_labels: ['cluster', 'namespace'],

cortex_p99_latency_threshold_seconds: 2.5,

// Whether resources dashboards are enabled (based on cAdvisor metrics).
Expand Down
2 changes: 1 addition & 1 deletion cortex-mixin/dashboards.libsonnet
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -31,5 +31,5 @@
(import 'dashboards/writes-resources.libsonnet') +
(import 'dashboards/alertmanager-resources.libsonnet')) +

{ _config:: $._config },
{ _config:: $._config + $._group_config },
}
15 changes: 11 additions & 4 deletions cortex-mixin/dashboards/writes.libsonnet
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,15 +12,22 @@ local utils = import 'mixin-utils/utils.libsonnet';
})
.addPanel(
$.panel('Samples / s') +
$.statPanel('sum(cluster_namespace_job:cortex_distributor_received_samples:rate5m{%s})' % $.jobMatcher($._config.job_names.distributor), format='reqps')
$.statPanel(
'sum(%(group_prefix_jobs)s:cortex_distributor_received_samples:rate5m{%(job)s})' % (
$._config {
job: $.jobMatcher($._config.job_names.distributor),
}
),
format='reqps'
)
)
.addPanel(
$.panel('Active Series') +
$.statPanel(|||
sum(cortex_ingester_memory_series{%(ingester)s}
/ on(namespace) group_left
max by (namespace) (cortex_distributor_replication_factor{%(distributor)s}))
||| % {
/ on(%(group_by_cluster)s) group_left
max by (%(group_by_cluster)s) (cortex_distributor_replication_factor{%(distributor)s}))
||| % ($._config) {
ingester: $.jobMatcher($._config.job_names.ingester),
distributor: $.jobMatcher($._config.job_names.distributor),
}, format='short')
Expand Down
45 changes: 45 additions & 0 deletions cortex-mixin/groups.libsonnet
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,45 @@
{
local makePrefix(groups) = std.join('_', groups),
local makeGroupBy(groups) = std.join(', ', groups),

local group_by_cluster = makeGroupBy($._config.cluster_labels),

_group_config+:: {
// Each group prefix is composed of `_`-separated labels
group_prefix_jobs: makePrefix($._config.job_labels),
group_prefix_clusters: makePrefix($._config.cluster_labels),

// Each group-by label list is `, `-separated and unique identifies
group_by_job: makeGroupBy($._config.job_labels),
group_by_cluster: group_by_cluster,
},

// The following works around the deprecation of `$._config.alert_aggregation_labels`
// - If an override of that value is detected, a warning will be printed
// - If no override was detected, it will be set to the `group_by_cluster` value,
// which will replace it altogether in the future.
local alert_aggregation_labels_override = (
Copy link
Collaborator

Choose a reason for hiding this comment

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

Good job here! I'm just wondering if we actually need this complexity. What we typically do is to introduce a breaking change (eg. stop using alert_aggregation_labels) and document it in the CHANGELOG. This mixin policy doesn't guarantee backward compatibility.

Copy link
Contributor Author

Choose a reason for hiding this comment

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

I was concerned that my change could cause a phantom failure for someone, where their custom override would just stop working without any warning/error.

A plain old "error if alert_aggregation_labels is defined" would be less complex and solve my concern.

Let me know if/how you want this addressed in this PR.

Copy link
Collaborator

Choose a reason for hiding this comment

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

Let's not block on this. What you did looks correct, I'm just a bit worried about its maintenance, but not a blocker.

{
alert_aggregation_labels: null,
} + super._config
).alert_aggregation_labels,

_config+:: {
alert_aggregation_labels:
if alert_aggregation_labels_override != null
then std.trace(
|||
Deprecated: _config.alert_aggregation_labels
This field has been explicitly overridden to "%s".
Instead, express the override in terms of _config.cluster_labels.
E.g., cluster_labels: %s will automatically convert to "%s".
||| % [
alert_aggregation_labels_override,
$._config.cluster_labels,
group_by_cluster,
],
alert_aggregation_labels_override
)
else group_by_cluster,
},
}
1 change: 1 addition & 0 deletions cortex-mixin/mixin.libsonnet
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,4 +1,5 @@
(import 'config.libsonnet') +
(import 'groups.libsonnet') +
(import 'dashboards.libsonnet') +
(import 'alerts.libsonnet') +
(import 'recording_rules.libsonnet')
22 changes: 11 additions & 11 deletions cortex-mixin/recording_rules.libsonnet
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,12 @@
local utils = import 'mixin-utils/utils.libsonnet';

{
local _config = {
max_series_per_ingester: 1.5e6,
max_samples_per_sec_per_ingester: 80e3,
max_samples_per_sec_per_distributor: 240e3,
limit_utilisation_target: 0.6,
} + $._config + $._group_config,
prometheusRules+:: {
groups+: [
{
Expand Down Expand Up @@ -51,20 +57,14 @@ local utils = import 'mixin-utils/utils.libsonnet';
name: 'cortex_received_samples',
rules: [
{
record: 'cluster_namespace_job:cortex_distributor_received_samples:rate5m',
record: '%(group_prefix_jobs)s:cortex_distributor_received_samples:rate5m' % _config,
expr: |||
sum by (cluster, namespace, job) (rate(cortex_distributor_received_samples_total[5m]))
|||,
sum by (%(group_by_job)s) (rate(cortex_distributor_received_samples_total[5m]))
||| % _config,
},
],
},
{
local _config = {
max_series_per_ingester: 1.5e6,
max_samples_per_sec_per_ingester: 80e3,
max_samples_per_sec_per_distributor: 240e3,
limit_utilisation_target: 0.6,
},
name: 'cortex_scaling_rules',
rules: [
{
Expand All @@ -89,7 +89,7 @@ local utils = import 'mixin-utils/utils.libsonnet';
ceil(
quantile_over_time(0.99,
sum by (cluster, namespace) (
cluster_namespace_job:cortex_distributor_received_samples:rate5m
%(group_prefix_jobs)s:cortex_distributor_received_samples:rate5m
)[24h:]
)
/ %(max_samples_per_sec_per_distributor)s
Expand Down Expand Up @@ -123,7 +123,7 @@ local utils = import 'mixin-utils/utils.libsonnet';
ceil(
quantile_over_time(0.99,
sum by (cluster, namespace) (
cluster_namespace_job:cortex_distributor_received_samples:rate5m
%(group_prefix_jobs)s:cortex_distributor_received_samples:rate5m
)[24h:]
)
* 3 / %(max_samples_per_sec_per_ingester)s
Expand Down