320 Commits (a69a83095cbefbf0ab330216ae580e9d7084f016)

Author	SHA1	Message	Date
Or Sela	8f9e456535	helm/loki-stack: render loki datasource only if grafana is enabled (#2086)	5 years ago
James Hollowell	10a1f28a85	Support Prometheus on a sub-path in Grafana config (#2099) ... Co-authored-by: Cyril Tovena <cyril.tovena@gmail.com>	5 years ago
beorn7	a5aa34137a	Adjust Loki dashboards to the new label order ... - container_name and pod_name have disappeared in K8s 1.16. - We are changing the instance name in Prometheus (and removing it in Loki altogether). The instance name used to be the pod name, so switch usage of the instance label to the pod label. The pod label will generally only appear as a target label with rolling out recent changes to target labels. This will unfortunately break dashboards if looking at older data.	5 years ago
beorn7	123e6f51fe	Keep scrape config in line with the new Prometheus scrape config ... This is triggered by https://github.com/grafana/jsonnet-libs/pull/261 . The above PR removes the `instance` label. As it has turned out (see PR linked above), a sane `instance` label in Prometheus has to be unique, and that includes the case where a single container exposes metrics on two different endpoints. However, that scenario would still only result in one log stream for Loki to scrape. Therefore, Loki and Prometheus need to sync via target labels uniquely identifying a container (rather than a metrics endpoint). Those labels are namespace, pod, container, also added here. This commit removes the `container_name` label. It is the same as the `container` label and was already added to Loki previously. However, the `container_name` label is deprecated and has disappeared in K8s 1.16, so that it will soon become useless for direct joining.	5 years ago
Owen Diehl	89d80a6fc4	Feature/querysharding ii (#1927) ... * [wip] sharding evaluator/ast * [wip] continues experimenting with ast mapping * refactoring in preparation for binops * evaluators can pass state to other evaluators * compiler alignment * Evaluator method renamed to StepEvaluator * chained evaluator impl * tidying up sharding code * handling for ConcatSampleExpr * downstream iterator * structure for downstreaming asts * outlines sharding optimizations * work on sharding mapper * ast sharding optimizations * test for different logrange positions * shard mapper tests * stronger ast sharding & tests * shardmapper tests for string->string * removes sharding evaluator code * removes unused ctx arg * Revert "removes sharding evaluator code" This reverts commit 55d41b9519. * interfaces for downstreaming, type conversions * sharding plumbing on frontend * type alignment in queryrange to downstream sharded queriers * downstreaming support for sharding incl storage code * removes chainedevaluator * comment alignment * storage shard injection * speccing out testware for sharding equivalence * [wip] shared engine refactor * sorting streams, sharding eval fixes * downstream evaluator embeds defaultevaluator * other pkgs adopt logql changes * metrics & logs use same middleware instantiation process * wires up shardingware * middleware per metrics/logfilter * empty step populating StepEvaluator promql.Matrix adapter * sharding metrics * log/span injection into sharded engine * sharding metrics avoids multiple instantiation * downstreamhandler tracing * sharding parameterized libsonnet * removes querier replicas * default 32 concurrency for workers * jsonnet correct level override * unquote true in yaml * lowercase error + downstreamEvaluator defaults to embedded defaultEvaluator * makes shardRecorder private * logs query on failed parse * refactors engine to be multi-use, minimizes logger injection, generalizes Query methods, removes Engine interface * basic tests for querysharding mware * [wip] concurrent evaluator * integrates stat propagation into sharding evaluator * splitby histogram * extends le bounds for bytes processed * byte throughput histogram buckets to 40gb * chunk duration mixin * fixes merge w/ field rename * derives logger in sharded engine via ctx & logs some downstream evaluators * moves sharded engine to top, adds comments * logs failed merge results in stats ctx * snapshotting stats merge logic is done more effectively * per query concurrency controlled via downstreamer * unexports decodereq * queryrange testware * downstreamer tests * pr requests	5 years ago
Frederik Nordahl Jul Sabroe	156023ae9a	Allow configuring more options for output configuration (#2089) ... * Allow configuring more options for output configuration The `TenantID`, `BatchWait`, `BatchSize` and `AutoKubernetesLabels` options in the fluent-bit loki output plugin were not configurable. This change allows configuring those options via the helm chart. * typo in production/helm/fluent-bit/README.md Co-authored-by: Joseph Petersen <josephp90@gmail.com> Co-authored-by: Joseph Petersen <josephp90@gmail.com> Co-authored-by: Cyril Tovena <cyril.tovena@gmail.com>	5 years ago
Ed Welch	26a4c53cbd	updating all the images to 1.5.0 release (#2100) ... * updating all the images to 1.5.0 release Signed-off-by: Ed Welch <edward.welch@grafana.com> * duration must have a unit Signed-off-by: Ed Welch <edward.welch@grafana.com>	5 years ago
Ed Welch	12c7eab8bb	Prep 1.5.0 release (#2098) ... * Updating the Changelog and Upgrade guide for 1.5.0 release. Signed-off-by: Ed Welch <edward.welch@grafana.com> * Changing release number in all the docs Signed-off-by: Ed Welch <edward.welch@grafana.com>	5 years ago
Björn Rabenstein	d343c09aa9	Accept suffixes to pod name in instance labels (#2080) ... Currently, the instance label is just the pod name, but that's insufficient, cf. https://github.com/grafana/deployment_tools/issues/3441 . In the near future, we'll suffix the pod name in the instance label with container name and/or port name. To make this mixin still work, adjust the `instance` label matcher. Signed-off-by: beorn7 <beorn@grafana.com>	5 years ago
Cyril Tovena	bce4470a5b	Recover from panic in http and grpc handlers. (#2059) ... * Recover from panic in http and grpc handlers. I don't see any good reason to crash any component during a bad request. Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * Add alerts to the mixin for panics. Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * 😡 gomod Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com>	5 years ago
Naoki Oketani	918ba41982	Docs: update Loki Helm Chart document to support Helm 3 (#2038) ... * update Loki Helm Chart document to support Helm 3 * update Loki Helm Chart document to support Helm 2 & 3	5 years ago
Ed Welch	2629e9a3d6	fixes the cpu usage graphs (#2044) ... Signed-off-by: Ed Welch <edward.welch@grafana.com>	5 years ago
avii-ridge	69c1d1c0e0	Add extraOutputs variable to support multiple outputs for fluent-bit (#2040) ... * Adding support for extra outputs * added documentation	5 years ago
Joe Elliott	236d3728d5	Swapped to container restarts over terminated reasons (#2043) ... Signed-off-by: Joe Elliott <number101010@gmail.com>	5 years ago
Ed Welch	fd99dbc926	Dashboard: Loki Operational improvements (#2041) ... * fixing indentation so I can see diffs better Signed-off-by: Ed Welch <edward.welch@grafana.com> * add rows for chunks, streams, dropped samples. success rate graphs now only look at 5xx errors (4xx are considered "successful") Limited the mouseover to only show the active line for a bunch of graphs. Signed-off-by: Ed Welch <edward.welch@grafana.com>	5 years ago
Josh Soref	0d9cf4946a	Spelling (#2014) ... * spelling: access * spelling: actual * spelling: asynchronously * spelling: circleci * spelling: circle * spelling: cognitive * spelling: comment * spelling: computation * spelling: correct * spelling: daemon * spelling: elliott * spelling: enabled * spelling: encbuf * spelling: extracted * spelling: explicitly * spelling: independent * spelling: languages * spelling: loose * spelling: message * spelling: metrics * spelling: mismatch * spelling: nonexistent * spelling: output * spelling: pipeline * spelling: pipe line * spelling: position * spelling: prepend * spelling: query * spelling: receive * spelling: received * spelling: reducible * spelling: registry * spelling: response * spelling: retryable * spelling: scanning * spelling: sourced * spelling: supported * spelling: template * spelling: timeout * spelling: unsupported * spelling: adapter * spelling: deferring * spelling: expirable	5 years ago
Etash Singh	521999c419	Change image tag from 1.4.1 to 1.4.0 in Helm chart (#2015) ... * Change image tag from 1.4.1 to 1.4.0 Change image tag from 1.4.1 to 1.4.0 to make the chart work as the image with 1.4.1 runs into errors while deploying * Update chart version value * update loki stack chert version * Changed image tag value from 1.4.1 to 1.4.0 * Update values.yaml * Update Chart.yaml * Update Chart.yaml * Update Chart.yaml * Update loki-stack chart version	5 years ago
Smit Shah	b652f0a3f2	added extraCommandlineArgs in values file (#1981) ... * added extraCommandlineArgs in values file * corrected yaml with adding space * upgrade promtail chart version * update loki-stack chart version	5 years ago
Cyril Tovena	2374d4891b	Log slow query from the frontend by default in ksonnet. (#1980) ... Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com>	5 years ago
Ed Welch	e90e3b769f	incorporate some better defaults into table-manager configs (#1976) ... * Set less aggressive poll interval and a larger grace period for creating new periodic tables in the ksonnet config. Add some rate limiting and retry settings for the table manager bigtable arguments. Signed-off-by: Ed Welch <edward.welch@grafana.com> * Removing the dynamodb provisioning of 0 from config files Signed-off-by: Ed Welch <edward.welch@grafana.com>	5 years ago
Marc Bihlmaier	be56144485	helm chart: add missing line feed (#1967) ... * add missing LF in helm charts * increase helm Chart version	5 years ago
Callum Styan	a5d1f283e0	Move build from source instructions to root readme. (#1951) ... Signed-off-by: Callum Styan <callumstyan@gmail.com>	5 years ago
Florian Ludwig	153bcb0575	version pin the docker image in docker-compose (#1945) ... While the [documentation](https://github.com/grafana/loki/blob/master/docs/installation/docker.md) does version pin the links to specific versions of the docker-compose file, the file it docker-compose file does use the `latest` tag for loki. For an example of issues this causes see #1941	5 years ago
Ed	3fb316e4dd	Reverting helm changes which were included in Cortex 1.0 vendor upgrade because it resulted in us pushing an unstable helm.	5 years ago
Owen Diehl	f86785b68e	bumps helm to keepalive-master after cortex vendoring (#1937) ... * bumps helm to keepalive-master after cortex vendoring * minor helm version bump + keepalive image for promtail	5 years ago
Peter Štibraný	6d5408c100	Update Cortex to latest master (#1869) ... * Update Cortex to master Signed-off-by: Peter Štibraný <peter.stibrany@grafana.com> * Fix compilation errors after Cortex update. Signed-off-by: Peter Štibraný <peter.stibrany@grafana.com> * Update to Cortex v1.0.0-rc.0 Signed-off-by: Peter Štibraný <peter.stibrany@grafana.com> * Added missing vendored file. Signed-off-by: Peter Štibraný <peter.stibrany@grafana.com> * Vendor Cortex 1.0.0 Signed-off-by: Peter Štibraný <peter.stibrany@grafana.com> * Configuration changes. Signed-off-by: Peter Štibraný <peter.stibrany@grafana.com> * Mention Cortex 1.0 changes and link to docs. Signed-off-by: Peter Štibraný <peter.stibrany@grafana.com> * Fix jsonnet files, and mention fields that were removed. Signed-off-by: Peter Štibraný <peter.stibrany@grafana.com>	5 years ago
Tom Wilkie	299ba1ff44	And the loki mixin	5 years ago
Tom Wilkie	0170249eae	Put loki-mixin and promtail-mixin dashboards in a folder.	5 years ago
Cyril Tovena	0e53190e72	Add cap_net_bind_service linux capabilities to Loki. (#1916) ... * Add cap_net_bind_service linux capabilities to Loki. This allow Loki to bind to port below port 1024 without being a root user. Also update tanka to use 80 as default port since this is more intuitive to remember. Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * Move setcap to build step Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * Doesn't work inside the build step. Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * Set 3100 default port. Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * Again back to 80 :p Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com>	5 years ago
Nicholas St. Germain	d9423f3379	Fix fluent-bit metrics (#1915) ... * Add http server to allow prometheus metrics * add correct prometheus path * version bump	5 years ago
Tom Wilkie	8e5fa03de3	s/dashboards/grafanaDashboards. ... Signed-off-by: Tom Wilkie <tom@grafana.com>	5 years ago
Edward Welch	4995c97ff3	updating helm and ksonnet to 1.4.1 ... updating Fluent Bit to correct image Signed-off-by: Edward Welch <edward.welch@grafana.com>	5 years ago
Stefan Andres	94cd4f4330	[helm loki/promtail] make UpdateStrategy configurable (#1898) ... This commit makes UpdateStrategy of the promtail daemonset configurable. On large installations, you may want to increase the value of maxUnavailable. This fixes https://github.com/grafana/loki/issues/1878	5 years ago
Cyril Tovena	87c63855dd	Sign helm package with GPG. (#1864) ... * Sign helm package with GPG. Also add GPG to the build image and fixes dlv not pinned (this was causing issue using latest 1.4). Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * Regen drone config. Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * Testing CI. Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * Test CI. Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * Include .prov files for charts. Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * Revert testing changes. Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * Add public GPG key for verifying helm charts. Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com>	5 years ago
Edward Welch	43ef2d85af	updating helm and ksonnet for 1.4.0 release ... Signed-off-by: Edward Welch <edward.welch@grafana.com>	5 years ago
Stefan Andres	733b24620e	Add extraMount, extraVolumeMount to promtail helm chart for systemd-journal (#1871) ... This commit adds the variables extraMount and extraVolumeMount to the promtail helm chart. It also adds an example on how to use the helm chart with systemd-journal support.	5 years ago
polar3130	0a3621822b	Helm/loki-stack: refresh default grafana.image.tag to 6.7.0 (#1825) ... * update grafana.image.tag in loki-stack' helm chart * bump loki-stack chart Co-authored-by: Cyril Tovena <cyril.tovena@gmail.com>	5 years ago
Cyril Tovena	fa51393bce	Move default port for Loki to 3100 everywhere. (#1838) ... Also fixes default value for the table manager that doesn't support 0 anymore on master. Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com>	5 years ago
Cyril Tovena	a0dc10a03a	Use 8080 for Loki http port binding in tanka. (#1823) ... * Use 8080 for Loki http port binding. Previously we were using 80 but now that we run a rootless Loki for security reason. We can't bind below 1000. Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * Update also docker file and default config. Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com>	5 years ago
Bruno Clermont	ba53328673	Prevent prometheus to scrap both services (#1817) ... Signed-off-by: Bruno Clermont <bruno@robotinfra.com>	5 years ago
Marco Pracucci	7c6cbc21aa	Enable global limits by default in production mixin (#1807) ... Signed-off-by: Marco Pracucci <marco@pracucci.com>	5 years ago
Aditya C S	89263b09a7	Add /ready endpoint to table-manager (#1736)	5 years ago
sh0rez	a9f25a35d6	fix typo: Not a binary operator: = (#1776)	5 years ago
Joe Elliott	61bfa20ad5	Dashboard Cleanup (#1767) ... * Removed env references from operational dashboard Signed-off-by: Joe Elliott <number101010@gmail.com> * Updated refresh on template vars Signed-off-by: Joe Elliott <number101010@gmail.com> * Standard formatting Signed-off-by: Joe Elliott <number101010@gmail.com> * Removed env references Signed-off-by: Joe Elliott <number101010@gmail.com> * Added queries Signed-off-by: Joe Elliott <number101010@gmail.com> * memory => working set Signed-off-by: Joe Elliott <number101010@gmail.com>	5 years ago
Joe Elliott	e0d9be19e7	Move dashboards out into their own json files (#1766) ... Signed-off-by: Joe Elliott <number101010@gmail.com>	5 years ago
Edward Welch	5cf1dc51e4	Add another variable to pick the name of the promtail pod, we don't always deploy promtail with the same name and sometimes put more than one deployment in a namespace ... Signed-off-by: Edward Welch <edward.welch@grafana.com>	5 years ago
sh0rez	d42859c03a	fix(ksonnet): named parameters for containerPort ... Because ksonnet-lib switched the order of the `containerPort.new()` parameters from `ksonnet.beta.3` to `ksonnet.beta.4`, we have to use named parameters to avoid invalid resources when using one of these versions.	5 years ago
Edward Welch	8df56cea6e	Increasing the threshold for a file lag and reducing the severity to warning ... Signed-off-by: Edward Welch <edward.welch@grafana.com>	5 years ago
Edward Welch	f9e37e2997	breakout promtail mixin to be separate. ... Signed-off-by: Edward Welch <edward.welch@grafana.com>	5 years ago
Jeff Billimek	d27f11970a	[promtail helm chart] - Expand promtail syslog svc to support values (#1731) ... * Expand promtail syslog svc to support values The promtail helm chart syslog service configuration in the values.yaml supports elements not accounted-for in the actual syslog service template. This change modifys the promtail syslog service to account for those missing elements. Signed-off-by: Jeff Billimek <jeff@billimek.com> * Simplifying service typer selection process Signed-off-by: Jeff Billimek <jeffrey_k_billimek@homedepot.com>	5 years ago