From 4029c539b9c6a8dfcd36e8c1d757bc19ede742c2 Mon Sep 17 00:00:00 2001
From: Dmytro Kozlov <kozlovdmitriyy@gmail.com>
Date: Fri, 10 Mar 2023 12:29:08 +0200
Subject: [PATCH] app/vmctl: update importing tips when migrating data with
 overlapping time range (#3941)

app/vmctl: update importing tips when migrating data with overlapping time range
---
 app/vmctl/README.md | 3 +++
 docs/vmctl.md       | 3 +++
 2 files changed, 6 insertions(+)

diff --git a/app/vmctl/README.md b/app/vmctl/README.md
index fd23c5ba5..000e463a0 100644
--- a/app/vmctl/README.md
+++ b/app/vmctl/README.md
@@ -790,6 +790,9 @@ to allocated CPU resources of your VictoriaMetrics installation.
 to the corresponding source address.
 9. `vmctl` supports `--vm-native-disable-http-keep-alive` to allow `vmctl` to use non-persistent HTTP connections to avoid
 error `use of closed network connection` when run a longer export.
+10. Migrating data with overlapping time range for destination data can produce duplicates series at destination.
+To avoid duplicates on the destination set `-dedup.minScrapeInterval=1ms` for `vmselect` and `vmstorage`.
+This will instruct `vmselect` and `vmstorage` to ignore duplicates with match timestamps.     
 
 In this mode `vmctl` acts as a proxy between two VM instances, where time series filtering is done by "source" (`src`)
 and processing is done by "destination" (`dst`). So no extra memory or CPU resources required on `vmctl` side. Only
diff --git a/docs/vmctl.md b/docs/vmctl.md
index bfa2b9157..d0bef4292 100644
--- a/docs/vmctl.md
+++ b/docs/vmctl.md
@@ -794,6 +794,9 @@ to allocated CPU resources of your VictoriaMetrics installation.
 to the corresponding source address.
 9. `vmctl` supports `--vm-native-disable-http-keep-alive` to allow `vmctl` to use non-persistent HTTP connections to avoid
 error `use of closed network connection` when run a longer export.
+10. Migrating data with overlapping time range for destination data can produce duplicates series at destination.
+To avoid duplicates on the destination set `-dedup.minScrapeInterval=1ms` for `vmselect` and `vmstorage`.
+This will instruct `vmselect` and `vmstorage` to ignore duplicates with match timestamps.
 
 In this mode `vmctl` acts as a proxy between two VM instances, where time series filtering is done by "source" (`src`)
 and processing is done by "destination" (`dst`). So no extra memory or CPU resources required on `vmctl` side. Only