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

Commit Graph

Author	SHA1	Message	Date
Christian Haudum	d4a6b159fa	chore: Cleanup label extraction expression (#8596 ) The aim of this is to remove redundancy from the parser grammar and make the expression name more descriptive. Signed-off-by: Christian Haudum <christian.haudum@gmail.com>	2 years ago
Bayan Taani	68a9fd6c0a	Add logfmt selective label extraction (#6675 ) <!-- 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 introduces extracting labels from a log line in `logfmt`, with an extra option to rename them. For example, this query: ``` {app=”foo”} \| logfmt msg,status ``` will extract the labels `msg` and `status` from the following logfmt line: ``` level=error ts=2021-02-12T19:18:10.037940878Z caller=client.go:294 component=client host=observability-loki-gateway msg="final error sending batch" status=400 error="server returned HTTP status 400 Bad Request (400): entry with timestamp 2021-02-12 19:18:08.917452 +0000 UTC ignored, reason: 'entry out of order' for stream..." ``` With the following results: ``` msg="final error sending batch" status=”400” ``` -------------- Another possible scenario with label renaming: ``` {app=”foo”} \| logfmt message="msg", status ``` That produces the following results: ``` message="final error sending batch" status=”400” ``` Which issue(s) this PR fixes: Fixes #3355 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 - [x] Documentation added - [x] Tests updated - [x] Is this an important fix or new feature? Add an entry in the `CHANGELOG.md`. - [x] Changes that require user attention or interaction to upgrade are documented in `docs/sources/upgrading/_index.md` --------- Signed-off-by: Christian Haudum <christian.haudum@gmail.com> Co-authored-by: Christian Haudum <christian.haudum@gmail.com>	2 years ago
Tristan Morgan	2e11ff00eb	Replace deprecated inet.af/netaddr (#8560 ) This removes the deprecated `inet.af/netaddr` that imports `go4.org/unsafe/assume-no-moving-gc` and panic when it's not updated. Fixes #8399 and replaces #8495	2 years ago
李国忠	322783e3d8	LogQL: [optimization] syntax: Replace "panic" in "/pkg/logql/syntax" with "error" (#7208 )	2 years ago
Kaviraj Kanagaraj	807aa459d4	logql: Support `urlencode` and `urldecode` template functions (#8271 ) Signed-off-by: Kaviraj <kavirajkanagaraj@gmail.com>	2 years ago
Ed Welch	c8760f31e2	Loki: Remove validation of matchers when performing label values lookup. (#8224 ) the curren validation fails on != style matchers when they are the only matchers present, requiring at least one `=` matcher. However for looking up label values in TSDB, by design it uses a `desired_label != ""` style query, as such it should be allowed to also include these as matchers. Additional matchers on this Label Value queries typically only exist with LBAC headers adding additional labels. Signed-off-by: Edward Welch <edward.welch@grafana.com> 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 - [ ] Changes that require user attention or interaction to upgrade are documented in `docs/sources/upgrading/_index.md` Signed-off-by: Edward Welch <edward.welch@grafana.com> Signed-off-by: Christian Haudum <christian.haudum@gmail.com> Co-authored-by: Owen Diehl <ow.diehl@gmail.com> Co-authored-by: Danny Kopping <danny.kopping@grafana.com> Co-authored-by: Christian Haudum <christian.haudum@gmail.com>	2 years ago
Owen Diehl	a7ea9c1422	Revert "fix(logql): remove __error_details__ label on drop __error__ or other way around" (#8200 ) Reverts grafana/loki#8150 I was talking with @adityacs about reverting this PR and found the following: Looking at why tests don’t keep `__error_details__` when `__error__` is dropped and I found this line of code: ```go // LabelsResult returns the LabelsResult from the builder. // No grouping is applied and the cache is used when possible. func (b *LabelsBuilder) LabelsResult() LabelsResult { // unchanged path. if len(b.del) == 0 && len(b.add) == 0 && b.err == "" { return b.currentResult } return b.toResult(b.labels()) }``` https://github.com/grafana/loki/blob/main/pkg/logql/log/labels.go#L314 We only add err details when err isn’t empty anyway, so I guess we’ll revert for simplicity but just keep the same functionality	2 years ago
Aditya C S	cf23194d05	fix(logql): remove __error_details__ label on drop __error__ or other way around (#8150 )	2 years ago
Danny Kopping	b9a9cdf2f7	Fixing test regression (#8130 ) Regression introduced by #7723 which was merged around the same time Signed-off-by: Danny Kopping <danny.kopping@grafana.com>	2 years ago
Cyril Tovena	2861c0036f	Improve the JSON parser performance. (#7723 ) Sorry I couldn't help myself and worked a bit on the JSON parser by using another library that allows way less allocations. On top of this, the parser code is also much simpler to understand. I realize that we could possibly also intern/cache (limited obviously) extracted JSON value per key. I would do it per key because otherwise, a high cardinality key can take over the intern/cache which will make it worthless. Happy to walk someone through the idea, I think this would be a big jump too. Would be great to run and test this in our dev environment using some JSON queries before releasing it. ``` ❯ benchstat before.txt after.txt name old time/op new time/op delta _Parser/json/no_labels_hints-10 2.53µs ± 0% 2.16µs ± 2% -14.59% (p=0.008 n=5+5) _Parser/json/labels_hints-10 1.93µs ± 0% 1.47µs ± 0% -23.96% (p=0.008 n=5+5) name old alloc/op new alloc/op delta _Parser/json/no_labels_hints-10 656B ± 0% 280B ± 0% -57.32% (p=0.008 n=5+5) _Parser/json/labels_hints-10 512B ± 0% 176B ± 0% -65.62% (p=0.008 n=5+5) name old allocs/op new allocs/op delta _Parser/json/no_labels_hints-10 46.0 ± 0% 18.0 ± 0% -60.87% (p=0.008 n=5+5) _Parser/json/labels_hints-10 39.0 ± 0% 12.0 ± 0% -69.23% (p=0.008 n=5+5) ```	2 years ago
Aditya C S	8df5803d9a	feat(logql): Support drop labels in logql pipeline (#7975 ) This PR introduces `drop` stage in logql pipeline. Fixes #7870, Fixes #7368	2 years ago
Aditya C S	a08a7324f4	fix(logql): fix caseinsensitive search in filter (#8037 ) What this PR does / why we need it: Fix caseinsensitive search in filter Which issue(s) this PR fixes: Fixes #7904 #7837 Checklist - [x] 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`	2 years ago
Michel Hollands	59eff5ca58	Add count to template functions (#7951 ) What this PR does / why we need it: Add a `count` template function to count occurrences of a substring. For example if `label="abc abd abc"` ``` {{ .label \| count "abc" }} ``` would return 2. Signed-off-by: Michel Hollands <michel.hollands@grafana.com> Co-authored-by: Danny Kopping <dannykopping@gmail.com>	2 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
Robert Fratto	2208ea7147	Update Prometheus dependency to latest release (v2.39.1) (#7425 )	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
Vladyslav Diachenko	615941234a	Logql Analyzer (#6171 ) * created logql-debugger page and API endpoint that returns information for each stage of the query * added docker-compose setup for local testing * refactored analyzer and added unit tests * removed docs page to release it in a separate PR when back-end API is published * changes structure of results object in response to return origin log line needed for case when only stream selector is sent in query * merged the latest main and resolved conflicts * added endpoint for readiness probe and configured CI * fixed drone.yml discrepancy * fixed method signature * fixed path to Dockerfile * added clean step	3 years ago
Ed Welch	6d54505485	Loki: add `__line__` and `__timestamp__` to label_format (#6983 ) <!-- 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: We have the `__line__` and `__timestamp__` functions currently available in the `line_format` stage but not in the `label_format` stage. This PR fixes that. Specifically `__timestamp__` can be valuable in a label because you can do things like ``` label_format hour_of_day=`{{__timestamp__ \| date "15"}}` ``` To be able to conditionaly select log lines based on the hour of the day they were created, this can be useful for conditional alerting as an example. Which issue(s) this PR fixes: Fixes #6402 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 - [x] Documentation added - [ ] 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` Signed-off-by: Edward Welch <edward.welch@grafana.com>	3 years ago
Christian Simon	ee2fe6514d	Propagate error detail properly in all cases (#6890 ) The error handling was missing appending the error detail properly in all cases. Fixes #6265	3 years ago
Kaviraj Kanagaraj	7b89f20378	fix(logql): Make `LabelSampleExtractor` work as expected when label doesn't exist in logline (#6766 ) * fix(logql): Make `StringLabelfilter` work as expected Consider Log lines ``` msg=hello subqueries=5 msg=hello subqueries=0 msg=hello ``` If you run following logql ``` {foo="bar"}\|logfmt\|subqueries!=0 ``` it returns ``` msg=hello subqueries=5 ``` But if you run following logql ``` {foo="bar"}\|logfmt\|subqueries!="0" # NOTE: "0" as string. ``` it returns ``` msg=hello subqueries=5 msg=hello ``` NOTE: it also returns log lines that doesn't contain `subqueries` label in the first place. This cause subtle error if used in metric queries using that `label` in unwrap. e.g: following logql on above log lines returns ``` sum_over_time({foo="bar"}\|logfmt\|subqueries!="0"\|unwrap subqueries[1m]) ``` Returns `pipeline error: 'SampleExtractionErr'`. Because, the lines without `subqueries` labels are not skipped during `ProcessLine` and during metric extraction, our extractor sets this `SampleExtractionerror`. Fixes: https://github.com/grafana/loki/issues/6713 * Add it on changelog Signed-off-by: Kaviraj <kavirajkanagaraj@gmail.com> * Fix linter Signed-off-by: Kaviraj <kavirajkanagaraj@gmail.com> * Add more tests and fix the edge case with equality to empty strings Signed-off-by: Kaviraj <kavirajkanagaraj@gmail.com> * typo on the comment Signed-off-by: Kaviraj <kavirajkanagaraj@gmail.com> * handle this edge case during metric extraction. Not in label filter Signed-off-by: Kaviraj <kavirajkanagaraj@gmail.com> * Fix linter Signed-off-by: Kaviraj <kavirajkanagaraj@gmail.com> * Remove commented code Signed-off-by: Kaviraj <kavirajkanagaraj@gmail.com> * Update changelog Signed-off-by: Kaviraj <kavirajkanagaraj@gmail.com>	3 years ago
Ed Welch	1a75bb8e22	Loki: Return an __error_details__ label for any line which incurs a __error__ while being processed (#6543 ) * adds an __error_details__ label which includes any details available about a logql error. Signed-off-by: Edward Welch <edward.welch@grafana.com> * fix lint, changed a test a little Signed-off-by: Edward Welch <edward.welch@grafana.com> * reset error details in Reset() function Signed-off-by: Edward Welch <edward.welch@grafana.com>	3 years ago
Ed Welch	33761194d9	Loki: Add the ability to use the log line timestamp in label and line formatter (#6517 ) * Change Stage interface Process method to include the timestamp of the line Signed-off-by: Edward Welch <edward.welch@grafana.com> * allow using log line timestamp in formatter Signed-off-by: Edward Welch <edward.welch@grafana.com> * update formatting in test Signed-off-by: Edward Welch <edward.welch@grafana.com> * update docs Signed-off-by: Edward Welch <edward.welch@grafana.com> * Update docs/sources/logql/template_functions.md Co-authored-by: Karen Miller <84039272+KMiller-Grafana@users.noreply.github.com> Co-authored-by: Karen Miller <84039272+KMiller-Grafana@users.noreply.github.com>	3 years ago
Cyril Tovena	8ee0d62d9e	Fixes bad initialization for uniqueString (#6432 )	3 years ago
Jesus Vazquez	db9dbe8e1d	Bump prometheus dependency (#6403 ) * Bump prometheus dependency We had to add a replace of prometheus in go mod because thanos is importing an older version and then replacing it. * Undo textparser.New variable mismatch * modtimevfs path update * Fix labels withlabels and withoutlabels usages WithoutLabels and WithLabels methods have dissappeared. Now labels.NewBuilder has to be used. * Fix RuleManager interface for Update() Update now accepts a new param called ruleGroupPostProcessFunc which is an optional function to be executed after rule group processing. In this case I've set it to nil in the usages i've found. * NewQueryRange now accepts QueryOpts The NewQueryRange function now accepts QueryOptions so I had to update all the calls. I've set the opts to nil since we're not using them. * gofmt	3 years ago
Mathew Heard	771e218b58	JSON label extraction field validation expression overly strict (#6372 ) * Extend json label extraction test * Add json label extraction to aggregation * fix lexeer issue, it's value to have numbers inside the field (just not at the start) Also simplified logic * integrate feedback from reviews	3 years ago
Cyril Tovena	7e53381eee	Avoid line copy during LogQL line_format (#6104 )	3 years ago
Cyril Tovena	67e7285bf5	Improve labels computation during LogQL pipeline (#6110 ) * Improve labels computation during LogQL pipeline * Fixes a bug that was mutating the base labels	3 years ago
Justin Burnham	97c7d7ca3c	LogQL: add the `default` sprig template function to logql label/line formatter (#6163 ) * add the `default` sprig template function to logql label/line formatter With the `default` sprig template function, we will be able to make it easier to output a default value if the source string is otherwise empty. Useful for json fields that can be missing, like HTTP headers in a log line that aren't required as in the following: ``` {job="access_log"} \| json \| line_format `{{.http_request_headers_x_forwarded_for \| default "-"}}` ``` Previously this could only be done (that I found) with the much more verbose ``` {{regexReplaceAllLiteral "^$" .http_request_headers_x_forwarded_for "-"}} ``` * update changelog entry * add pr number * enhance docs * Update docs/sources/logql/template_functions.md Co-authored-by: Kaviraj Kanagaraj <kavirajkanagaraj@gmail.com> * Update docs/sources/logql/template_functions.md Co-authored-by: Vladyslav Diachenko <82767850+vlad-diachenko@users.noreply.github.com> Co-authored-by: Kaviraj Kanagaraj <kavirajkanagaraj@gmail.com> Co-authored-by: Owen Diehl <ow.diehl@gmail.com> Co-authored-by: Vladyslav Diachenko <82767850+vlad-diachenko@users.noreply.github.com>	3 years ago
Owen Diehl	93de7a7061	renames skip -> ok in ProcessString (#6064 ) * renames skip -> ok in ProcessString Signed-off-by: Owen Diehl <ow.diehl@gmail.com> * streampipeline matches convention Signed-off-by: Owen Diehl <ow.diehl@gmail.com>	3 years ago
Cyril Tovena	986fe5cf05	Remove empty match within concat of regex (#6026 ) * Remove empty match within concat of regex * add back tests	3 years ago
Travis Patterson	5939d5e9d8	Query filtering in the ingester and storage (#5629 ) * Add timestamp to pipeline and extractor interfaces * implement filter extractor and pipeline * implement storage filtering * implement ingester filtering * test cleanup * Refactor pipeline setup to util * linter * enable filtering based on mode * review feedback * review comments * review feedback * no need to include label filters in the test pipeline * review feedpack	3 years ago
Cyril Tovena	79f0e349f5	Refactor LogQL syntax into its own package (#5539 ) * Refactor LogQL syntax into its own package Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * Merge upstream Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * Update pkg/loghttp/series.go Co-authored-by: Susana Ferreira <ssncferreira@gmail.com> * lint Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * lint Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> Co-authored-by: Susana Ferreira <ssncferreira@gmail.com>	3 years ago
Bryan Boreham	a50cac7674	filters: use faster regexp package (#5315 ) * regexp-filter: improve benchmark Add a few more cases based on real-world usage. Also simplify the loop to just run the number of times requested. Go benchmarks run 'N' times to reach a duration, default 1 second. Previously the benchmark was running N1000000 times, which took 7 seconds minimum for some cases. regexp filter: use modified package with optimisations See https://github.com/grafana/regexp/tree/speedup#readme Includes the following changes proposed upstream: * [regexp: allow patterns with no alternates to be one-pass](https://go-review.googlesource.com/c/go/+/353711) * [regexp: speed up onepass prefix check](https://go-review.googlesource.com/c/go/+/354909) * [regexp: handle prefix string with fold-case](https://go-review.googlesource.com/c/go/+/358756) * [regexp: avoid copying each instruction executed](https://go-review.googlesource.com/c/go/+/355789) * [regexp: allow prefix string anchored at beginning](https://go-review.googlesource.com/c/go/+/377294) * Add grafana/regexp to vendor directory	3 years ago
Cyril Tovena	e3b74158a5	Fixes log deduplication when mutating Labels using LogQL (#5289 ) * Changes the iterator interface to include labelshash. Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * Improve unused args and fixes build Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * Implement the new hash in the memchunk. Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * Add the hash in logproto Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * Insert the hash from ingester-querier Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * Rename LabelsHash to StreamHash Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * Working through fixing tests. Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * Refactor HeapIterator into merge and sort Iterator. Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * lint Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * Order alphabetically when ordering samples. Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * Do not JSON encode the hash for legacy model. Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * Add more regression tests. Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * Update CHANGELOG.md Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * Review feedback	3 years ago
Vladyslav Diachenko	ea37f4595d	added upgrade notes (#5311 )	3 years ago
Cyril Tovena	21e61315df	LogQL: add __line__ function to \| line_format template (#4879 ) * LogQL: add __line__ function to \| line_format template Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * add changelog entry Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * Update docs/sources/logql/template_functions.md Co-authored-by: Karen Miller <84039272+KMiller-Grafana@users.noreply.github.com> * Clone the function map to avoid a race between multiple expr Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> Co-authored-by: Karen Miller <84039272+KMiller-Grafana@users.noreply.github.com> Co-authored-by: Owen Diehl <ow.diehl@gmail.com>	3 years ago
Cyril Tovena	0a944670ef	Improve LogQL format stages requireLabel (#4769 ) * Improve LogQL format stages. This only includes required labels. ``` ➜ go test -benchmem -run=^$ -bench ^Benchmark_Pipeline ./pkg/logql/log -v -count 5 > after.txt && benchstat before.txt after.txt name old time/op new time/op delta _Pipeline/pipeline_bytes-16 13.8µs ± 6% 11.5µs ± 1% -16.42% (p=0.008 n=5+5) _Pipeline/pipeline_string-16 13.9µs ± 2% 11.5µs ± 2% -17.27% (p=0.008 n=5+5) _Pipeline/line_extractor_bytes-16 14.0µs ± 4% 11.6µs ± 0% -17.45% (p=0.016 n=5+4) _Pipeline/line_extractor_string-16 13.5µs ± 0% 11.5µs ± 0% -14.67% (p=0.008 n=5+5) _Pipeline/label_extractor_bytes-16 13.7µs ± 1% 11.7µs ± 1% -14.69% (p=0.008 n=5+5) _Pipeline/label_extractor_string-16 14.1µs ±11% 11.6µs ± 0% -17.89% (p=0.008 n=5+5) name old alloc/op new alloc/op delta _Pipeline/pipeline_bytes-16 9.51kB ± 0% 7.50kB ± 0% -21.19% (p=0.008 n=5+5) _Pipeline/pipeline_string-16 9.51kB ± 0% 7.50kB ± 0% -21.19% (p=0.008 n=5+5) _Pipeline/line_extractor_bytes-16 9.51kB ± 0% 7.50kB ± 0% -21.19% (p=0.008 n=5+5) _Pipeline/line_extractor_string-16 9.51kB ± 0% 7.50kB ± 0% ~ (p=0.079 n=4+5) _Pipeline/label_extractor_bytes-16 9.51kB ± 0% 7.50kB ± 0% -21.18% (p=0.008 n=5+5) _Pipeline/label_extractor_string-16 9.51kB ± 0% 7.50kB ± 0% -21.18% (p=0.008 n=5+5) name old allocs/op new allocs/op delta _Pipeline/pipeline_bytes-16 46.0 ± 0% 45.0 ± 0% -2.17% (p=0.008 n=5+5) _Pipeline/pipeline_string-16 46.0 ± 0% 45.0 ± 0% -2.17% (p=0.008 n=5+5) _Pipeline/line_extractor_bytes-16 46.0 ± 0% 45.0 ± 0% -2.17% (p=0.008 n=5+5) _Pipeline/line_extractor_string-16 46.0 ± 0% 45.0 ± 0% -2.17% (p=0.008 n=5+5) _Pipeline/label_extractor_bytes-16 46.0 ± 0% 45.0 ± 0% -2.17% (p=0.008 n=5+5) _Pipeline/label_extractor_string-16 46.0 ± 0% 45.0 ± 0% -2.17% (p=0.008 n=5+5) ``` Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * rollback model change Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * remove unused variable Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com>	4 years ago
Marco Pracucci	c53457feb9	Upgrade Cortex, Prometheus and Thanos (#4830 ) * Upgrade Cortex, Prometheus and Thanos Signed-off-by: Marco Pracucci <marco@pracucci.com> * Fixed imported ordering Signed-off-by: Marco Pracucci <marco@pracucci.com> * Fixed data type used for series ref when recovering from checkpoint/WAL Signed-off-by: Marco Pracucci <marco@pracucci.com> * Upgraded Thanos to fix arm build Signed-off-by: Marco Pracucci <marco@pracucci.com> * Run go mod tidy Signed-off-by: Marco Pracucci <marco@pracucci.com>	4 years ago
Cyril Tovena	45ed6d560d	Fix up a merge conflict on log filter (#4815 ) * Fix up a merge conflict on log filter Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * use the same parameter order as the stdlib Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com>	4 years ago
Hiroki Sakamoto	b1b82a6000	[Hot Fix] Merge improving case sensitive search correctly (#4804 ) * Improve case sensitive search to avoid allocations. ``` ❯ benchcmp before.txt after.txt benchmark old ns/op new ns/op delta Benchmark_LineFilter/default_true_(?i)foo-16 2400 2233 -6.96% Benchmark_LineFilter/simplified_true_(?i)foo-16 201 228 +13.13% Benchmark_LineFilter/default_false_(?i)foo-16 2443 2376 -2.74% Benchmark_LineFilter/simplified_false_(?i)foo-16 185 231 +24.96% benchmark old allocs new allocs delta Benchmark_LineFilter/default_true_(?i)foo-16 0 0 +0.00% Benchmark_LineFilter/simplified_true_(?i)foo-16 1 0 -100.00% Benchmark_LineFilter/default_false_(?i)foo-16 0 0 +0.00% Benchmark_LineFilter/simplified_false_(?i)foo-16 1 0 -100.00% benchmark old bytes new bytes delta Benchmark_LineFilter/default_true_(?i)foo-16 0 0 +0.00% Benchmark_LineFilter/simplified_true_(?i)foo-16 128 0 -100.00% Benchmark_LineFilter/default_false_(?i)foo-16 0 0 +0.00% Benchmark_LineFilter/simplified_false_(?i)foo-16 128 0 -100.00% ``` It's not much but for a billions line it makes a big difference. Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * Simplified version Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * Cleanup Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> Co-authored-by: Cyril Tovena <cyril.tovena@gmail.com>	4 years ago
Cyril Tovena	dc222dc98d	Improve case insensitive search to avoid allocations. (#4394 ) * Improve case sensitive search to avoid allocations. ``` ❯ benchcmp before.txt after.txt benchmark old ns/op new ns/op delta Benchmark_LineFilter/default_true_(?i)foo-16 2400 2233 -6.96% Benchmark_LineFilter/simplified_true_(?i)foo-16 201 228 +13.13% Benchmark_LineFilter/default_false_(?i)foo-16 2443 2376 -2.74% Benchmark_LineFilter/simplified_false_(?i)foo-16 185 231 +24.96% benchmark old allocs new allocs delta Benchmark_LineFilter/default_true_(?i)foo-16 0 0 +0.00% Benchmark_LineFilter/simplified_true_(?i)foo-16 1 0 -100.00% Benchmark_LineFilter/default_false_(?i)foo-16 0 0 +0.00% Benchmark_LineFilter/simplified_false_(?i)foo-16 1 0 -100.00% benchmark old bytes new bytes delta Benchmark_LineFilter/default_true_(?i)foo-16 0 0 +0.00% Benchmark_LineFilter/simplified_true_(?i)foo-16 128 0 -100.00% Benchmark_LineFilter/default_false_(?i)foo-16 0 0 +0.00% Benchmark_LineFilter/simplified_false_(?i)foo-16 128 0 -100.00% ``` It's not much but for a billions line it makes a big difference. Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * typo Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * Simplified version Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * Cleanup Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com>	4 years ago
Karsten Jeschkies	4c9e4f5f06	Make `` and `+` non-greedy to double regex filter speed. (#4775 ) Optimize regular expressions in filters. * Run original RE2 tests. * Run both tests. * Update pkg/logql/log/filter.go Co-authored-by: Bryan Boreham <bjboreham@gmail.com> * Update pkg/logql/log/filter.go Co-authored-by: Bryan Boreham <bjboreham@gmail.com> * Fix tests. * Format code. Co-authored-by: Bryan Boreham <bjboreham@gmail.com>	4 years ago
Karsten Jeschkies	c19a6c5bbc	Join contain filters into one. (#4528 ) * Join contain filters into one. * Update pkg/logql/log/filter.go Co-authored-by: Dylan Guedes <djmgguedes@gmail.com> * Test Filter method. * Test old and new filter consturctors. * Remove comment. * Define more benchmark cases. * Take care of empty contain filters. * Satisfy linter. * USe different method to filter. * Simplify all filter concat. * Format code. * Push regex filters to the end. * Remove unused test. * Add a few more tests. * Make filters private again. * Fix test failure. * Only accumulate successive contain filters. * Reverse accumulation. * Satisfy linter. * Remove old version. Co-authored-by: Dylan Guedes <djmgguedes@gmail.com>	4 years ago
Garrett	1c7df620c4	add date time sprig template functions in logql label/line formatter (#4603 ) * add date time sprig template functions in logql label/line formatter With date time sprig template functions, we will be able to filter test jobs which create time is 1 day before as the following: ``` {job="test"} \| label_format nowEpoch=`{{(unixEpoch now)}}`,createDateEpoch=`{{unixEpoch (toDate "2006-01-02" .createDate)}}` \| label_format dateTimeDiff="{{sub .nowEpoch .createDateEpoch}}" \| dateTimeDiff > 86400 ``` * add documents for date time sprig template functions * resolve comments * add golang datetime layout doc reference * update CHANGELOG.md accordingly to include this PR	4 years ago
Cyril Tovena	a6c7ecc2fa	Fixes the pattern parser validation. (#4308 ) We should never have 2 consecutive capture, since when we do we can't figure which data goes where. The previous validation was validating two nodes at the time and moving to the next two nodes so in the end it wasn't validating properly. 🤦 Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com>	4 years ago
Cyril Tovena	32e55c3043	Fork cortex chunk storage into Loki. (#3842 ) The chunk package has been forked from this commit for future reference and licensing. `09fe17c0e6`	4 years ago
Kaviraj	bfb24e6a4c	Ip matcher for LogQL (#3986 ) * IP matcher and filter via netaddr.IP package Signed-off-by: Kaviraj <kavirajkanagaraj@gmail.com> * Adding IP filter into label filter gramer 1. ast tests 2. parse tests Signed-off-by: Kaviraj <kavirajkanagaraj@gmail.com> * Label filters all test cases passes Signed-off-by: Kaviraj <kavirajkanagaraj@gmail.com> * Rename ipFilter -> ipLabelFilter Signed-off-by: Kaviraj <kavirajkanagaraj@gmail.com> * More tests Signed-off-by: Kaviraj <kavirajkanagaraj@gmail.com> * Remove testlogs.txt Signed-off-by: Kaviraj <kavirajkanagaraj@gmail.com> * Make linter happy Signed-off-by: Kaviraj <kavirajkanagaraj@gmail.com> * Add more cases to the benchmark Signed-off-by: Kaviraj <kavirajkanagaraj@gmail.com> * PR remarks 1. support NEQ 2. reduce allocation. 3. Minor tweaks Signed-off-by: Kaviraj <kavirajkanagaraj@gmail.com> * Make IP filter work as line filter 1. Add necessary interface implementation 2. More tests for parser and ast Signed-off-by: Kaviraj <kavirajkanagaraj@gmail.com> * Make linter happy Signed-off-by: Kaviraj <kavirajkanagaraj@gmail.com> * Bug with filter matcher Its wierd bug. Not handing string() method breaks the parser :( Signed-off-by: Kaviraj <kavirajkanagaraj@gmail.com> * Split Line and Label filter to different struct. It got out of hands to handle both in same struct. Signed-off-by: Kaviraj <kavirajkanagaraj@gmail.com> * Add basic doc for IP matcher Signed-off-by: Kaviraj <kavirajkanagaraj@gmail.com> * Add docs feedback from Karen * Label filter add patter error report * PR remarks 1. Indendation fix 2. Remove string() interface for ip label filte * Linter happy * PR remaks * Fix ip label matcher string() method if pattern is invalid * Fix label filter tests * Have local copy of `pattern` in label filter to use in string() * Rebase conflicts * ast.go fixes	4 years ago
Jarrod Funnell	267107b75e	Parser: Allow literal control chars in logfmt decoder (#3932 ) * Allow literal control chars in decoder * Ensure decoding control chars in logfmt works	4 years ago
Cyril Tovena	5d8414fcd6	Intern label keys for LogQL parser. (#3990 ) * Intern label keys for LogQL parser. They often will see the same key over and over and instead of allocating/sanitizing we can re-use the previous value. This mostly speeds up parser that were using a dynamic list of keys like json and logfmt. We could intern values too specially for metrics queries. ``` ❯ benchcmp before.txt after.txt benchmark old ns/op new ns/op delta Benchmark_Parser/json/no_labels_hints-16 3894 2885 -25.91% Benchmark_Parser/json/labels_hints-16 2084 2163 +3.79% Benchmark_Parser/unpack/no_labels_hints-16 915 890 -2.72% Benchmark_Parser/unpack/labels_hints-16 848 846 -0.32% Benchmark_Parser/logfmt/no_labels_hints-16 2229 1146 -48.59% Benchmark_Parser/logfmt/labels_hints-16 1230 1150 -6.50% Benchmark_Parser/regex_greedy/no_labels_hints-16 3881 3754 -3.27% Benchmark_Parser/regex_greedy/labels_hints-16 5443 3737 -31.34% Benchmark_Parser/regex_status_digits/no_labels_hints-16 318 290 -8.89% Benchmark_Parser/regex_status_digits/labels_hints-16 335 289 -13.85% Benchmark_Parser/pattern/no_labels_hints-16 151 151 -0.20% Benchmark_Parser/pattern/labels_hints-16 129 131 +1.24% benchmark old allocs new allocs delta Benchmark_Parser/json/no_labels_hints-16 69 43 -37.68% Benchmark_Parser/json/labels_hints-16 32 36 +12.50% Benchmark_Parser/unpack/no_labels_hints-16 19 17 -10.53% Benchmark_Parser/unpack/labels_hints-16 16 17 +6.25% Benchmark_Parser/logfmt/no_labels_hints-16 46 16 -65.22% Benchmark_Parser/logfmt/labels_hints-16 20 16 -20.00% Benchmark_Parser/regex_greedy/no_labels_hints-16 2 2 +0.00% Benchmark_Parser/regex_greedy/labels_hints-16 2 2 +0.00% Benchmark_Parser/regex_status_digits/no_labels_hints-16 2 2 +0.00% Benchmark_Parser/regex_status_digits/labels_hints-16 2 2 +0.00% Benchmark_Parser/pattern/no_labels_hints-16 2 2 +0.00% Benchmark_Parser/pattern/labels_hints-16 1 1 +0.00% benchmark old bytes new bytes delta Benchmark_Parser/json/no_labels_hints-16 1089 624 -42.70% Benchmark_Parser/json/labels_hints-16 456 480 +5.26% Benchmark_Parser/unpack/no_labels_hints-16 569 537 -5.62% Benchmark_Parser/unpack/labels_hints-16 553 537 -2.89% Benchmark_Parser/logfmt/no_labels_hints-16 576 336 -41.67% Benchmark_Parser/logfmt/labels_hints-16 256 336 +31.25% Benchmark_Parser/regex_greedy/no_labels_hints-16 194 194 +0.00% Benchmark_Parser/regex_greedy/labels_hints-16 194 194 +0.00% Benchmark_Parser/regex_status_digits/no_labels_hints-16 52 52 +0.00% Benchmark_Parser/regex_status_digits/labels_hints-16 52 52 +0.00% Benchmark_Parser/pattern/no_labels_hints-16 35 35 +0.00% Benchmark_Parser/pattern/labels_hints-16 32 32 +0.00% ``` Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com> * Move to labels.go and cap the amount of keys that can be interned to 1024. Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com>	4 years ago
Cyril Tovena	e9c0687c15	Reject labels with invalid runes when using implicit extraction parser. (#3983 ) `json` and `logfmt` parser will attempt to extract all possible keys and values as labels. But if a values currently contains the rune error (replace when there's an invalid rune) PromQL parser will fail and the request will be fully rejected. This PR removes those labels values when using those two parsers. For other parser you can just decide to not extract thoses values.( At least that's my choice for now). Unfortunatelly this comes with a small cpu overhead prices: ``` ~/go/src/github.com/grafana/loki utf8-parser* ❯ benchcmp before.txt after.txt benchmark old ns/op new ns/op delta Benchmark_Parser/json/no_labels_hints-16 3647 4210 +15.44% Benchmark_Parser/json/labels_hints-16 2210 2247 +1.67% benchmark old allocs new allocs delta Benchmark_Parser/json/no_labels_hints-16 72 72 +0.00% Benchmark_Parser/json/labels_hints-16 35 35 +0.00% benchmark old bytes new bytes delta Benchmark_Parser/json/no_labels_hints-16 1121 1121 +0.00% Benchmark_Parser/json/labels_hints-16 472 472 +0.00% ``` Fixes #2859 Signed-off-by: Cyril Tovena <cyril.tovena@gmail.com>	4 years ago

1 2

86 Commits (d4a6b159fab621e4072a5173ce082d7f5d54ec13)