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

Commit Graph

Author	SHA1	Message	Date
Travis Patterson	e89c30ce6b	Fix Flaky Integration Tests (#7360 ) Allow the OS kernel to decide what's an open port instead of trying to guess	3 years ago
Owen Diehl	289190ee26	removes unused tsdb files (#7353 ) These were introduced in an earlier revision of the TSDB work, but are unused now.	3 years ago
Owen Diehl	eaa0919e53	Use labelbuilder in syntax.ParseLabels to remove empty label values (#7355 ) There's a bug in TSDB where empty label values alter the series fingerprints used in chunk addresses, but are stripped out and create different fingerprints in our TSDB based index. I plan to open a PR to Prometheus, but I also prefer normalizing labels here by removing empty label values. According to [Prometheus](https://prometheus.io/docs/concepts/data_model/), > A label with an empty label value is considered equivalent to a label that does not exist. Our bug comes from the Prometheus model `LabelBuilder` stripping out empty label values, [here](https://github.com/grafana/loki/blob/main/vendor/github.com/prometheus/prometheus/model/labels/labels.go#L419-L423). This means that `Labels.NewBuilder({job="foo",bar=""}) => {job="foo"}`. The [labels.Hash()](https://github.com/grafana/loki/blob/main/vendor/github.com/prometheus/prometheus/model/labels/labels.go#L136-L182) function does not skip empty label values, meaning it's easy to end up with different fingerprints for effectively the same series (an empty label value is supposed to be equivalent to a missing label in prometheus, but they generate different label hashes).	3 years ago
Robert Fratto	c657857e95	Update Thanos dependency to latest version (#7348 )	3 years ago
Dylan Guedes	b1d4efab12	Loki: Per-tenant stream sharding (#7311 ) What this PR does / why we need it: - Move stream sharding configuration to its own package to avoid cyclic imports - Change stream sharding to be a per-tenant configuration - Change ingesters to reject whole streams due to rate-limit based on per-tenant stream sharding - Change stream sharding flags prefix from `distributor.shard-stream` to `shard-stream`	3 years ago
Travis Patterson	e3e4e27c8c	Implement ingester API to return stream rates (#7176 ) This PR introduces an API on the Ingester that reports all current streams in memory and their current rate in bytes/second. The reponse from the API contains a slice of `StreamRate` structs: ```go type StreamRate struct { StreamHash uint64 `protobuf:"varint,1,opt,name=streamHash,proto3" json:"streamHash,omitempty"` StreamHashNoShard uint64 `protobuf:"varint,2,opt,name=streamHashNoShard,proto3" json:"streamHashNoShard,omitempty"` Rate int64 `protobuf:"varint,3,opt,name=rate,proto3" json:"rate,omitempty"` } ``` In the interest of space on the caller and in transmission, only hashes and rates are sent. - `StreamHash` will be used to deduplicate replicas - `StreamHashNoShard` will be used to aggregate a sharded stream's total rate. Co-authored-by: Dylan Guedes <djmgguedes@gmail.com>	3 years ago
Dylan Guedes	da7acb49e5	Loki: Add sharding support for negative/zeroed desired rate (#7342 ) What this PR does / why we need it: Adds the handling of negative/zeroed desired rate to our distributor stream sharding. Without this, invalid desired rates results in a -Inf number of shards. This also adds a log line to report that such a thing happened. Signed-off-by: Danny Kopping <danny.kopping@grafana.com> Co-authored-by: Danny Kopping <danny.kopping@grafana.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
Sandeep Sukhani	10d3fa0029	mmap tsdb index file while opening it for querying (#7338 ) What this PR does / why we need it: In PR #7321, we re-introduced mmap for reading tsdb index, but it does not cover all the paths where we open the index. This PR takes care of it. I have also removed the code for decompressing the index file since we always get a decompressed file from the `indexshipper`.	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
Periklis Tsirakidis	e9419747c9	Fix internal server bootstrap for query frontend (#7328 ) Adds a fix for setting up the internal server module	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
Florian Lehner	4861914c81	pkg/ruler/rulestore/configdb: Preallocate memory (#7326 ) Preallocate memory instead of enforcing an incremental growth. This will result in less work for the garbage collector.	3 years ago
Owen Diehl	044d06015d	adds result cache key version comparison metrics (#7323 ) Followup to https://github.com/grafana/loki/pull/7300 which gives more metric visibility into the underlying errors when comparing cache key versions on the queriers vs query frontends.	3 years ago
Owen Diehl	2bfe626605	reintroduce mmap for tsdb indices (#7321 ) This reverts a [previous decision](`b45efd4c2d (diff-e3f5966d58063d2852aa51899b5aae14e65bb05f29fec7302eb982c3d18c60ab)`) which removed `mmap` support when loading TSDB indices. I did this primarily for simplicity, but since then have reneged on my original choice for two reasons: 1) Retention in Loki is variable and can be very long. Requiring we keep all loaded tsdb indices in memory creates a cost burden in terms of memory required to hold all queried indices at once. 2) A high number of tenants also exacerbates the memory claims which mmap helps alleviate.	3 years ago
Owen Diehl	2bd9c91564	Bugfix include cache generations on querier response (#7300 ) We check the cache number (incremented by deletes) after executing a query with the value returned by the queriers which executed it. If they’re different, we know we can’t safely cache. However, we’re never actually returning a number from the queriers. This means that as soon as this cache number is incremented by a delete, it’ll disable the results cache for that tenant completely. I suspect this has been a latent bug since we forked cortex (goes back to missing some code from this [PR](https://github.com/cortexproject/cortex/pull/2279) into cortex from @sandeepsukhani back in the day!). It just showed itself once we started using deletes.	3 years ago
wilfriedroset	8886800bf6	Add support for username to redis cache configuration (#7270 ) What this PR does / why we need it: Add support for username to redis cache configuration Which issue(s) this PR fixes: Fixes #<issue number> Special notes for your reviewer: Checklist - [x] Reviewed the `CONTRIBUTING.md` guide - [x] Documentation added - [ ] Tests updated - [x] `CHANGELOG.md` updated - [ ] Changes that require user attention or interaction to upgrade are documented in `docs/sources/upgrading/_index.md` Signed-off-by: Wilfried Roset <wilfriedroset@users.noreply.github.com>	3 years ago
wilfriedroset	54f16c008a	Fix typo in flags definitions (#7257 ) What this PR does / why we need it: As all flags defined for `TSDBShipperConfig` are already prefixed by a dot, suffixing tsdb with one makes it redundant. Signed-off-by: Wilfried Roset <wilfriedroset@users.noreply.github.com>	3 years ago
Dylan Guedes	f9f1eddfc5	Loki: Improve query timeouts behavior (#7230 ) What this PR does / why we need it: Improves the different query-related timeouts behavior by fixing a bug and having better defaults. The changes are: - Fix the timeout middleware by creating a new request that will use the new WithDeadline context - `engine:timeout` was being ignored. This PR improves such scenario by using the query timeout as a fallback instead of the default `5min` for all cases	3 years ago
Michel Hollands	8868957783	Log delete request ID upon creation (#7307 ) Log the request ID when a new delete request is created. Signed-off-by: Michel Hollands <michel.hollands@grafana.com>	3 years ago
Bryan Boreham	9498ea2c02	chunks: decode varints directly from byte buffer; stop panicing on some corrupt inputs (#7264 ) What this PR does / why we need it: This is much faster, since we don't pay for byte-at-a-time reads from the buffer. For best performance with gzip and lz4, we still want a bufio.Reader. We don't need the pool of buffered readers now. Other compression formats have their own internal buffers so don't need extra buffering. Checklist - [x] Reviewed the `CONTRIBUTING.md` guide - NA Documentation added - NA Tests updated - [x] `CHANGELOG.md` updated - NA Changes that require user attention or interaction to upgrade are documented in `docs/sources/upgrading/_index.md` (using changes from https://github.com/grafana/loki/pull/7246, ignoring lines that didn't change much) ``` name old time/op new time/op delta Write/ordered-lz4-256k-4 24.1ms ± 7% 22.1ms ± 2% -8.52% (p=0.008 n=5+5) Write/unordered-lz4-4 54.6ms ±29% 49.1ms ± 1% -10.04% (p=0.016 n=5+4) Read/none_66_kB-4 4.16ms ± 6% 2.54ms ± 2% -38.87% (p=0.008 n=5+5) Read/gzip_66_kB-4 177ms ± 4% 161ms ± 4% -9.09% (p=0.008 n=5+5) Read/lz4-64k_66_kB-4 59.5ms ± 2% 49.5ms ± 3% -16.80% (p=0.008 n=5+5) Read/lz4-256k_66_kB-4 62.9ms ± 0% 51.2ms ± 1% -18.58% (p=0.016 n=4+5) Read/lz4-1M_66_kB-4 62.9ms ± 2% 51.5ms ± 1% -18.03% (p=0.008 n=5+5) Read/lz4_66_kB-4 63.7ms ± 4% 51.2ms ± 1% -19.68% (p=0.008 n=5+5) Read/snappy_66_kB-4 51.1ms ± 2% 41.9ms ± 1% -17.95% (p=0.016 n=5+4) Read/flate_66_kB-4 168ms ± 4% 150ms ± 2% -10.49% (p=0.008 n=5+5) Read/zstd_66_kB-4 334ms ±16% 272ms ± 6% -18.52% (p=0.008 n=5+5) Read/none_262_kB-4 3.93ms ± 1% 2.48ms ± 2% -36.91% (p=0.008 n=5+5) Read/gzip_262_kB-4 159ms ± 5% 143ms ± 2% -10.12% (p=0.008 n=5+5) Read/lz4-64k_262_kB-4 53.5ms ± 1% 43.2ms ± 3% -19.35% (p=0.008 n=5+5) Read/lz4-256k_262_kB-4 55.8ms ± 1% 45.3ms ± 2% -18.75% (p=0.008 n=5+5) Read/lz4-1M_262_kB-4 57.5ms ± 2% 45.5ms ± 2% -20.91% (p=0.008 n=5+5) Read/lz4_262_kB-4 56.8ms ± 2% 45.3ms ± 2% -20.26% (p=0.008 n=5+5) Read/snappy_262_kB-4 46.0ms ± 2% 37.5ms ± 1% -18.57% (p=0.008 n=5+5) Read/flate_262_kB-4 153ms ± 3% 134ms ± 2% -12.33% (p=0.008 n=5+5) Read/none_524_kB-4 3.85ms ± 2% 2.46ms ± 2% -36.04% (p=0.008 n=5+5) Read/gzip_524_kB-4 128ms ± 2% 119ms ± 2% -7.64% (p=0.008 n=5+5) Read/lz4-64k_524_kB-4 44.6ms ± 3% 34.7ms ± 1% -22.19% (p=0.008 n=5+5) Read/lz4-256k_524_kB-4 45.4ms ± 1% 36.7ms ± 5% -19.07% (p=0.008 n=5+5) Read/lz4-1M_524_kB-4 47.3ms ± 1% 38.6ms ± 2% -18.36% (p=0.008 n=5+5) Read/lz4_524_kB-4 47.4ms ± 1% 38.0ms ± 1% -19.70% (p=0.008 n=5+5) Read/snappy_524_kB-4 37.3ms ± 1% 31.1ms ± 3% -16.47% (p=0.008 n=5+5) Read/flate_524_kB-4 122ms ± 1% 111ms ± 3% -9.04% (p=0.008 n=5+5) Read/sample_none_66_kB-4 7.81ms ± 1% 6.62ms ± 6% -15.24% (p=0.008 n=5+5) Read/sample_gzip_66_kB-4 237ms ± 2% 219ms ± 1% -7.57% (p=0.008 n=5+5) Read/sample_lz4-64k_66_kB-4 105ms ± 2% 92ms ± 1% -12.03% (p=0.008 n=5+5) Read/sample_lz4-256k_66_kB-4 109ms ± 1% 98ms ± 2% -9.65% (p=0.008 n=5+5) Read/sample_lz4-1M_66_kB-4 113ms ± 8% 98ms ± 2% -13.66% (p=0.008 n=5+5) Read/sample_lz4_66_kB-4 109ms ± 1% 99ms ± 2% -9.29% (p=0.008 n=5+5) Read/sample_snappy_66_kB-4 86.6ms ± 1% 77.2ms ± 6% -10.92% (p=0.008 n=5+5) Read/sample_flate_66_kB-4 236ms ± 8% 214ms ± 3% -9.11% (p=0.008 n=5+5) Read/sample_none_262_kB-4 7.70ms ± 2% 6.34ms ± 2% -17.67% (p=0.008 n=5+5) Read/sample_lz4-64k_262_kB-4 94.7ms ± 1% 84.4ms ± 2% -10.90% (p=0.008 n=5+5) Read/sample_lz4-256k_262_kB-4 98.4ms ± 1% 87.7ms ± 2% -10.93% (p=0.008 n=5+5) Read/sample_lz4-1M_262_kB-4 101ms ± 2% 90ms ± 2% -11.19% (p=0.008 n=5+5) Read/sample_lz4_262_kB-4 100ms ± 2% 90ms ± 3% -10.78% (p=0.008 n=5+5) Read/sample_snappy_262_kB-4 77.9ms ± 2% 68.7ms ± 2% -11.84% (p=0.008 n=5+5) Read/sample_flate_262_kB-4 211ms ± 2% 190ms ± 2% -9.86% (p=0.008 n=5+5) Read/sample_none_524_kB-4 7.82ms ± 2% 6.43ms ± 2% -17.80% (p=0.008 n=5+5) Read/sample_gzip_524_kB-4 176ms ± 1% 164ms ± 2% -6.80% (p=0.008 n=5+5) Read/sample_lz4-64k_524_kB-4 77.7ms ± 1% 69.8ms ± 4% -10.10% (p=0.008 n=5+5) Read/sample_lz4-256k_524_kB-4 81.7ms ± 2% 72.1ms ± 3% -11.79% (p=0.008 n=5+5) Read/sample_lz4-1M_524_kB-4 83.8ms ± 1% 74.4ms ± 1% -11.21% (p=0.008 n=5+5) Read/sample_lz4_524_kB-4 83.2ms ± 0% 76.1ms ± 8% -8.53% (p=0.016 n=4+5) Read/sample_snappy_524_kB-4 63.6ms ± 2% 56.5ms ± 2% -11.23% (p=0.008 n=5+5) Read/sample_flate_524_kB-4 172ms ± 2% 157ms ± 1% -8.88% (p=0.008 n=5+5) name old speed new speed delta Write/ordered-lz4-256k-4 687MB/s ± 6% 749MB/s ± 2% +9.09% (p=0.008 n=5+5) Write/unordered-lz4-4 313MB/s ±24% 341MB/s ± 1% +9.23% (p=0.016 n=5+4) Read/none_66_kB-4 1.82GB/s ± 6% 2.97GB/s ± 2% +63.42% (p=0.008 n=5+5) Read/gzip_66_kB-4 669MB/s ± 4% 736MB/s ± 4% +10.00% (p=0.008 n=5+5) Read/lz4-64k_66_kB-4 1.42GB/s ± 2% 1.71GB/s ± 3% +20.21% (p=0.008 n=5+5) Read/lz4-256k_66_kB-4 1.43GB/s ± 0% 1.76GB/s ± 1% +22.83% (p=0.016 n=4+5) Read/lz4-1M_66_kB-4 1.43GB/s ± 2% 1.74GB/s ± 1% +21.99% (p=0.008 n=5+5) Read/lz4_66_kB-4 1.41GB/s ± 4% 1.76GB/s ± 1% +24.44% (p=0.008 n=5+5) Read/snappy_66_kB-4 1.29GB/s ± 2% 1.57GB/s ± 1% +21.86% (p=0.016 n=5+4) Read/flate_66_kB-4 710MB/s ± 4% 793MB/s ± 2% +11.70% (p=0.008 n=5+5) Read/zstd_66_kB-4 425MB/s ±14% 517MB/s ± 6% +21.74% (p=0.008 n=5+5) Read/none_262_kB-4 1.93GB/s ± 1% 3.06GB/s ± 2% +58.54% (p=0.008 n=5+5) Read/gzip_262_kB-4 701MB/s ± 5% 779MB/s ± 2% +11.20% (p=0.008 n=5+5) Read/lz4-64k_262_kB-4 1.47GB/s ± 1% 1.82GB/s ± 3% +24.03% (p=0.008 n=5+5) Read/lz4-256k_262_kB-4 1.48GB/s ± 1% 1.82GB/s ± 2% +23.09% (p=0.008 n=5+5) Read/lz4-1M_262_kB-4 1.46GB/s ± 2% 1.84GB/s ± 2% +26.44% (p=0.008 n=5+5) Read/lz4_262_kB-4 1.48GB/s ± 2% 1.85GB/s ± 2% +25.42% (p=0.008 n=5+5) Read/snappy_262_kB-4 1.31GB/s ± 2% 1.61GB/s ± 1% +22.80% (p=0.008 n=5+5) Read/flate_262_kB-4 729MB/s ± 3% 831MB/s ± 2% +14.05% (p=0.008 n=5+5) Read/none_524_kB-4 1.97GB/s ± 2% 3.08GB/s ± 2% +56.33% (p=0.008 n=5+5) Read/gzip_524_kB-4 714MB/s ± 2% 773MB/s ± 2% +8.27% (p=0.008 n=5+5) Read/lz4-64k_524_kB-4 1.47GB/s ± 3% 1.88GB/s ± 1% +28.48% (p=0.008 n=5+5) Read/lz4-256k_524_kB-4 1.50GB/s ± 1% 1.86GB/s ± 5% +23.63% (p=0.008 n=5+5) Read/lz4-1M_524_kB-4 1.49GB/s ± 1% 1.83GB/s ± 2% +22.51% (p=0.008 n=5+5) Read/lz4_524_kB-4 1.49GB/s ± 1% 1.86GB/s ± 1% +24.53% (p=0.008 n=5+5) Read/snappy_524_kB-4 1.34GB/s ± 1% 1.60GB/s ± 3% +19.75% (p=0.008 n=5+5) Read/flate_524_kB-4 750MB/s ± 1% 824MB/s ± 3% +9.97% (p=0.008 n=5+5) Read/sample_none_66_kB-4 967MB/s ± 1% 1142MB/s ± 6% +18.13% (p=0.008 n=5+5) Read/sample_gzip_66_kB-4 501MB/s ± 2% 542MB/s ± 1% +8.19% (p=0.008 n=5+5) Read/sample_lz4-64k_66_kB-4 811MB/s ± 2% 922MB/s ± 1% +13.66% (p=0.008 n=5+5) Read/sample_lz4-256k_66_kB-4 825MB/s ± 1% 913MB/s ± 2% +10.68% (p=0.008 n=5+5) Read/sample_lz4-1M_66_kB-4 795MB/s ± 7% 920MB/s ± 1% +15.67% (p=0.008 n=5+5) Read/sample_lz4_66_kB-4 824MB/s ± 1% 909MB/s ± 2% +10.25% (p=0.008 n=5+5) Read/sample_snappy_66_kB-4 759MB/s ± 1% 852MB/s ± 6% +12.37% (p=0.008 n=5+5) Read/sample_flate_66_kB-4 506MB/s ± 7% 556MB/s ± 4% +9.88% (p=0.008 n=5+5) Read/sample_none_262_kB-4 983MB/s ± 2% 1194MB/s ± 2% +21.47% (p=0.008 n=5+5) Read/sample_lz4-64k_262_kB-4 830MB/s ± 1% 932MB/s ± 2% +12.24% (p=0.008 n=5+5) Read/sample_lz4-256k_262_kB-4 839MB/s ± 1% 942MB/s ± 2% +12.28% (p=0.008 n=5+5) Read/sample_lz4-1M_262_kB-4 829MB/s ± 2% 934MB/s ± 2% +12.60% (p=0.008 n=5+5) Read/sample_lz4_262_kB-4 836MB/s ± 2% 937MB/s ± 3% +12.11% (p=0.008 n=5+5) Read/sample_snappy_262_kB-4 774MB/s ± 2% 878MB/s ± 2% +13.44% (p=0.008 n=5+5) Read/sample_flate_262_kB-4 529MB/s ± 2% 587MB/s ± 2% +10.94% (p=0.008 n=5+5) Read/sample_none_524_kB-4 970MB/s ± 2% 1180MB/s ± 2% +21.64% (p=0.008 n=5+5) Read/sample_gzip_524_kB-4 521MB/s ± 1% 559MB/s ± 2% +7.31% (p=0.008 n=5+5) Read/sample_lz4-64k_524_kB-4 842MB/s ± 1% 937MB/s ± 4% +11.30% (p=0.008 n=5+5) Read/sample_lz4-256k_524_kB-4 834MB/s ± 2% 945MB/s ± 3% +13.37% (p=0.008 n=5+5) Read/sample_lz4-1M_524_kB-4 842MB/s ± 1% 949MB/s ± 1% +12.63% (p=0.008 n=5+5) Read/sample_lz4_524_kB-4 849MB/s ± 0% 929MB/s ± 7% +9.53% (p=0.016 n=4+5) Read/sample_snappy_524_kB-4 783MB/s ± 2% 882MB/s ± 2% +12.65% (p=0.008 n=5+5) Read/sample_flate_524_kB-4 534MB/s ± 2% 585MB/s ± 1% +9.73% (p=0.008 n=5+5) name old alloc/op new alloc/op delta Write/unordered-lz4-4 18.8MB ± 4% 17.2MB ± 7% -8.28% (p=0.008 n=5+5) Read/none_66_kB-4 41.9kB ± 0% 43.7kB ± 0% +4.41% (p=0.008 n=5+5) Read/gzip_66_kB-4 713kB ± 0% 741kB ± 0% +3.85% (p=0.016 n=5+4) Read/lz4-64k_66_kB-4 730kB ± 1% 748kB ± 0% +2.53% (p=0.016 n=5+4) Read/lz4-256k_66_kB-4 770kB ± 0% 806kB ± 3% +4.70% (p=0.016 n=4+5) Read/lz4-1M_66_kB-4 770kB ± 0% 812kB ± 4% +5.43% (p=0.016 n=4+5) Read/lz4_66_kB-4 770kB ± 0% 792kB ± 0% +2.83% (p=0.029 n=4+4) Read/snappy_66_kB-4 361kB ± 0% 376kB ± 0% +4.02% (p=0.008 n=5+5) Read/flate_66_kB-4 715kB ± 0% 743kB ± 0% +3.88% (p=0.016 n=5+4) Read/none_262_kB-4 11.2kB ± 0% 11.7kB ± 0% +4.12% (p=0.016 n=5+4) Read/gzip_262_kB-4 198kB ± 0% 203kB ± 0% +2.93% (p=0.016 n=4+5) Read/lz4-64k_262_kB-4 169kB ± 0% 174kB ± 0% +2.75% (p=0.008 n=5+5) Read/lz4-256k_262_kB-4 177kB ± 0% 182kB ± 0% +2.79% (p=0.008 n=5+5) Read/lz4-1M_262_kB-4 181kB ± 0% 186kB ± 0% +2.77% (p=0.008 n=5+5) Read/lz4_262_kB-4 181kB ± 0% 186kB ± 0% +2.73% (p=0.016 n=5+4) Read/snappy_262_kB-4 87.8kB ± 0% 90.1kB ± 0% +2.58% (p=0.008 n=5+5) Read/flate_262_kB-4 197kB ± 0% 203kB ± 0% +2.63% (p=0.016 n=5+4) Read/zstd_262_kB-4 553MB ± 0% 552MB ± 0% -0.15% (p=0.032 n=5+5) Read/none_524_kB-4 5.94kB ± 0% 6.16kB ± 0% +3.76% (p=0.016 n=5+4) Read/gzip_524_kB-4 96.2kB ± 0% 98.3kB ± 0% +2.16% (p=0.016 n=4+5) Read/lz4-64k_524_kB-4 70.9kB ± 0% 72.8kB ± 0% +2.71% (p=0.016 n=4+5) Read/lz4-256k_524_kB-4 73.8kB ± 0% 75.8kB ± 0% +2.68% (p=0.008 n=5+5) Read/lz4-1M_524_kB-4 76.5kB ± 0% 78.7kB ± 0% +2.77% (p=0.016 n=4+5) Read/lz4_524_kB-4 76.5kB ± 0% 78.6kB ± 0% +2.74% (p=0.016 n=5+4) Read/snappy_524_kB-4 39.1kB ± 0% 39.5kB ± 0% +1.19% (p=0.008 n=5+5) Read/flate_524_kB-4 95.4kB ± 0% 97.3kB ± 0% +2.06% (p=0.008 n=5+5) Read/sample_none_66_kB-4 39.9kB ± 0% 41.7kB ± 0% +4.62% (p=0.008 n=5+5) Read/sample_gzip_66_kB-4 686kB ± 0% 715kB ± 0% +4.23% (p=0.016 n=4+5) Read/sample_lz4-64k_66_kB-4 707kB ± 0% 728kB ± 0% +2.88% (p=0.016 n=5+4) Read/sample_lz4-256k_66_kB-4 749kB ± 0% 783kB ± 4% +4.65% (p=0.008 n=5+5) Read/sample_lz4_66_kB-4 748kB ± 0% 770kB ± 0% +2.91% (p=0.016 n=5+4) Read/sample_snappy_66_kB-4 350kB ± 0% 364kB ± 0% +4.04% (p=0.029 n=4+4) Read/sample_flate_66_kB-4 688kB ± 0% 716kB ± 0% +4.08% (p=0.029 n=4+4) Read/sample_none_262_kB-4 10.6kB ± 0% 11.0kB ± 0% +4.23% (p=0.016 n=5+4) Read/sample_gzip_262_kB-4 194kB ± 0% 199kB ± 0% +2.48% (p=0.016 n=4+5) Read/sample_lz4-64k_262_kB-4 165kB ± 0% 169kB ± 0% +2.86% (p=0.016 n=4+5) Read/sample_lz4-256k_262_kB-4 172kB ± 0% 177kB ± 0% +2.92% (p=0.016 n=4+5) Read/sample_lz4-1M_262_kB-4 176kB ± 0% 181kB ± 0% +2.91% (p=0.029 n=4+4) Read/sample_lz4_262_kB-4 176kB ± 0% 181kB ± 0% +2.81% (p=0.016 n=5+4) Read/sample_snappy_262_kB-4 88.1kB ± 0% 90.8kB ± 0% +3.11% (p=0.029 n=4+4) Read/sample_flate_262_kB-4 194kB ± 0% 198kB ± 0% +2.13% (p=0.029 n=4+4) Read/sample_none_524_kB-4 5.56kB ± 0% 5.76kB ± 0% +3.65% (p=0.016 n=5+4) Read/sample_gzip_524_kB-4 95.4kB ± 0% 96.9kB ± 0% +1.56% (p=0.008 n=5+5) Read/sample_lz4-64k_524_kB-4 69.0kB ± 0% 71.0kB ± 0% +2.87% (p=0.008 n=5+5) Read/sample_lz4-256k_524_kB-4 71.8kB ± 0% 73.8kB ± 0% +2.79% (p=0.016 n=4+5) Read/sample_lz4-1M_524_kB-4 74.5kB ± 0% 76.6kB ± 0% +2.75% (p=0.008 n=5+5) Read/sample_lz4_524_kB-4 74.5kB ± 0% 188.4kB ±89% +152.98% (p=0.008 n=5+5) Read/sample_snappy_524_kB-4 40.7kB ± 1% 41.3kB ± 1% +1.58% (p=0.008 n=5+5) Read/sample_flate_524_kB-4 95.2kB ± 0% 96.2kB ± 0% +0.95% (p=0.008 n=5+5) name old allocs/op new allocs/op delta Read/gzip_66_kB-4 13.2k ± 0% 13.2k ± 0% -0.04% (p=0.000 n=5+4) Read/zstd_66_kB-4 18.0k ± 1% 17.8k ± 1% -1.06% (p=0.032 n=5+5) Read/gzip_262_kB-4 4.61k ± 0% 4.61k ± 0% -0.11% (p=0.029 n=4+4) Read/flate_262_kB-4 4.61k ± 0% 4.60k ± 0% -0.11% (p=0.008 n=5+5) Read/zstd_262_kB-4 7.10k ± 2% 6.88k ± 3% -3.01% (p=0.008 n=5+5) Read/gzip_524_kB-4 2.62k ± 0% 2.62k ± 0% -0.06% (p=0.000 n=4+5) Read/zstd_524_kB-4 6.13k ± 2% 5.94k ± 2% -3.03% (p=0.024 n=5+5) Read/sample_flate_66_kB-4 13.2k ± 0% 13.2k ± 0% -0.01% (p=0.029 n=4+4) Read/sample_flate_262_kB-4 4.62k ± 0% 4.61k ± 0% -0.22% (p=0.029 n=4+4) Read/sample_zstd_262_kB-4 7.12k ± 1% 6.97k ± 1% -2.11% (p=0.008 n=5+5) Read/sample_gzip_524_kB-4 2.63k ± 0% 2.63k ± 0% -0.11% (p=0.008 n=5+5) Read/sample_flate_524_kB-4 2.63k ± 0% 2.62k ± 0% -0.15% (p=0.008 n=5+5) Read/sample_zstd_524_kB-4 6.11k ± 2% 5.89k ± 1% -3.68% (p=0.016 n=5+4) ``` Co-authored-by: Owen Diehl <ow.diehl@gmail.com>	3 years ago
Susana Ferreira	10e9695b21	Fix rangemapper with VectorExpr support (#7288 ) What this PR does / why we need it: The new expression type VectorExpr introduced recently in [PR](https://github.com/grafana/loki/pull/7007), is not taken into account in the query mapping in the AST mapper rangemapper. The AST mapper shardmapper was updated in [PR](https://github.com/grafana/loki/pull/7045). Special notes for your reviewer: Checklist - [ ] Reviewed the `CONTRIBUTING.md` guide - [ ] Documentation added - [x] Tests updated - [x] `CHANGELOG.md` updated - [ ] Changes that require user attention or interaction to upgrade are documented in `docs/sources/upgrading/_index.md`	3 years ago
Gerard Vanloo	bcdf17b0a5	Configurable TLS MinVersion and CipherSuites (#7227 )	3 years ago
Michel Hollands	77c10da81d	Remove deletion filter debug log messages (#7277 ) These debug messages are a slightly misleading and appear too frequent. Signed-off-by: Michel Hollands <michel.hollands@grafana.com>	3 years ago
Michel Hollands	391b9cea10	Logcli: load remote schema before validation (#7258 ) The --remote-schema option didn't work as validation would be called before the remote schema was loaded. An error about a missing schema section is the result. This was tested by running against a remote bucket containing a schemaconfig.yaml file with a local config file that does not contain a schema config section. Signed-off-by: Michel Hollands <michel.hollands@grafana.com>	3 years ago
Dylan Guedes	dcfba366de	Loki: Shard streams based on desired rate (#7199 ) What this PR does / why we need it: Modify our sharding mechanism to be dictated by the desired rate and ingestion rate. Once the number of shards is calculated, the stream entries will be distributed across the shards. If the number of entries is lower than number of shards, we report a metric and use number of entries as the number of shards.	3 years ago
Owen Diehl	dab6a24541	correct heartbeat timeout to distributor lifecycler for ring page (#7253 ) Passes the correct `heartbeat_timeout` to the created lifecycler config in distributor ring construction. When we didn't include this, it would create a visual bug incorrectly marking many distributors as unhealthy when in fact they were in fine condition. Including The _lifecycler_ `heartbeat_timeout` is _only_ referenced in the status page [here](`74c8cf03ba/vendor/github.com/grafana/dskit/ring/lifecycler.go (L892-L894)`). The _ring_ `heartbeat_timeout` continues to be updated correctly.	3 years ago
李国忠	4089b73bc3	[Optimization] logql client: When logql fails to execute, return the failure information to the caller (#6671 ) …ail msg to the client caller. When logql fails to execute, return the failure information to the caller. Solve the problem that grafana's loki alarm cannot see the cause of failure. Co-authored-by: Christian Haudum <christian.haudum@gmail.com>	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
李国忠	557fdf2206	[bugfix] logql: vectorExpr run fail when query_range: parallelise_shardable_queries: true (#7045 ) …hardable_queries: true <!-- 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: A grafana error occurred when I deployed the vector(0) function to our online loki cluster. It was finally found because online loki has a configuration `unexpected expr type (syntax.VectorExpr) for ASTMapper type (logql.ShardMapper) ` ```yaml query_range: parallelise_shardable_queries: true ``` Which issue(s) this PR fixes: Fixes # Special notes for your reviewer: error snapssnapshot ![image](https://user-images.githubusercontent.com/9583245/188357957-e573e589-a789-4e5e-a1c3-b91c019ceb62.png) with this pr，success snapshot ![image](https://user-images.githubusercontent.com/9583245/188358036-5ca0d9be-c2b6-4516-807c-c027445ee62d.png) <!-- 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 - [ ] 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`	3 years ago
Sandeep Sukhani	2af3d680cb	tsdb supports data deletion, this PR takes care of todo to enable delete request client for tsdb (#7223 ) What this PR does / why we need it: `tsdb` index store supports deletion of data, but we had missed enabling the creation of delete requests client, which is required for query time filtering of data requested for deletion. This PR takes care of it. I have also removed some unused code and adjusted the tests accordingly. Checklist - [x] Tests updated	3 years ago
Sandeep Sukhani	7f298ff72f	avoid using bloomfilters for chunks in stats calls by avoiding duplicates (#7209 ) What this PR does / why we need it: Avoid using bloomfilters for chunks deduplication in tsdb `Stats` calls by avoiding fetching duplicate entries. The idea is to split and align queries by [ObjectStorageIndexRequiredPeriod](`61794710a7/pkg/storage/config/schema_config.go (L47)`) and make each split process chunks with a start time >= start time of the table interval. In other terms, table interval that contains start time of the chunk, owns it. For e.g. if the table interval is 10s, and we have chunks 5-7, 8-12, 11-13. Query with range 6-15 would be split into 6-10, 10-15. query1 would process chunks 5-7, 8-12 and query2 would process chunks 11-13. This check is not applied for the first split so that we do not eliminate any chunks that overlaps the original query intervals but starts at the previous table. For e.g. if the table interval is 10s, and we have chunks 5-7, 8-13, 14-13. Query with range 11-12 should process chunk 8-13 even though its start time <= start time of table we will query for index. The caveat here is that we will overestimate the data we will be processing if the index is not compacted yet since it could have duplicate chunks when RF > 1. I think it is okay since the Stats call is just an estimation and need not be accurate. Removing all the extra processing saves us quite a bit of CPU and memory, as seen from the benchmark comparison between the two implementations: ``` name old time/op new time/op delta IndexClient_Stats-10 187µs ± 0% 34µs ± 1% -82.00% (p=0.008 n=5+5) name old alloc/op new alloc/op delta IndexClient_Stats-10 61.5kB ± 4% 12.5kB ± 2% -79.69% (p=0.008 n=5+5) name old allocs/op new allocs/op delta IndexClient_Stats-10 1.46k ± 0% 0.48k ± 0% -67.28% (p=0.008 n=5+5) ``` Checklist - [x] Tests updated	3 years ago
Owen Diehl	bac52165d3	highly sharded subqueries are ellided in expr.String (#7203 ) This is mostly a quality of life improvement. Highly sharded subqueries can be burdensome to view and they're logged on every request. This PR will only display the first four subqueries in a `Concat{Log,Sample}Expr`, which is easier on the eyes when viewing logs and is more efficient.	3 years ago
Sandeep Sukhani	4e7105d018	add index requests panel to reads/writes dashboard (#7163 ) What this PR does / why we need it: In PR #7154, I refactored the code also to measure the latency on index write requests. This PR adds the index requests panel to the Loki Reads/Writes dashboard.	3 years ago
Christian Haudum	67b12f2357	Update distributor push tests (#6641 ) Adds a test case for concurrent pushes (which would have spotted a bug in https://github.com/grafana/loki/pull/5248) and a test case for tolerance regarding replication factor. The PR also replaces the mock ring with an actual in-memory kv store ring. And variables and function names were changed to meet Loki/logs terminology. Signed-off-by: Christian Haudum <christian.haudum@gmail.com>	3 years ago
Ashwanth	3245625c40	ingester_querier: early exit queries on reaching quorum (#7172 ) Currently `forGivenIngesters` and its derivatives don't allow passing down context to the work function as a result of which we are not making use of the tracking functionality provided by `replicationSet.Do` to exit early and relieve pressure on the ingesters. This change ensures that Unary rpc calls receive ctx populated by the replication set tracker to exit early on reaching quorum or max failures. The same cannot be applied to stream rpc calls given how the code is structured at the moment where we are processing the streams outside of the work func	3 years ago
Owen Diehl	12ebd33f94	TSDB indexBuckets helper logs instead of errors (#7187 ) indexBuckets will return the list of overlapping TSDB buckets for a time range. Historically, this errored when there was no overlap, which is generally impossible. While mutating period configurations as a test (which should never be done in production), I realized this would preemptively error us when there were old TSDB WALs from a bucket that no longer belonged to a TSDB period (i.e. by replacing a `tsdb` index with a `boltdb-shipper` one). This PR changes this to a log line instead as we don't need to error and stop computation; it's better to return an empty overlap instead. Co-authored-by: Dylan Guedes <djmgguedes@gmail.com>	3 years ago
Owen Diehl	f1114fd221	[Bugfix] TSDB index bucket calculation (#7178 ) When calculating the index buckets a chunk belongs to, we shouldn't error when a bucket falls outside the current TSDB store's periodConfig range. This can happen when a chunk is written across the bounds between a previous index type (i.e. boltdb-shipper) and a new tsdb type or vice versa. Instead of erroring when this happens, we'll now correctly return only the buckets matching the TSDB store (the other buckets are handled by their respective stores). I think we can also remove the error from this function signature, although I left it in as it's conceivably helpful in the case of a chunk range that doesn't overlap at all with the store (which shouldn't happen). Dependent on https://github.com/grafana/loki/pull/7177	3 years ago
Owen Diehl	85e8f00958	fixes azure client secret duplicate CLI registration (#7185 ) Fixes duplicate key registration from https://github.com/grafana/loki/pull/7179/files#diff-665c7ce7ef736bdf9f9da76a8e389d5b6fcdc90099db913ce36b4c0da2a97ef2R121-R122	3 years ago
Owen Diehl	ab24efe1bf	allow backup store metrics registration by including backend type in labels (#7177 ) The refactor [here](https://github.com/grafana/loki/blob/main/pkg/storage/store.go#L237) introduced in https://github.com/grafana/loki/pull/7154 will cause a panic due to duplicate registration via the same store prefix when using a backup index store (for use during tsdb migrations). This PR adds more granularity to the label value to differentiate these and avoid the problem.	3 years ago
Vladyslav Diachenko	7f629c8849	[loki] Azure auth using Service Principal (#7179 ) This PR adds the ability to use Service Principal credentials to authenticate through Azure OAuth. It might be needed when we want to access Azure Blob storage from another tenant. Previously we could use only StorageAccount key that can not be used cross-tenant.	3 years ago
Owen Diehl	2766e0a8c7	fixes nil interface value with non-nil type in cache construction (#7175 ) Fixes a bug introduced in https://github.com/grafana/loki/pull/7013 This will panic when the interface's _value_ is nil because it's been assigned a concrete type already and then passed as an interface. Instead, we treat it as a nil interface with no type bound to it: ``` panic: runtime error: invalid memory address or nil pointer dereference goroutine 40570 [running]: github.com/grafana/loki/pkg/util/server.onPanic({0x2209320, 0x3dfae20}) /src/enterprise-logs/vendor/github.com/grafana/loki/pkg/util/server/recovery.go:41 +0x53 github.com/grafana/loki/pkg/util/server.glob..func1.1.1() /src/enterprise-logs/vendor/github.com/grafana/loki/pkg/util/server/recovery.go:29 +0x3a panic({0x2209320, 0x3dfae20}) /usr/local/go/src/runtime/panic.go:838 +0x207 github.com/grafana/loki/pkg/storage/stores/indexshipper/compactor/generationnumber.(GenNumberLoader).getCacheGenNumber(0x0, {0xc0081e14d7, 0x2}) /src/enterprise-logs/vendor/github.com/grafana/loki/pkg/storage/stores/indexshipper/compactor/generationnumber/gennumber_loader.go:120 +0x60 github.com/grafana/loki/pkg/storage/stores/indexshipper/compactor/generationnumber.(GenNumberLoader).getCacheGenNumbersPerTenants(0x2?, {0xc000e8c4d0?, 0x1, 0x0?}) /src/enterprise-logs/vendor/github.com/grafana/loki/pkg/storage/stores/indexshipper/compactor/generationnumber/gennumber_loader.go:98 +0xe7 github.com/grafana/loki/pkg/storage/stores/indexshipper/compactor/generationnumber.(*GenNumberLoader).GetResultsCacheGenNumber(0x2b26298?, {0xc000e8c4d0?, 0xc0050ad080?, 0x2dcdeeb9bcc040e4?}) ```	3 years ago
Travis Patterson	071118a090	revert stream push error response (#7160 ) We've changed approaches to how we want to do per-stream rate limiting. The new approach will rely on creating an api on the ingester so the distributors can decide how streams should be sharded. As a result, the work to return rate-limited streams in the error body is no longer needed.	3 years ago
Sandeep Sukhani	c8c8303a6a	refactor index storage code to index reader/writer interfaces for better monitoring (#7154 ) What this PR does / why we need it: Currently, it is hard to monitor just the index latencies since the chunks PUT code is coupled with both object storage and index writer. This PR refactors the code to have separate interfaces defined for index readers/writers. I have also added a wrapper called `monitoredReaderWriter`, which measures the latency on the index operations. Special notes for your reviewer: Cortex had added `PutChunksAndIndex` for DynamoDB, which would write both chunks and index in a single request when configured for storing both. I have removed the use of it since it would be too complex to support it with this refactoring. Loki, anyways does not recommend using chunks in the index store, and using a hosted object storage would be much cheaper. Also, this is not a breaking change, it might somewhat impact the performance of flushing the chunks. Co-authored-by: Owen Diehl <ow.diehl@gmail.com>	3 years ago
Ashwanth	88503ae2e1	tsdb: clean-up metrics (#7150 ) What this PR does / why we need it: clean-up unused tsdb metrics and refactor others to use a `CounterVec` Signed-off-by: Ashwanth Goli <iamashwanth@gmail.com>	3 years ago
Kaviraj Kanagaraj	6f0690e7bf	fix(ruler): Initialize multi remote clients config (#7151 ) <!-- 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: To be backward compatible with single client config. ``` ts=2022-09-14T08:24:04.549961234Z caller=memberlist_logger.go:74 level=debug msg="Initiating push/pull sync with: 10.132.18.167:7946" │ │ panic: assignment to entry in nil map │ │ goroutine 1 [running]: │ │ github.com/grafana/loki/pkg/ruler.NewRuler({{{0xc00002a120}, {0x0, 0x0, 0x0}, {0x6400000, 0x6400000, {0x0, 0x0}, 0x0, 0x0, ...}, ...}, ...}, ...) │ │ /src/enterprise-logs/vendor/github.com/grafana/loki/pkg/ruler/ruler.go:21 +0x1b7 ``` Which issue(s) this PR fixes: Fixes #<issue number> 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 - [ ] 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`	3 years ago
Karol Bakunowski	4e7e51ceb5	Remove duplicated `loki_boltdb_shipper` prefix from `tables_upload_operation_total` metric (#7040 ) This PR removes the duplicated `loki_boltdb_shipper` prefix from the `tables_upload_operation_total` metric.	3 years ago
Travis Patterson	3f0ecd3c7b	Implement all-or-nothing stream ingestion (#7094 ) Per-stream sharding will introduce automatic stream sharding on rate-limited streams. The ingester's current line-by-line behavior may result in duplicates when a partially-accepted stream is resharded and sent to other ingesters. This PR makes the ingester either rate limit the entire stream before attempting to write lines to a chunk. The PR 1. Separates validation, storage, and wal replay into their own functions 2. Implements a flag to rate limit the whole stream or not 3. Checks rate limits for the whole stream rather than line-by-line 4. Returns an error with information on the whole stream when rate limited	3 years ago
李国忠	461f844044	[bugfix] fix query_frontend init fail error (#7013 ) Co-authored-by: Ashwanth <iamashwanth@gmail.com>	3 years ago
Sandeep Sukhani	61794710a7	tsdb: boltdb shipper backup store tsdb (#7130 ) <!-- 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: Since we are planning to start deploying `tsdb` to critical cells, we want to have a disaster recovery without any data loss. This PR adds support to `tsdb` store itself to configure `boltdb-shipper` as a backup chunk writer. When the flag is enabled, `tsdb` store would also index chunks in `boltdb-shipper` index store with table names set to `<configured-tsdb-table-prefix>`+`backup_`+`<table-number>`. This is the simplest option that I can come up with without maintaining a lot of code and we can simply drop the support before marking `tsdb` as production ready. The rollback/recovery plan would just be to change the schema config to use `boltdb-shipper` as index store with `<configured-tsdb-table-prefix>`+`backup_` as prefix like below: tsdb enabled: ``` schema_config: configs: - from: 2022-01-01 store: boltdb-shipper index: prefix: index_ - from: 2022-09-10 store: tsdb index: prefix: index_ ``` rollback: ``` schema_config: configs: - from: 2022-01-01 store: boltdb-shipper index: prefix: index_ - from: 2022-09-10 store: boltdb-shipper index: prefix: index_backup_ ``` <!-- 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] Tests updated	3 years ago

1 2 3 4 5 ...

1745 Commits (e89c30ce6badcfa15cf435ebf86bc98ca71f4a42)