Merge branch 'feature/sidecar-flags' of github.com:yeya24/thanos into…

… feature/sidecar-flags

docs/components/sidecar.md

@@ -14,7 +14,8 @@ Prometheus servers connected to the Thanos cluster via the sidecar are subject t
 * The minimum Prometheus version is 2.2.1
 * The `external_labels` section of the configuration implements is in line with the desired label scheme (will be used by query layer to filter out store APIs to query).
 * The `--web.enable-lifecycle` flag is enabled if you want to use `reload.*` flags.
-* The `--storage.tsdb.min-block-duration` and `--storage.tsdb.max-block-duration` must be set to equal values to disable local compaction. The default of `2h` is recommended. Mentioned parameters set to equal value disable internal Prometheus compaction, which is needed to avoid the uploaded data corruption when thanos compactor does its job, this is critical for data consistency and should not be ignored if you plan to use thanos compactor. Even though you set mentioned parameters equal, you might observe Prometheus internal metric `prometheus_tsdb_compactions_total` being incremented, don't be confused by that: Prometheus passes all the complete data blocks via internal compaction mechanism with no exception, but if you followed recommendations - data won't be modified by Prometheus before sidecar uploads it. Thanos sidecar will also check sanity of the flags set to Prometheus on the startup and log errors if they have been configured improperly (#838).
+* The `--storage.tsdb.min-block-duration` and `--storage.tsdb.max-block-duration` must be set to equal values to disable local compaction on order to use Thanos sidecar upload. Leave local compaction on if sidecar just exposes StoreAPI and your retention is normal. The default of `2h` is recommended. 
+  Mentioned parameters set to equal values disable the internal Prometheus compaction, which is needed to avoid the uploaded data corruption when thanos compactor does its job, this is critical for data consistency and should not be ignored if you plan to use Thanos compactor. Even though you set mentioned parameters equal, you might observe Prometheus internal metric `prometheus_tsdb_compactions_total` being incremented, don't be confused by that: Prometheus writes initial head block to filestem via internal compaction mechanis, but if you followed recommendations - data won't be modified by Prometheus before sidecar uploads it. Thanos sidecar will also check sanity of the flags set to Prometheus on the startup and log errors or warning if they have been configured improperly (#838).
 
 The retention is recommended to not be lower than three times the block duration. This achieves resilience in the face of connectivity issues
 to the object storage since all local data will remain available within the Thanos cluster. If connectivity gets restored the backlog of blocks gets uploaded to the object storage.