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

Commit Graph

Author	SHA1	Message	Date
Karsten Jeschkies	6578a00481	feat: Support usage trackers for received and discarded bytes. (#11840 )	2 years ago
Christian Haudum	2f54f8d5be	chore(blooms): Implement BloomStore as a service (#12044 ) The bloomstore service is instantiated once at startup and used as dependency for both the bloom compactor and the bloom gateway. This will prevent instantiating the store in different ways. Signed-off-by: Christian Haudum <christian.haudum@gmail.com>	2 years ago
Christian Haudum	1f56da206d	refactor: Simplify fingerprint partitioning in bloom gateway client (#12018 ) Use the recently introduced `tokenRangesForInstance()` function to determine the the ranges for each replication set. For this purpose, the function moves into the `bloomutils` packages. This PR eliminates the structs `addrsWithBounds` and `instanceWithFingerprints` (which had somewhat similar usage) and unifies them into `rsWithRanges` that holds both the fingerprint ranges and the assigned GroupedChunkRefs. Signed-off-by: Christian Haudum <christian.haudum@gmail.com>	2 years ago
Owen Diehl	e835b6192f	refactor(blooms): refactors `blockLoadingIter`, exposing more error cases + additional metrics (#12015 )	2 years ago
Owen Diehl	09aa0f80b5	Blooms/contiguous bounds union (#12009 )	2 years ago
Owen Diehl	166229818d	[Blooms] Consistent hashing via tokens for bloomcompactor (#12002 )	2 years ago
Owen Diehl	85f7baaeda	Blooms/integration fixes (#11979 )	2 years ago
Salva Corts	ef40136715	Use TSDB index prefix on blooms directory (#11977 )	2 years ago
Owen Diehl	5d1798f16d	Blooms/misc fixes (#11974 )	2 years ago
Owen Diehl	f158b5bc3e	correctly set block iter when no overlapping blocks are found (#11973 )	2 years ago
Owen Diehl	87ae2efe41	Pull/11950 ammendments (#11972 ) Signed-off-by: Christian Haudum <christian.haudum@gmail.com> Signed-off-by: Owen Diehl <ow.diehl@gmail.com> Co-authored-by: Christian Haudum <christian.haudum@gmail.com>	2 years ago
Salva Corts	543aaab055	(Blooms) Add metrics to compactor (#11966 )	2 years ago
Christian Haudum	5967ceeb43	Replace min/max token with `TokenRange` in bloom ring utilities (#11960 ) This PR replaces min/max token fields with a `TokenRange` field that uses the `Range[uint32]` type. The `Range[uint32]` uses similar semantics as the `FingerprintBounds` we use for fingerprint ranges. Signed-off-by: Christian Haudum <christian.haudum@gmail.com>	2 years ago
Salva Corts	6434df7345	Bloom compactor shuffle-sharding (#11947 )	2 years ago
Owen Diehl	bd12e16339	[Blooms] Use correct table address function (#11955 )	2 years ago
Owen Diehl	3c06b360ec	meta & removal support for bloom compaction (#11941 ) Signed-off-by: Owen Diehl <ow.diehl@gmail.com>	2 years ago
Christian Haudum	1c43991ddc	Remove `bloomcompactor.DayTable` in favour of `config.DayTime` (#11917 ) Both structs shared the same semantics. This PR moves additional functionality from the `DayTable` to the `DayTime` struct. To get the table name of a day (ordinal number of day since unix epoch), call `DayTime.Table()`. Signed-off-by: Christian Haudum <christian.haudum@gmail.com>	2 years ago
Christian Haudum	eb8464a641	Bloom compactor: Load blocks lazily in batches (#11919 ) To avoid loading possibly lots of blocks upfront, this PR introduces lazy loading of blocks in batches using an iterator that loads blocks on demand. Signed-off-by: Christian Haudum <christian.haudum@gmail.com>	2 years ago
Owen Diehl	0bb2574040	makes batchedLoader generic + removes unnecessary interfaces & adapters (#11924 ) While reviewing https://github.com/grafana/loki/pull/11919, I figured it'd be nice to make `batchedLoader` generic so we can reuse it's logic. This let me test it easier and remove a lot of now-unnecessary adapter code (interfaces, types)	2 years ago
Owen Diehl	7560067439	Bloom/running (#11918 ) Includes a few small fixes in order to get the compactor running, including * Wires up i/o from `v1` pkg and the `bloomshipper` pkg in order to archive+write blocks remotely after building * config error handling * fix an error iterating chunks * handles gzipping when loading tsdbs from storage * bypasses internal cache in `BloomTSDBStore`'s internal `indexshipper/storage.Client` to get up to date results, which is easier for prototyping * `ClosableReadSeekerAdapter` --------- Signed-off-by: Christian Haudum <christian.haudum@gmail.com> Signed-off-by: Owen Diehl <ow.diehl@gmail.com> Co-authored-by: Christian Haudum <christian.haudum@gmail.com>	2 years ago
Salva Corts	7a95cb8663	Configurable bloom tokenizer and block settings (#11889 ) What this PR does / why we need it: This PR makes the max block size configurable and wires up the already existing nGramLen and nGramSkip settings. 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`)	2 years ago
Salva Corts	681bb57971	Limit bloom block size (#11878 ) What this PR does / why we need it: This PR limits the size of the blocks created by the compactor. The block builder keeps adding series' blooms to a block until the limit is exceeded, meaning that blocks may grow beyond the configured maximum. This is needed so we avoid having tiny blocks which had space for small blooms but later a bigger blooms didn't fit. Blocks are built lazily: the generator returns an iterator that builds one block at a time. Special notes for your reviewer: The maximum size is currently set to 50 MBs. We will make this configurable on a followup PR. Checklist - [x] Reviewed the [`CONTRIBUTING.md`](https://github.com/grafana/loki/blob/main/CONTRIBUTING.md) guide (required) - [x] Documentation added - [x] 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`) --------- Co-authored-by: Christian Haudum <christian.haudum@gmail.com>	2 years ago
Owen Diehl	d58031e853	Blooms/bloomcompactor wiring (#11914 ) * Moves `router` struct into the `Compactor` struct * Implements period aware `TSDBStore` * Integrates everything together, hooks into `modules.go` * Removes some now unused compactor code * Light refactoring to make types align where appropriate	2 years ago
Owen Diehl	3a1bd41ef5	[Blooms] Routing & wiring for bloom building (#11894 ) A few changes wiring up bloom building logic: * A `Router` as the controlling mechanism for iterating tenants+periods, queueing work for parallel computation, determining which fingerprint ranges and tenants are owned by a replica, and waiting for it's completion * Updates the `SimplerBloomController` to accept ownership ranges, periods, and tenants as arguments for use by the `Router` * Builds a `BloomTSDBStore` struct around the `indexshipper/storage.Client` interface in order to iterate relevant TSDBs/etc	2 years ago
Owen Diehl	ba0538c2a5	[Blooms] tsdb iterator implementation for bloom construction (#11886 ) Builds a lazy iterator over a tsdb file for bloom construction. This is just the library utility, not the integration.	2 years ago
Christian Haudum	c4ac168a8c	BloomStore: Change signature of `FetchBlocks()` to return `[]CloseableBlockQuerier` (#11884 ) What this PR does / why we need it: In order to encapsulate the reference counting on the `BlockDirectory`, the `BloomStore` now returns a slice of `CloseableBlockQuerier` instead of the directory itself. The caller is responsible for closing the returned querier in order to release the reader resource and decrement the ref counter. The PR also renames `ClosableBlockQuerier` to `CloseableBlockQuerier`. --------- Signed-off-by: Christian Haudum <christian.haudum@gmail.com>	2 years ago
Owen Diehl	af66ecec6a	[Blooms] bloomshipper meta integration (#11874 ) Builds on top of https://github.com/grafana/loki/pull/11872, standardizing on the `bloomshipper` types `Meta` and `MetaRef`.	2 years ago
Christian Haudum	73edf7a943	(chore) Bloomshipper: Separate store and client (#11865 ) What this PR does / why we need it: This PR removes the `StoreAndClient` interface that was accepted by the `BloomShipper`. Since the `BloomStore` had to not only implement the `Store` interface, but also the `Client` interface, it caused re-implementation of the same methods in different ways. Now the shipper solely relies on the `Store` interface. See individual commit messages for more context. Tests have been rewritten from scratch and placed in their own respective test files for store and client. --------- Signed-off-by: Christian Haudum <christian.haudum@gmail.com>	2 years ago
Owen Diehl	d413c28fa3	Blooms: Align compactor to shipper (#11855 ) Does some alignment work between the `bloomcompactor` and the `bloomshipper` pkgs. Notably: * Uses `bloomshipper.BlockRef` everywhere (removes old bloomshipper struct * Integrates `v1.FingerprintBounds` in `Ref` struct * `Location` interface to distinguish local paths vs paths in object storage for certain types (`{Meta,Bloom}Ref`s) * Introduces `KeyResolver` interface to generate locations from these structs * Integrates `KeyResolver` into our bloom store. In the future, this will allow us to change key structures across schema boundaries when we want to change\|improve them. * Removes `BlockPath` from `BlockRef` in favor of the new resolving functionality. This is also beneficial because it lets us _calculate_ locations from the pure Ref objects, rather than tie some arbitrary state to them which can change (or not be populated by accident).	2 years ago
Owen Diehl	928e3f34f8	Blooms: Prepare compactor refactor (#11847 ) Prepares `bloomcompactor` pkg for integration with new logic, primarily removes/thins out existing code.	2 years ago
Christian Haudum	4411649a0e	(chore) Bloom shipper: Replace `Keyspace` struct with `v1.FingerprintBounds` (#11839 ) The latter struct has more utility functions to compare and operate on bounds. --------- Signed-off-by: Christian Haudum <christian.haudum@gmail.com>	2 years ago
Owen Diehl	509f6f6b66	Bloom/controller wiring (#11831 ) Functionality & testware for building bloom block plans.	2 years ago
Christian Haudum	6902130c9a	Bloom shipper: Restructure bloom store (#11828 ) What this PR does / why we need it: This PR changes the structure of how the bloom client * The `BloomStore` is the top component interface which implements the `Store` and the `Client` interfaces. The store holds a store entry for each schema period. * The `bloomStoreEntry` implements the `Store` and `Client` interfaces. It holds a bloom client for a single schema period. Additionally, the store entry also exposes a fetcher for metas (and in the future for blocks), which is responsible for getting data from cache or storage (if not available in cache). * The `BloomClient` implement the `Client` interface. The bloom client uses an object client for the schema period of the bloom client. * The `Fetcher` can fetch and cache metas. This structure is very similar to what we use for the chunk store. Note Before implementing `FetchBlocks()` in the `BloomStore`, I want to implement the new `FingerprintBounds` type also in the shipper and bloom gateway code to be consistent across components, as well as to make use of the new utilities exposed by them. The store implementation probably needs some specific test cases around using the correct store entry for different schema period configs. The code however is mostly taken from the chunk store implementation. --------- Signed-off-by: Christian Haudum <christian.haudum@gmail.com>	2 years ago
Owen Diehl	a56b56c8b0	[Bloom] Keyspace aware bloom creation, library utilities (#11810 ) This PR continues work refactoring the bloom compactor pkg and adds a bunch of utilities for keyspace manipulation/mgmt. A lot of the bloom block logic is contained here as a skeleton. This code is not complete, but is fine to merge as we don't use this yet and should be easier to review before it gets too large.	2 years ago
Christian Haudum	ecab33a932	Bloom Gateway: Extract processing logic from bloom worker into separate struct (#11812 ) What this PR does / why we need it: The processor executes a set of tasks. It cleanly separates I/O from logic and therefore is more testable than the current worker implementation. `9602214abf/pkg/bloomgateway/processor.go (L33-L36)` Note: The processor is not used yet in the worker. Certain parts must be refactored when #11810 is merged. --------- Signed-off-by: Christian Haudum <christian.haudum@gmail.com>	2 years ago
Owen Diehl	de4f56e42d	[Blooms] Refactoring bloom compactor to isolate state from logic (#11793 ) This PR refactors a bunch of code to separate state+I/O related complexity from logic so we can test and extend it more easily. It also adds more logic for the bloom generator to use.	2 years ago
Salva Corts	d97ac48681	Add timing metrics for compaction (#11785 ) What this PR does / why we need it: This PR modifies the metrics of the bloom compactor to: - Add new histogram metrics to track the time spent to process compaction cycles, tenants and jobs. - Replace `_succeeded` and `_failed` metrics with a new `_completed` metric with a status label. 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`)	2 years ago
Owen Diehl	06bb20914b	[Blooms] Specs out BloomGenerator (#11755 ) This is my attempt at making more testable implementations of the work done by the `bloomcompactor` pkg. It's largely inspired by what already exists there, but without depending on things like object storage, chunk clients, etc, it's much easier to locally prove behavior. I'm planning to build on top of this, continuing to integrate work already done in this pkg.	2 years ago
Vladyslav Diachenko	a5aa8b315d	[bloom-compactor] downloading chunks in batches (#11649 ) What this PR does / why we need it: Added chunks batches iterator to download chunks in batches instead of downloading all of them at once. Otherwise, when the stream contains a lot of chunks, it can lead to OOM. Special notes for your reviewer: Checklist - [x] Reviewed the [`CONTRIBUTING.md`](https://github.com/grafana/loki/blob/main/CONTRIBUTING.md) guide (required) - [x] Documentation added - [x] 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: Vladyslav Diachenko <vlad.diachenko@grafana.com>	2 years ago
Christian Haudum	6bcac00c70	Add e2e tests for bloom filtering (#11645 ) What? This PR adds an end-to-end integration test for creating bloom filters and using them in the gateway to filter chunks. Why? This test helped to identify bugs in the bloom shipper code that would have taken a very long time to discover in other ways, such as deploying to a dev environment or running a long-running docker compose setup. Notes The following commits of this PR are actual changes to the compactor/gateway code to make e2e test work: * `2dcc761ba0` * `84052ddb34` * `3ba44f83c8` The bloom gateway code path for processing blocks was cleaned up, because it still contained the unused "sequention processing" path of blocks, which was initially kept to verify the callback based processing works alike: * `f55d79b84f` --------- Signed-off-by: Christian Haudum <christian.haudum@gmail.com>	2 years ago
Poyzan	10c88aaefd	Bloom compactor/debug compacting with existing metas (#11638 ) What this PR does / why we need it: This branch was used to debug uploading and getting existing metas correctly. Uncommented functions to clean up archive directories will be addressed once debugging bloom filters is completed. The work was verified in the dev cell. --------- Co-authored-by: Paul Rogers <paul.rogers@grafana.com>	2 years ago
Poyzan	44b817451a	[bloom-compactor] Clean up work to create blooms (#11591 ) What this PR does / why we need it: This PR is the result of debugging sprint on bloom-compactor on a dev cell. With these changes, compactor now can create and upload blooms and relevant meta files from scratch. - Queries tsdb correctly with empty matchers to get all series - Seeks to the beginning of the bloom data file before uploading - Adds uuid strings to local file name, as checksum is not available at the time of directory creation - Creates meta object with all fields populated - Additional logging edit: confirmed that it runs as expected in the dev cell. --------- Signed-off-by: Christian Haudum <christian.haudum@gmail.com> Co-authored-by: Christian Haudum <christian.haudum@gmail.com> Co-authored-by: Paul Rogers <paul.rogers@grafana.com>	2 years ago
Poyzan	599eed7c52	[bloom-compactor] remove BloomCompactorMinTableAge check (#11546 ) What this PR does / why we need it: `BloomCompactorMinTableAge` was added with the idea that we will skip most recent index files from ingesters that are not compacted into TSDB indexes yet. The default value is an hour. This blocks bloom-compacter processing TSBD of the day, because end timestamp is always in the last 15 mins. We can either reduce it something lower than the frequency of TSDB indexes being built (< 15mins). Here I choose to remove it altogether, assuming uncompacted indexes from ingesters will not be processed as a table as there won't be a schema for that table with the [check here](https://github.com/grafana/loki/blob/main/pkg/bloomcompactor/bloomcompactor.go#L268-L272).	2 years ago
Christian Haudum	0a0b7c83ea	Bloomshipper: Use `model.Time` in `MetaRef` and `BlockRef` (#11566 ) The `Ref` of `MetaRef` and `BlockRef` do have start and end timestamps, which are of type `int64`. This however, can lead to the problem that different components might use different precisions (`s`, `ms`, `ns`) for these timestamps. This PR changes the current implementation of the bloom shipper to use `ms` precision timestamps (using `model.Time`) instead of `s`, to match the behaviour that we have to encode ChunkRefs to object store keys. Signed-off-by: Christian Haudum <christian.haudum@gmail.com>	2 years ago
Poyzan	cc450dcaf9	Add labels to bloom tokenizer metrics (#11542 ) What this PR does / why we need it: Bloom-compactor panics for duplicate metrics registration in bloom-tokenizers. This PR adds labels to bloom-tokenizer metrics, adopting same solution in https://github.com/grafana/loki/pull/11536	2 years ago
Christian Haudum	a760cdfa3f	Bloom Compactor: Fix panic at startup if multiple TSDB period configs are defined (#11536 ) ### Bloom Compactor: Fix duplicate metrics registration If there are multiple period configs with TSDB, then each shipper instance needs to register their metrics with a unique static label set, otherwise, the bloom compactor would panic at startup due to duplicate metric registration attempt. ### Bloom compactor: Start indexshipper in RO mode We don't want/need to upload any index files in the bloom compactor, but we only download files. --------- Signed-off-by: Christian Haudum <christian.haudum@gmail.com>	2 years ago
Poyzan	e56e4c92bc	Run compaction over existing bloom blocks (#11486 ) What this PR does / why we need it: Upon changes to TSDB indexes, the existing blooms needs to be amended with newly added chunks, or new series should be added to blocks. This PR compares the existing meta information with the most up to date TSDB information in the job. In case of no changes to indexes, compaction is skipped; if there's any block needs amending, it does that and merges all blocks in fp range; if there's no previous compaction, it compacts from scratch. Special notes for your reviewer: This logic does not support cutting blocks upon topology changes to bloom-compactors. It can create blocks with series outside of the fp range of the compactor. Cutting blocks will be addressed in a follow-up PR.	2 years ago
Paul Rogers	23780f0cd4	Compress bloom data before uploading it to object storage (#11460 ) What this PR does / why we need it: Currently, the block_downloader.go file is preparing to use the UnTarGz function when retrieving a bloom block from object storage. This PR is to put the corresponding compression in for the upload to object storage. 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`)	3 years ago
Christian Haudum	5f4285963c	Avoid reading block files into memory (#11447 ) Bloom blocks in the wild likely exceed the memory resources of a compactor. Therefore we need to make sure that we don't copy block data into memory, but rather write to and read from disk. Signed-off-by: Christian Haudum <christian.haudum@gmail.com> Co-authored-by: Paul Rogers <paul.rogers@grafana.com>	3 years ago
Christian Haudum	c4f5a57bc8	Bloom Compactor: Optimize check for fingerprint ownership (#11389 ) Calling `c.sharding.OwnsFingerprint(tenant, uint64(fingerprint))` for each Series of a TSDB index is very expensive, because it not only creates the tenant's sub-ring but also needs to check the fingerprint against it. Instead, we can pre-calculate the current instance's token ranges and check if the (uint32 converted) fingerprint is contained within these ranges. Signed-off-by: Christian Haudum <christian.haudum@gmail.com>	3 years ago

1 2

66 Commits (4cae003ecedd474e4c15feab4ea2ef435afff83f)