Incremental sync stuck after the initial sync is completed for a VM on vSphere Replication.
search cancel

Incremental sync stuck after the initial sync is completed for a VM on vSphere Replication.

book

Article ID: 385124

calendar_today

Updated On:

Products

VMware vSphere ESXi

Issue/Introduction

Symptoms:

  • VM newly configured for replication fails to complete the first incremental sync. However, the initial sync completes successfully.

  • All other VMs complete first replication with no issues and continues to replicate.

Environment

vSphere Replication 8.7

Cause

  • You may see file transfer failure events from /var/run/log/hostd.log on ESXi host.

2025-02-18T09:48:44.966Z info hostd[2102433] [Originator@6876 sub=Hbrsvc] ReplicationConfig file (vm=lctcvp10123): (type=2) (path=/vmfs/volumes/64eae1c8-a0530608-b19d-########/vm-name/vm-name.nvram)
2025-02-18T09:48:44.972Z info hostd[2102433] [Originator@6876 sub=Vimsvc.ha-eventmgr] Event 22755 : Sync started by VR Scheduler for virtual machine vm-name on host host.example.com in cluster host.example.com in ha-datacenter.
2025-02-18T09:48:44.972Z info hostd[2102433] [Originator@6876 sub=Hbrsvc] ReplicationGroup started new instance (groupID=GID-7fb83442-b2c0-4c20-9fd7-########) (instanceId=replica-52f0116e-e050-c1a4-473f-########)
2025-02-18T09:48:44.975Z info hostd[2101564] [Originator@6876 sub=Vimsvc.TaskManager opID=f58d9e7a user=vpxuser] Task Created : haTask--vim.event.EventHistoryCollector.readNext-19709960
2025-02-18T09:48:44.976Z info hostd[2101559] [Originator@6876 sub=Vimsvc.TaskManager opID=f58d9e7a user=vpxuser] Task Completed : haTask--vim.event.EventHistoryCollector.readNext-19709960 Status success
2025-02-18T09:48:44.976Z error hostd[2101557] [Originator@6876 sub=Hbrsvc] Error computing digest: Is a directory
2025-02-18T09:48:44.976Z error hostd[2101557] [Originator@6876 sub=Hbrsvc] ReplicationGroup: Transfer config files failed (groupID=GID-7fb83442-b2c0-4c20-9fd7-d04bd6e4a6ef) (instanceId=replica-52f0116e-e050-c1a4-473f-71bd93d74097): Fault cause: vmodl.fault.SystemError
-->
2025-02-18T09:48:44.976Z warning hostd[2101557] [Originator@6876 sub=Hbrsvc] Replication group aborting replica (instanceId=replica-52f0116e-e050-c1a4-473f-########) (deltaState=file transfer failed)
2025-02-18T09:48:44.976Z info hostd[2101557] [Originator@6876 sub=Hbrsvc] ReplicatedDisk abort, state: inactive (diskID=RDID-891b6367-99e8-46cc-838f-########) (groupID=GID-7fb83442-b2c0-4c20-9fd7-########)
2025-02-18T09:48:44.976Z info hostd[2101557] [Originator@6876 sub=Vimsvc.ha-eventmgr] Event 22756 : Sync stopped for virtual machine vm-name on host host.example.com in cluster host.example.com in ha-datacenter: File transfer failed
2025-02-18T09:48:44.976Z warning hostd[2101557] [Originator@6876 sub=Hbrsvc] ReplicationGroup finished aborting delta (groupID=GID-7fb83442-b2c0-4c20-9fd7-d04bd6e4a6ef) (instanceId=replica-52f0116e-e050-c1a4-473f-########) (deltaState=file transfer failed)
2025-02-18T09:48:44.980Z info hostd[2101552] [Originator@6876 sub=Vimsvc.TaskManager opID=f58d9e7c user=vpxuser] Task Created : haTask--vim.event.EventHistoryCollector.readNext-19709961

  • From the hbrsrv.log multiple sync start and aborts are observed:

2024-12-04T15:26:39.609+13:00 info hbrsrv[01166] [Originator@6876 sub=Delta groupID=GID-ab1234a1-1234-12a1-ab12-##### opID=hsl-2830df45] New instance start triggering old instance abort (group GID-ab1234a1-1234-12a1-ab12-#####)
2024-12-04T15:26:39.609+13:00 warning hbrsrv[01166] [Originator@6876 sub=Delta groupID=GID-ab1234a1-1234-12a1-ab12-##### opID=hsl-2830df45] Aborting in-progress instance for group GID-ab1234a1-1234-12a1-ab12-#####

  • These aborts are only seen for specific VM GID. Due to the sync aborts, the incremental sync never completes.

  • On comparing the vmx file of a working VM and the non-working VM, it can be seen that the "extendedConfigFile" parameter is populated with the vmxf file for the working VM, and empty for the non-working VM:

Working VM's vmx file: extendedConfigFile = "vm_name.vmxf" 

Non-working VM's vmx file: extendedConfigFile = ""

  • Even though the vmxf file is present in the VM home folder, the "extendedConfigFile" parameter is not populated with this information for the non-working VM.

  • An issue with the vmx file of the VM missing the value for extendedConfigFile parameter.

Resolution

Workaround:

NOTICE: Ensure you have valid backup for the VM and VMX before performing this.

  1. Remove the VM from replication.

  2. Power OFF the VM.

  3. Ensure that the vmname.vmxf file is present in the VM home folder.

  4. Then, edit the vmx file to add the name of the vmxf file in the "extendedConfigFile" parameter.

  5. Power ON the VM and verify if it is working fine.

  6. Configure the VM for replication and verify if the replication completes.

Powered by Wolken Software