[#1349] node: Evacuate objects without setting mode to MAINTENANCE
All checks were successful
DCO action / DCO (pull_request) Successful in 1m31s
Vulncheck / Vulncheck (pull_request) Successful in 1m32s
Pre-commit hooks / Pre-commit (pull_request) Successful in 2m12s
Tests and linters / Run gofumpt (pull_request) Successful in 2m21s
Build / Build Components (pull_request) Successful in 2m32s
Tests and linters / gopls check (pull_request) Successful in 2m36s
Tests and linters / Staticcheck (pull_request) Successful in 2m55s
Tests and linters / Tests with -race (pull_request) Successful in 3m26s
Tests and linters / Lint (pull_request) Successful in 3m31s
Tests and linters / Tests (pull_request) Successful in 3m40s

Signed-off-by: Anton Nikiforov <an.nikiforov@yadro.com>
This commit is contained in:
Anton Nikiforov 2024-09-03 12:18:10 +03:00
parent b3deb893ba
commit 108e4e07be
18 changed files with 156 additions and 17 deletions

View file

@ -10,6 +10,12 @@ First of all, by the evacuation the data is transferred to other shards of the s
Only one running evacuation process is allowed on the node at a time.
It is not necessary to turn maintenance mode on storage node.
Once evacuation from shard started, it is impossible to read data from it via public API, except the case when evacuation stopped manually or node restarted.
Because it is necessary to prevent removing by policer objects with policy `REP 1 ...` from remote node during evacuation.
`frostfs-cli` utility is used to manage evacuation.
## Commands