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

Commit Graph

Author	SHA1	Message	Date
Benoît Knecht	489898f529	promtail/server: Disable profiling by default (#5400 ) What this PR does / why we need it: Add a `server.debug` option (that defaults to `false`) to enable the `/debug/fgprof` endpoint in Promtail. Which issue(s) this PR fixes: Fixes #5005 Co-authored-by: Trevor Whitney <trevorjwhitney@gmail.com>	3 years ago
李国忠	fe49e6642e	[bugfix] add Promtail cloudflare close (#7394 ) What this PR does / why we need it: If the http server closes the connection, the client may also return a non-empty reader io.ReadCloser. Which issue(s) this PR fixes: Fixes #6150 Co-authored-by: Christian Haudum <christian.haudum@gmail.com>	3 years ago
gelo22	8b89343b29	batchRetries to countersWithHost (#7392 ) What this PR does / why we need it: Metric promtail_batch_retries_total is absent when has value == 0 Which issue(s) this PR fixes: Fixes #7261	3 years ago
Furkan Türkal	28d6a97991	fluent-bit: clarify docs, use 1.9.9 (#6464 ) What this PR does / why we need it: * Pins digests for security enhancements * Pass `-ldflags` using `--build-args` * Use fluent-bit image `1.9.9` * Use golang image `1.19.2` * Clarify fluent-bit readme Signed-off-by: Furkan <furkan.turkal@trendyol.com> Co-authored-by: Michel Hollands <42814411+MichelHollands@users.noreply.github.com>	3 years ago
Max Vorobev	c5911d5342	Loki: implement decolorize filter (#7602 ) Implement `decolorize` filter that get rids of ANSI color codes in the log message. Which issue(s) this PR fixes: Fixes #7601	3 years ago
MarNicGit	17c36d60f3	Promtail: Fix collecting userdata field from Windows Event Log (#7461 ) What this PR does / why we need it: The windows event log target for Promtail had a typo that led to `EventData` being assigned to `UserData`. This PR fixes this typo. Which issue(s) this PR fixes: Fixes #6167	3 years ago
Robert Fratto	3808fb6ac3	promtail: register reload metrics after running options (#7561 ) What this PR does / why we need it: This allows the reload metrics to be sent to a custom registerer, which is required for Grafana Agent to be able to use the Promtail code. Which issue(s) this PR fixes: N/A Special notes for your reviewer: Is it possible for us to cherry-pick this into k121? That's the version we're trying to update Grafana Agent to. Checklist - [x] Reviewed the `CONTRIBUTING.md` guide - [x] Documentation added - [x] Tests updated - [x] `CHANGELOG.md` updated - [x] Changes that require user attention or interaction to upgrade are documented in `docs/sources/upgrading/_index.md`	3 years ago
Christian Haudum	0f5dd2bceb	Revert "fluentd: Add un-escaping of control characters in JSON" (#7468 ) Reverts grafana/loki#7396 according to discussing on PR	3 years ago
BenjaminVouillaumeJT	2d649a7fdd	Promtail: Add additional fields for cloudflare (#7463 )	3 years ago
Pablo	5f665651d5	Promtail: Add basic tracing support (#7414 ) We've been using Promtail's HTTP/gRPC server-based targets with real workloads for some weeks now. The server that these spin out is based on [weaveworks commons](https://github.com/weaveworks/common) HTTP/gRPC servers. These include by default some middlewares that if enabled, inject tracing support. We found this useful when tracking down request handling time, or even more, if the caller also support Jaeger tracing support, cross-correlated requests between the upstream server and Promtail's. This PR adds support for enabling tracing the same way Loki does. At the moment, this will just generate traces for the before-mentioned targets. In the future, the quality and completeness of tracing support can grow organically to enrich spans.	3 years ago
Robert Fratto	2208ea7147	Update Prometheus dependency to latest release (v2.39.1) (#7425 )	3 years ago
Richard Treu	5fe714e466	fluentd: Add un-escaping of control characters in JSON (#7396 )	3 years ago
Danny Kopping	2caf5f7c5d	Promtail: correctly hashes the config on reload (#7421 )	3 years ago
Pablo	e73584a3cd	Promtail: Add handler timeout for GCP Logs Push target (#7401 ) What this PR does / why we need it: After been running the GCP Logs Push target in actual workloads, we've noticed that GCP sends a lot of traffic, and adjusts that send rate according to the rate at which the receiver can process. To perform that adjustment, GCP considers as a NACK [non 2xx HTTP responses](https://cloud.google.com/pubsub/docs/push#receive_push). This PR adds a target handler timeout (that includes as well the `api.Entry` channel send) to allow the user to configure this maximum processing time, therefore giving GCP notice that the sending rate is too high. This new timeout is optional, and disabled by default. Which issue(s) this PR fixes: Related to https://github.com/grafana/cloud-onboarding/issues/2067 Special notes for your reviewer: Checklist - [ ] Reviewed the `CONTRIBUTING.md` guide - [x] Documentation added - [x] Tests updated - [x] `CHANGELOG.md` updated - [x] Changes that require user attention or interaction to upgrade are documented in `docs/sources/upgrading/_index.md`	3 years ago
李国忠	61bbe15171	[promtail] log: simplify log, only print hash (#7405 ) Compute and print MD5 hash of config	3 years ago
Vijal Patel	5f7543d047	Fix promtail scrape config job name validation (#7402 )	3 years ago
Danny Kopping	33bf36aa21	Promtail: flush logs on exit (#7400 ) Promtail is using the line-buffered logger (https://github.com/grafana/loki/pull/6954), and log messages were not being printed on exit	3 years ago
李国忠	fdbb06d123	add promtail reload changelog and doc (#7386 )	3 years ago
李国忠	fb26baa5b1	[new feature] promtail: Add config reload endoint / signal to promtail (#7247 )	3 years ago
Brandon W Maister	381735d3d4	logging: Change Docker target discovery log level from Error to Info (#7312 ) What this PR does / why we need it: Discovering a new node is not an error, so having it send errors to dashboards is confusing. Special notes for your reviewer: Checklist - [x] Reviewed the `CONTRIBUTING.md` guide - [x] Documentation added - [ ] Tests updated - [x] `CHANGELOG.md` updated - [x] Changes that require user attention or interaction to upgrade are documented in `docs/sources/upgrading/_index.md`	3 years ago
Hemanth Krishna	0e54477309	area(promtail): add -check-syntax functionality (#7363 ) Adds `-check-syntax` to promtail which will exit just by validating the config file. Fixes #6862. Signed-off-by: Hemanth Krishna <hkpdev008@gmail.com>	3 years ago
Danny Kopping	c3e92b876c	Restoring lint checks that were previously disabled (#7340 ) I removed these checks in https://github.com/grafana/loki/pull/7337 to not make https://github.com/grafana/loki/pull/7243 overly complex to review. Refactored code to fix valid lint issues, ignored the irrelevant/dangerous ones. I also removed `unused` and `unparam` as these produce low-quality results that are largely useless.	3 years ago
Danny Kopping	355175ff89	Upgrade to Go 1.19 (#7243 ) Upgrade Loki and all related components to Go 1.19.2.	3 years ago
Danny Kopping	6bf5b5d8c1	Use a line-buffered logger to deamplify write syscalls (#6954 ) We initialise a global logger in `pkg/util/log/log.go` and use it extensively throughout the Loki codebase. Every time we write a log message, a `write` syscall is invoked. Syscalls are problematic because they transition the process from userspace to kernelspace, which means: - a context-switch is incurred, which is inherently expensive ([1-2 microseconds](https://eli.thegreenplace.net/2018/measuring-context-switching-and-memory-overheads-for-linux-threads/)) - the goroutine executing the code is blocked - the underlying OS thread (_M_ in the go scheduler model) is also blocked - the goroutine has to be rescheduled once the syscall exits - the go scheduler may need to spawn additional OS threads if all are blocked in syscalls - which can also be expensive This change introduces the use of a line-buffered logger. It has a buffer of [256](https://gist.github.com/dannykopping/0704db32c0b08751d1d2494efaa734c2) entries, and once that buffer is filled it will flush to disk. However, a situation will arise in which that buffer remains somewhat empty for a period of time, so there is a periodic flush mechanism, configured to flush every 100ms. There is also a preallocated bytes slice of 10MB which is reused, to avoid excessive slice resizing & garbage collection. This does mean that we could lose up to 256 log messages in case of an ungraceful termination of the process, but this would need to be precisely timed within the 100ms flushes - in other words, the likelihood is low, and generally we shouldn't `kill -9` any Loki process.	3 years ago
Irina	d1a996afce	Added mapstructure tag to promtail configs structs (#7320 ) Added mapstructure tags to configs structs to promtail client. It is needed for [opentelemetry-collector-contrib](https://github.com/open-telemetry/opentelemetry-collector-contrib). I working on adding promtail receiver component there. Opentelemetry collector decodes configs with mapstructure	3 years ago
Dylan Guedes	f571282ffc	Tests: Improve several tests behavior (#7283 ) What this PR does / why we need it: Follow-up of https://github.com/grafana/loki/pull/7273 * Tune a filetarget test to remove created files to guarantee that all tailers are interrupted/fred * Call pubsub close for our GCP `pull target` * Append testName to the job names used in GCP tests. This will help us to identify which tests are struggling.	3 years ago
Dylan Guedes	2dc5a71a67	Tests: Reduce flakiness of GCPLog and integration tests (#7273 ) What this PR does / why we need it: - Reduce flakiness of GCPLog tests by reusing context and stopping GCP topic - Reduce flakiness of integration tests by: - using port as they're assigned - reusing context - adding timeout to calls (hence, if something get stuck, tests fail earlier)	3 years ago
Dylan Guedes	f3b72c8b59	Promtail: Change max support compressed line size to be 2MB (#7268 ) What this PR does / why we need it: Change Promtail compression support max line size to 2MB instead of 4096 bytes (~4KB). This however doesn't change the default buffer size. 4096 bytes means we can't ever have a buffer size bigger than ~4KB but since the default mem page size is 4096 ([source](https://unix.stackexchange.com/questions/128213/how-is-page-size-determined-in-virtual-address-space)), we end up ignoring all lines.	3 years ago
Pablo	64fe0af9dd	Expose tenant ID through relabel in push promtail targets (#7250 ) What this PR does / why we need it: Some Promtail targets that support pushing logs through HTTP, allow using a specific header for writing logs for multiple tenants. This log is propagated through a Promtail internal label, that is then consumed by the remote write client. If Promtail is used in a multi-tenant manner, it'd be nice to somehow be able to track how much is each tenant's share from the logs received. This PR exposes the received tenant id value through relabeling. This allows for example, keeping it as a label and exposing a metric though the `metrics` pipeline stage, which can be segregated by tenant. Which issue(s) this PR fixes: Related to https://github.com/grafana/cloud-onboarding/issues/2067	3 years ago
Dylan Guedes	73bea7e4fa	Promtail: Add compressed files support (#6708 ) What this PR does / why we need it: Adds to Promtail the ability to read compressed files. It works by: 1. Infer which compression format to use based on the file extension 2. Uncompress the file with the native `golang/compress` packages 3. Iterate over uncompressed lines and send them to Loki Its usage is the same as our current file tailing. Which issue(s) this PR fixes: Fixes #5956 Co-authored-by: Danny Kopping <dannykopping@gmail.com>	3 years ago
Pablo	42bfc22d23	Fix promtail `metric` stage causing failure when getting extracted metrics (#7216 ) What this PR does / why we need it: Promtail allows ingested log entries to be modified before pushing them to Loki with the use of pipelines. One of the available pipeline "stages" is called `metrics`, and allows extracting metrics from incoming logs. These metrics are labeled with the labels available in the processed `Entry`. There's some edge cases in which if either a entry label is invalid, or it starts with the prometheus internal label prefix, `__`, calling the promtail `/metrics` fails. One way to easily reproduce this is by having the following configuration: ```yaml pipeline_stages: - static_labels: good_label: "1" - tenant: value: "2" - metrics: loki_count: type: Counter description: "should count all entries" config: match_all: true action: inc ``` Which exposes a `promtail_custom_loki_count` metrics that counts processed log entries. In this case, the step before the `metrics` one injects an internal label `__tenant_id__` to be used by the remote write client. Since this label starts with the prometheus reserved prefix, this will make calling the `metrics` endpoint fail with an error like the following: ``` An error has occurred while serving metrics: 1 error(s) occurred: * "__tenant_id__" is not a valid label name for metric "promtail_custom_loki_count" ``` This PR attempts to fix that by, before creating the corresponding metrics, dropping invalid or internal labels. That way promtail will just use labels that are valid. Which issue(s) this PR fixes: Related to https://github.com/grafana/cloud-onboarding/issues/2067 Special notes for your reviewer: Checklist - [ ] Reviewed the `CONTRIBUTING.md` guide - [ ] Documentation added - [x] Tests updated - [ ] `CHANGELOG.md` updated - [ ] Changes that require user attention or interaction to upgrade are documented in `docs/sources/upgrading/_index.md`	3 years ago
Eng Zer Jun	e2842c69c5	Replace deprecated `io/ioutil` with `io` and `os` packages (#7212 ) The `io/ioutil` package has been deprecated in Go 1.16 (See https://pkg.go.dev/io/ioutil). This PR replaces the existing `io/ioutil` functions with their new definitions in `io` and `os` packages. Signed-off-by: Eng Zer Jun <engzerjun@gmail.com>	3 years ago
Jeremy Wu	161fcac646	support custom label selector in kubernetes_sd_configs (#5920 ) make promtail scrape logs of pod with specific field or label selector, so that it will watch less logs and make fewer calculations and get a better performance in some scenarios. Signed-off-by: Zhimin Wu <hoozecn@gmail.com> Co-authored-by: Karen Miller <84039272+KMiller-Grafana@users.noreply.github.com>	3 years ago
Chris Hodges	d1f42fff50	promtail: Fix issue with dropping logs when a file based SD target's labels are updated (#7016 ) <!-- Thanks for sending a pull request! Before submitting: 1. Read our CONTRIBUTING.md guide 2. Name your PR as `<Feature Area>: Describe your change`. a. Do not end the title with punctuation. It will be added in the changelog. b. Start with an imperative verb. Example: Fix the latency between System A and System B. c. Use sentence case, not title case. d. Use a complete phrase or sentence. The PR title will appear in a changelog, so help other people understand what your change will be. 3. Rebase your PR if it gets out of sync with main --> What this PR does / why we need it: Allows for modifying labels for a target and continuing to scrape logs afterwards. Which issue(s) this PR fixes: Fixes #7001 Special notes for your reviewer: <!-- Note about CHANGELOG entries, if a change adds: * an important feature * fixes an issue present in a previous release, * causes a change in operation that would be useful for an operator of Loki to know then please add a CHANGELOG entry. For documentation changes, build changes, simple fixes etc please skip this step. We are attempting to curate a changelog of the most relevant and important changes to be easier to ingest by end users of Loki. Note about the upgrade guide, if this changes: * default configuration values * metric names or label names * changes existing log lines such as the metrics.go query output line * configuration parameters * anything to do with any API * any other change that would require special attention or extra steps to upgrade Please document clearly what changed AND what needs to be done in the upgrade guide. --> Checklist - [ ] Documentation added - [x] Tests updated - [x] Is this an important fix or new feature? Add an entry in the `CHANGELOG.md`. - [ ] Changes that require user attention or interaction to upgrade are documented in `docs/sources/upgrading/_index.md`	3 years ago
李国忠	a290549a59	Promtail: Add support for max stream limit (#7101 ) What this PR does / why we need it: Add Promtail support to a new feature/configuration named `max_stream_limit` * This new feature limits the max number of streams Promtail can materialize. The new feature is also useful to avoid OOMing Promtail Which issue(s) this PR fixes: Fixes #6968	3 years ago
Pablo	76c1198d58	Improvements to Promtail GCP push target (#7057 ) <!-- Thanks for sending a pull request! Before submitting: 1. Read our CONTRIBUTING.md guide 2. Name your PR as `<Feature Area>: Describe your change`. a. Do not end the title with punctuation. It will be added in the changelog. b. Start with an imperative verb. Example: Fix the latency between System A and System B. c. Use sentence case, not title case. d. Use a complete phrase or sentence. The PR title will appear in a changelog, so help other people understand what your change will be. 3. Rebase your PR if it gets out of sync with main --> What this PR does / why we need it: This PR includes some pending improvements to newish promtail GCP push target: - Pending low hanging optimization for replacing a regex that gets evaluated on every incoming request from GCP, for a simpler solution. [Related comments](https://github.com/grafana/loki/pull/6777#discussion_r935804383) - Some small refactors. See: [this](https://github.com/grafana/loki/pull/6777#pullrequestreview-1059064564) - Add some validations over the payload the target receives. Payload details are [here](https://cloud.google.com/pubsub/docs/push). This PR adds just validation over the log line, the incoming GCP message ID to add some traceability if needed, and the subscription name because it can be used as label. - Add `subscription` GCP payload field as label, which includes the PubSub subscription identifier. - Add `cause` label to error metrics. Since there's multiple failure point this might help debugging / notice strange behaviours. Also, by running the push target in some dev workloads, we noticed it has some labels different to the ones added in the pull target. This PR also refactors the logic to re-use the same GCP log entry parsing logic in both targets. That way the labels exposed to the user in the push target are $Labels_{pull} \cup Labels_{push}$. Which issue(s) this PR fixes: Related to: https://github.com/grafana/cloud-onboarding/issues/2067 Special notes for your reviewer: This PR is a follow up [this other one](https://github.com/grafana/loki/pull/6777). It works on some comments made after merge, and adds some changes we saw necessary upon running this target in real environments. <!-- Note about CHANGELOG entries, if a change adds: * an important feature * fixes an issue present in a previous release, * causes a change in operation that would be useful for an operator of Loki to know then please add a CHANGELOG entry. For documentation changes, build changes, simple fixes etc please skip this step. We are attempting to curate a changelog of the most relevant and important changes to be easier to ingest by end users of Loki. Note about the upgrade guide, if this changes: * default configuration values * metric names or label names * changes existing log lines such as the metrics.go query output line * configuration parameters * anything to do with any API * any other change that would require special attention or extra steps to upgrade Please document clearly what changed AND what needs to be done in the upgrade guide. --> Checklist - [x] Documentation added - [x] Tests updated - [ ] Is this an important fix or new feature? Add an entry in the `CHANGELOG.md`. - [ ] Changes that require user attention or interaction to upgrade are documented in `docs/sources/upgrading/_index.md` Co-authored-by: Kaviraj Kanagaraj <kavirajkanagaraj@gmail.com>	3 years ago
Trevor Whitney	c0eb73c34a	Enable build in env w/o internet or GCP creds (#7029 ) What this PR does / why we need it: The build is currently dependent upon you building in an environment with internet connectivity and GCP credentials. This PR fixes the few parts of our code and test that rely on these external dependencies so it can be built in an environment without internet or GCP credentials (such as when building a [nix flake](https://github.com/grafana/loki/pull/7020)) Which issue(s) this PR fixes: Fixes #7028	3 years ago
Callum Styan	4a21923913	update to alpine:3.6.2 (#6979 ) Updates the alpine image version we use to 3.16.2 and will also build/publish a new loki build image Signed-off-by: Callum Styan <callumstyan@gmail.com> Signed-off-by: Callum Styan <callumstyan@gmail.com>	3 years ago
Christian Simon	88e2705034	Update alpine and golang base images (#6897 ) * Upgrade base images to alpine 3.15.6 everywhere https://alpinelinux.org/posts/Alpine-3.13.12-3.14.8-3.15.6-3.16.2-released.html * Update golang images to 1.18.5	3 years ago
carlospeon	171a09b822	Add support for journal matches (only conjuntions) (#6656 )	3 years ago
Kaviraj Kanagaraj	e8b44172d8	test(promtail): Fix flaky test with promtail bind port. (#6859 ) * test(promtail): Fix flaky test with promtail bind port. You cannot run promtail tests if there is already running promtail in your local. e.g: ``` $ go test -run TestPromtail ./clients/pkg/promtail/ ``` It gives you error ``` --- FAIL: TestPromtail (0.02s) promtail_test.go:112: error creating promtail listen tcp 127.0.0.1:9080: bind: address already in use FAIL FAIL github.com/grafana/loki/clients/pkg/promtail 0.043s ``` Cause for the error is, test uses static default port (the same we use for promtail normally) in tests. This is not too annonying if seen in isolation. But mainly annoying in two cases particularly 1. Hard to run tests parallely in general with these kind of static ports 2. These are causing some real problems in some integeration tests on GEL which starts docker-compose to run some test suites and even stopping containers won't release ports soon enough making CI fail randomly. This PR takes advantage of Go idiom of passing `port=0` to http (and also grpc) servers enforce Go's `net` package to bind some unused random ports. Making each test run choose random ports enabling easy running theses tests in parallel. NOTE: This PR also contains upgrade in `common/server` package that made this fix possible. https://github.com/weaveworks/common/pull/249 Signed-off-by: Kaviraj <kavirajkanagaraj@gmail.com> * go mod tidy Signed-off-by: Kaviraj <kavirajkanagaraj@gmail.com>	3 years ago
Alexandre de Sá	6b47aab2de	promtail/client: add bot fields (#6828 ) * chore: add bot fields * Update CHANGELOG.md * chore: update cloudflare configuration section URI * docs: include BotScore and BotScoreSrc to all set of fields * Update CHANGELOG.md	3 years ago
Pablo	c9ef0621d1	Promtail: Add new target for receiving GCP PubSub Push messages (#6777 )	3 years ago
Kaviraj Kanagaraj	0d99d80acd	upgrade `go:1.17.9` to `go:1.18.4` (#6726 )	3 years ago
Pablo	d4cdc37682	promtail: Inject tenant ID when receiving X-Scope-OrgID in heroku target (#6695 ) * inject tenant ID header * fix import order	3 years ago
Pablo	0e28452f1f	Targets: Add Heroku HTTPS drain target (#6448 )	3 years ago
Dylan Guedes	f17d3d768c	Promtail: Add text encoding conversion to file targets (#6395 ) Adds support for text encoding conversion for file targets. To use it, add `encoding: <encoding_name>` into the scrapeconfig.	3 years ago
Johan Fleury	9b980c45ed	Promtail: update certificates used for syslog’s tests (#6534 ) All the certificates now have a 20 year lifetime. This reverts commit `2fc20af9a4`.	3 years ago
Ed Welch	2fc20af9a4	Promtail: disable syslog tests which have expired certs (#6520 ) * disable tests which are failing from hardcoded certs which are now expired. Signed-off-by: Edward Welch <edward.welch@grafana.com> * lint Signed-off-by: Edward Welch <edward.welch@grafana.com>	3 years ago
Tristan Colgate-McFarlane	21f69643b3	promtail: fix infinite loop in file target (fixes #5350 ) (#6135 )	3 years ago

1 2 3 4

178 Commits (9c2da08ef5d7208a104e5f7665db79e5f8340f4e)