NSX Manager Cluster Repository Sync Fails with Error Code 21057
search cancel

NSX Manager Cluster Repository Sync Fails with Error Code 21057

book

Article ID: 440852

calendar_today

Updated On:

Products

VMware NSX

Issue/Introduction

After recovering, replacing, or redeploying an NSX Manager node within a cluster, the repository synchronization (Repo_Sync) task enters a FAILED state on the newly attached manager nodes.

The NSX Manager UI shows Repo_Sync status as FAILED.

Automated features or edge deployments that rely on installation bundles fail to execute.

Checking the /var/log/proton/nsxapi.log reveals the following signature error:
<time_stamp> INFO http-nio-127.0.0.1-7440-exec-18 ManagerApplianceFacadeImpl ###### SYSTEM [nsx@6876 comp="nsx-manager" level="INFO" reqId="####-####-####" subcomp="manager" username="admin"] Repo Sync status for node ####-####-#### is RepoSyncStatusReportDto{status='FAILED', statusMessage='', failureMessage='Unable to connect to File /repository/<NSX_version_Bundle>/Edge/ovf/<file_name> on source <NSX_Manager_IP>. Please verify that file exists on source and install-upgrade service is up.', failureCode='21057'}



Environment

VMware NSX-T

Cause

An interrupted or dirty manual transfer can leave stale .filepart fragments (e.g., nsx-edge.vmdk.filepart) in the directory tree, causing the synchronization engine validation check to fail.

Resolution

To resolve this issue, you must remove any residual partial transfer files and recursively restore the correct ownership across the repository partition.

1. Identify the Broken Nodes:

Log in via SSH to the NSX Manager nodes reporting the failure as root and inspect the directory structure.
ls -lR /repository/4.x.x.x.x.xxxxxxx/Edge/ovf

Look for files owned by root:root instead of uuc:grepodir, or look for any trailing .filepart extensions.

2. Clean the Directory:

If any incomplete temporary files exist, remove them:
rm /repository/<bundle_version>/Edge/ovf/*.filepart

3. Correct Permissions and Ownership: 

Recursively change the user and group ownership back to the native NSX parameters:
chown -R uuc:grepodir /repository

4. Force a Repo Sync:

Log in to NSX UI and click on "Resolve" button besides the "Failed" status of Repo_Sync Service.

Additional Information

Please refer to this article for more insights: After replacing Managers or while running Upgrade prechecks, Repo_Sync is Failed.