# Load imbalance after replacing a failed node

**URL:** https://forum.scylladb.com/t/load-imbalance-after-replacing-a-failed-node/3174
**Category:** ScyllaDB
**Tags:** administration, repair
**Created:** [November 13, 2024, 12:53pm UTC](https://forum.scylladb.com/t/load-imbalance-after-replacing-a-failed-node/3174 "2024-11-13T12:53:19Z")
**Posts on this page:** 1
**Page:** 1

<div class="post-metadata">

### Author: ![Guy](https://sea2.discourse-cdn.com/flex016/user_avatar/forum.scylladb.com/guy/32/9_2.png) [@Guy](https://forum.scylladb.com/u/Guy)
#### Post date: [November 13, 2024, 12:53pm UTC](https://forum.scylladb.com/t/load-imbalance-after-replacing-a-failed-node/3174/1 "2024-11-13T12:53:19Z")

</div>

> [@](#):
>
> [**Originally from the User Slack**](https://scylladb-users.slack.com/)
> 
> ![Mr._Layup](https://us1.discourse-cdn.com/flex016/uploads/scylladb/original/2X/1/135de6033831a5f504b573c4a62dfc32f2fc9a42.png) **@Mr.\_Layup:** Hello!  
> I have a cluster with 3 nodes, and yesterday the disk on the DB2 node failed. Today, I replaced the disk and rebuilt the RAID0 (which resulted in data loss). After that, I re-added the node using the replace\_node\_first\_boot option. However, I am now experiencing some imbalance in load:
> 
> ```auto
> Status=Up/Down
> |/ State=Normal/Leaving/Joining/Moving
> -- Address Load Tokens Owns Host ID Rack
> UN DB1 1.35 GB 256 ? 266ce553-11ab-413c-a3fa-2fdfd1a3d452 RACK1
> UN DB2 1.36 GB 256 ? d932241c-0fe9-4965-833e-599d3be68555 RACK2
> UN DB3 423.32 MB 256 ? cc38257d-ac6e-4767-8806-4c970e84b986 RACK3
> 
> ```
> 
> Before performing operations on the production cluster, I tested the process on a similar development cluster using synthetic data, and I did not observe such a significant load imbalance.  
> Is it safe to run a repair on the cluster? I have some concerns about potentially losing data from my cluster.  
> Thank you for your help!
> 
> ![Felipe_Cardeneti_Mendes](https://us1.discourse-cdn.com/flex016/uploads/scylladb/original/1X/233a9d524fbc97e39f08633c16b3dd2575d1d1f4.jpeg) **@Felipe\_Cardeneti\_Mendes:** Yes, you may repair - though if you are running a reasonably recent version replace should’ve already repaired on its own. Doesn’t seem like a problem, keep in mind the load column refreshes whenever nodes get restarted, etc
> 
> ![Mr._Layup](https://us1.discourse-cdn.com/flex016/uploads/scylladb/original/2X/1/135de6033831a5f504b573c4a62dfc32f2fc9a42.png) **@Mr.\_Layup:** 5.4.4.  
> Thank yu, didn’t know about refreshing after restart.