apitech/loki - loki - Gitea: Git with a cup of tea

Commit Graph

Author	SHA1	Message	Date
Quentin Bisson	a03846b436	fix: Fix compactor matcher in the loki-deletion dashboard (#12790 )	1 year ago
Michel Hollands	ec1a057a32	feat: Add a version of the mixin dashboards for meta monitoring (#12700 ) Signed-off-by: Michel Hollands <michel.hollands@gmail.com>	1 year ago
Christian Haudum	807d692444	chore(loki-mixin): Fixes for `loki-reads` dashboard (#12788 ) * Update HTTP and GRPC routes * Add QPS and latency panels for Index Gateway * Rename row Index to TSDB Index and BoltDB Shipper to BoltDB Index for consistency Signed-off-by: Christian Haudum <christian.haudum@gmail.com>	1 year ago
Callum Styan	91ef72f742	fix(mixin): dashboards $__auto fix (#12707 ) Signed-off-by: Callum Styan <callumstyan@gmail.com>	1 year ago
Quentin Bisson	006f88cef1	fix: incorrect compactor matcher in loki-deletion dashboard mixin (#12567 ) Signed-off-by: QuentinBisson <quentin@giantswarm.io>	1 year ago
Callum Styan	be1a77255c	ci: add a github action for linting jsonnet (#12527 ) Signed-off-by: Callum Styan <callumstyan@gmail.com>	1 year ago
Salva Corts	9fe31c14ee	fix: Update deprecated angular graph panel in dashboards mixin (#12430 )	1 year ago
Paul Rogers	d4e359dea8	Change distributor latency to have a regular expression equals (#11827 ) Currently, the distributor latency panel uses a "=" in the code as opposed to a "=~", causing the panel to render as empty. This is to fix that issue. What this PR does / why we need it: Which issue(s) this PR fixes: Fixes #<issue number> Special notes for your reviewer: Checklist - [ ] Reviewed the [`CONTRIBUTING.md`](https://github.com/grafana/loki/blob/main/CONTRIBUTING.md) guide (required) - [ ] Documentation added - [ ] Tests updated - [ ] `CHANGELOG.md` updated - [ ] If the change is worth mentioning in the release notes, add `add-to-release-notes` label - [ ] Changes that require user attention or interaction to upgrade are documented in `docs/sources/setup/upgrade/_index.md` - [ ] For Helm chart changes bump the Helm chart version in `production/helm/loki/Chart.yaml` and update `production/helm/loki/CHANGELOG.md` and `production/helm/loki/README.md`. [Example PR](`d10549e3ec`) - [ ] If the change is deprecating or removing a configuration option, update the `deprecated-config.yaml` and `deleted-config.yaml` files respectively in the `tools/deprecated-config-checker` directory. [Example PR](`0d4416a4b0`)	1 year ago
Dylan Guedes	9964a30b62	Jsonnet: Change by-pod source code to reuse the `latency` util function (#11783 ) What this PR does / why we need it: Modify our implementation of `p99LatencyByPod` to reuse the `latencyPanel` util function.	1 year ago
Joao Marcal	0694d797de	mixins: add route to write Distributor Latency dashboard (#11637 ) Co-authored-by: Periklis Tsirakidis <periklis@redhat.com>	1 year ago
Dylan Guedes	31f8c91771	Jsonnet: Fix and update loki-mixin panels (#11529 ) What this PR does / why we need it: * Update our legacy panels (Loki/Logs and Loki/Operation) do use `"type": "timeseries"` instead of `"type": "graph"`. The graph type is deprecated on recent Grafana versions. I tested the changes with Grizzly and it worked fine. * Fix our p99 panel unit. I previously made it to be `s` instead of `short` but when replicating the change on our source code I put the change at the wrong identation level. This time I made sure to test the change with Grizzly.	1 year ago
Joao Marcal	bd505f8e2d	mixins: Adds structured metadata panels (#11087 ) Co-authored-by: Periklis Tsirakidis <periklis@redhat.com>	2 years ago
Sandeep Sukhani	6a62b8cf42	compaction: Separate metrics for tracking retention and compaction (#11263 ) What this PR does / why we need it: In PR #9884, we separated the retention loop from compaction to avoid blocking compaction for too long due to some intensive delete requests. Currently, we track retention and compaction using the same metrics. This PR adds separate metrics for monitoring retention operation. I have also updated the Retention dashboard to use the new metrics.	2 years ago
Dylan Guedes	4309add1b2	Fix per-pod panel unit (from 'ms' to 's') (#11245 ) What this PR does / why we need it: Change per-pod latency panel unit from 'ms' to 's' (the metric is in the 's' unit)	2 years ago
Dylan Guedes	6d822928b4	Jsonnet: Fix unit used by per-pod latency panel (#11104 ) What this PR does / why we need it: The per-pod latency panel should use "ms" as the unit, instead of the default (which is "short"). I'm also wrapping everything around "defaults", because that's how it is used everywhere. Which issue(s) this PR fixes: N/A	2 years ago
Michel Hollands	e93f5bfd93	Add dashboard changes for migrating from cortex_ metrics to loki_ metrics (#11082 ) What this PR does / why we need it: - add missing namespace param for ruler manager metrics - update the loki mixin to use the renamed metrics so both are supported - created the compiled mixin - make the prefix of the cortex_distributor_ingester_clients metrics configurable Checklist - [X] Reviewed the [`CONTRIBUTING.md`](https://github.com/grafana/loki/blob/main/CONTRIBUTING.md) guide (required) - [ ] Documentation added - [ ] Tests updated - [ ] `CHANGELOG.md` updated - [ ] If the change is worth mentioning in the release notes, add `add-to-release-notes` label - [ ] Changes that require user attention or interaction to upgrade are documented in `docs/sources/setup/upgrade/_index.md` - [ ] For Helm chart changes bump the Helm chart version in `production/helm/loki/Chart.yaml` and update `production/helm/loki/CHANGELOG.md` and `production/helm/loki/README.md`. [Example PR](`d10549e3ec`) - [ ] If the change is deprecating or removing a configuration option, update the `deprecated-config.yaml` and `deleted-config.yaml` files respectively in the `tools/deprecated-config-checker` directory. [Example PR](`0d4416a4b0`) --------- Signed-off-by: Michel Hollands <michel.hollands@gmail.com>	2 years ago
Poyzan	be2ecdc241	(Fix) cpu calculation queries in loki / operational (#10477 ) What this PR does / why we need it: irate is debated to show finer details, however it discards samples and only uses last two in range. This results in some smoothing out in our CPU utilisation graphs which mislead us during an investigation a while back. See the screenshot below from a random ingester. ![Screenshot 2023-09-06 at 15 58 45](https://github.com/grafana/loki/assets/31743851/d57483f2-e207-485b-b4b5-fd43c1d0ddca) Which issue(s) this PR fixes: Fixes #<issue number> Special notes for your reviewer: I choose to change the entire query to be in line with rest of the Loki Read/Write Resources dashboards. Checklist - [x] Reviewed the [`CONTRIBUTING.md`](https://github.com/grafana/loki/blob/main/CONTRIBUTING.md) guide (required) - [ ] Documentation added - [ ] Tests updated - [ ] `CHANGELOG.md` updated - [ ] If the change is worth mentioning in the release notes, add `add-to-release-notes` label - [ ] Changes that require user attention or interaction to upgrade are documented in `docs/sources/setup/upgrade/_index.md` - [ ] For Helm chart changes bump the Helm chart version in `production/helm/loki/Chart.yaml` and update `production/helm/loki/CHANGELOG.md` and `production/helm/loki/README.md`. [Example PR](`d10549e3ec`)	2 years ago
Trevor Whitney	67002f3465	Add per pod latency plot to reads dashboard (#10194 ) This PR adds a Per Pods P99 Latency plot to each row of the Loki/Reads dashboard.	2 years ago
Michel Hollands	10e3d80731	Fix distributor latency panel (#8760 ) What this PR does / why we need it: https://github.com/grafana/loki/pull/8247 changes a metric generated by a recording rule. This fixes the "Distributor Panel" to use this different metric. Tested by changing the "Distributor Panel" manually to see it showed results. Checklist - [ ] Reviewed the [`CONTRIBUTING.md`](https://github.com/grafana/loki/blob/main/CONTRIBUTING.md) guide (required) - [ ] Documentation added - [ ] Tests updated - [ ] `CHANGELOG.md` updated - [ ] Changes that require user attention or interaction to upgrade are documented in `docs/sources/upgrading/_index.md` --------- Signed-off-by: Michel Hollands <michel.hollands@grafana.com>	2 years ago
Michel Hollands	1c4dd6ad40	Use changed metric in operational dashboard (#8749 ) What this PR does / why we need it: The metric `job_route` was changed to `cluster_job_route` in https://github.com/grafana/loki/pull/8247. This PR updates this metric in the operational dashboard. Checklist - [ ] Reviewed the [`CONTRIBUTING.md`](https://github.com/grafana/loki/blob/main/CONTRIBUTING.md) guide (required) - [ ] Documentation added - [ ] Tests updated - [ ] `CHANGELOG.md` updated - [ ] Changes that require user attention or interaction to upgrade are documented in `docs/sources/upgrading/_index.md` --------- Signed-off-by: Michel Hollands <michel.hollands@grafana.com>	2 years ago
Dylan Guedes	ea6abbfd07	Loki-Mixin: Remove query-readiness panel (#8735 ) What this PR does / why we need it: Remove from loki-mixin the panel that shows the query_readiness metric. The metric was removed a while ago.	2 years ago
Trevor Whitney	dac3b84d08	Loki cloud integration instructions (and necessary mixin changes) (#8492 ) What this PR does / why we need it: This PR is a first stab at documentation for setting up the Grafana Cloud Loki Integration to monitor a local Loki cluster, installed using the helm chart, using Grafana Cloud. In addition to instructions on how to collect the necessary kubernetes metrics, there were also some things with the Loki mixin (which provides dashboards, alerts, and rules to the integration) that needed to be fixed. For example, our dashboards rely on there being a cluster label on our recording rules, yet our mixin does not include this? Also, we are still including `cortex-gw` panels even when the mixin is compile with `internal_components: false`. Special notes for your reviewer: The changes to the mixin will not be reflected in the current version of the integration. We need to merge this PR first then update the integration. Checklist - [ ] Reviewed the [`CONTRIBUTING.md`](https://github.com/grafana/loki/blob/main/CONTRIBUTING.md) guide (required) - [ ] Documentation added - [ ] Tests updated - [ ] `CHANGELOG.md` updated - [ ] Changes that require user attention or interaction to upgrade are documented in `docs/sources/upgrading/_index.md` --------- Co-authored-by: Michel Hollands <42814411+MichelHollands@users.noreply.github.com> Co-authored-by: J Stickler <julie.stickler@grafana.com>	2 years ago
Charlie N	340f62ac7a	[Mixin] Update clusterlabel usage (#8247 ) Signed-off-by: Whyeasy <Whyeasy@users.noreply.github.com>	2 years ago
liam-howe-maersk	4681f9f487	Alerts: Exclude '/schedulerpb.SchedulerForQuerier/QuerierLoop' route in LokiRequestLatency alert (#8138 ) Excludes the route `/schedulerpb.SchedulerForQuerier/QuerierLoop` from the `LokiRequestLatency` alert query config. We have found that this route often has extremely high request latencies which will trigger the alert constantly for us. Looking through the source code it seems like it is expected to have high request latencies for this route and so in our configuration have excluded the route. We can see that this route is also being excluded in the [respective mimir config](https://github.com/grafana/mimir/blob/main/operations/mimir-mixin-compiled/alerts.yaml#L33) so believe it might be relevant to exclude here in loki as well	2 years ago
Dbzman	cd1ba1911e	Fix labels for alerts (#7886 ) This fixes wrong labels for the LokiCanaryLatency alert and introduces "cluster" as label for LokiTooManyCompactorsRunning to be able to have multiple loki instances in the same namespace. documented in `docs/sources/upgrading/_index.md`	2 years ago
Bryan Boreham	2bad98dc8e	dashboards: add k8s resource requests to CPU and memory panels (#8047 ) What this PR does / why we need it: This lets administrators see at a glance if requests are too low, which risks overloading nodes, or too high which wastes provisioning. Example: <img width="356" alt="image" src="https://user-images.githubusercontent.com/8125524/210851875-f9ccbb60-1a3c-4276-b65f-645aec0268be.png"> Checklist - [x] Reviewed the `CONTRIBUTING.md` guide - NA Documentation added - NA Tests updated - [x] `CHANGELOG.md` updated - NA Changes that require user attention or interaction to upgrade are documented in `docs/sources/upgrading/_index.md` Co-authored-by: Owen Diehl <ow.diehl@gmail.com>	2 years ago
Christian Haudum	48c3fb4c7c	Fix label selector for gateway service in operational dashboard Use `job=~\"$namespace/cortex-gw(-internal)?\"` throughout the dashboard. This fixes the occurrence where a regex value was used in combination with an exact label match. Signed-off-by: Christian Haudum <christian.haudum@gmail.com>	3 years ago
Christian Haudum	6c7a4dcdbd	[loki-mixin] Make "Success Rate" panels in operational dashboard more accurate (#7756 )	3 years ago
Christian Haudum	fbd5bbb09d	Improve error logs query in operational dashboard Instead of parsing all lines and filtering on `level` label, just filter by `level=error` string. Signed-off-by: Christian Haudum <christian.haudum@gmail.com>	3 years ago
Goutham Veeramachaneni	1260adc756	[mixin] Add cortex-gw-internal to dashboards (#7481 )	3 years ago
Karsten Jeschkies	62707bf517	Loki mixin: Correct Loki datasource variable name. (#7422 )	3 years ago
Callum Styan	a7e7a8e013	[dashboards] update dashboards for zone aware ingesters (#7313 ) Mostly related to the job selector; I deployed these against a local grafana instance along the way and I think I got everything correct but we might still need to update some of these again. Signed-off-by: Callum Styan <callumstyan@gmail.com> Signed-off-by: Callum Styan <callumstyan@gmail.com>	3 years ago
Michel Hollands	682e9c881d	Add list of delete requests to deletion dashboard (#7309 ) Add 2 panels to the deletion dashboard that show: - the in progress/finished delete requests during the selected period - the delete request that have been entered, with the query This will help with trouble-shooting deletions. Signed-off-by: Michel Hollands <michel.hollands@grafana.com>	3 years ago
Karsten Jeschkies	c53c6072cf	[Mixin] Rename `logs` template variable to `loki_datasource`. (#7278 ) What this PR does / why we need it: Grafana Cloud integration have the convention to name data source variables after `<type>_datasource`. That means `$logs` should become `loki_datasource`.	3 years ago
Michel Hollands	74c8cf03ba	Update deletions dashboard (#7244 ) Add a panel for the deletions in progress (which is received - processed). This is similar to the existing 'Number of pending requests' only panel but seems more reliable. I'll remove one later after this has been running for a while. Add some compactor CPU, memory and processing duration statistics so you can see how the compactor is doing. Signed-off-by: Michel Hollands <michel.hollands@grafana.com>	3 years ago
Karsten Jeschkies	a1dce32d9a	Abide to Mixin linting rules. (#7010 ) Cloud integrations need to pass linting rules such as proper dashboard names. It should be `Data Source` not `datasource`. See also grafana/cloud-onboarding#2228	3 years ago
Salva Corts	7314d95a91	Support dashboards for SSD mode (#6540 ) * Support dashboards for SSD mode * Fix issues * Add compiled mixins * Flip $.config.ssd on if conditions * Use prefix matcher for ssd * Add compiled SSD mixin * Update loki-mixin-check to check ssd compiled mixin * Checkout compiled ssd mixin	3 years ago

37 Commits (a03846b4367cbb5a0aa445e539d92ae41e3f481a)