Restore jobs initiated from Enterprise Vault Migrated data fail reporting an error "0xe000150a - The operation was terminated by the database application."

book

Article ID: 100014692

calendar_today

Updated On:

Description

Error Message

Bengine debug log shows the following

[ndmp\ndmpsrvr]      | Session 2 started
[ndmp\ndmpcomm]      - ERROR: ndmp_process_messages: unexpected reply message: 0x502.
[ndmp\ndmpcomm]      - ndmp_readit: Caught message on closed connection. Socket 0xc7c len 0x0
[ndmp\ndmpcomm]      - ndmp_readit: ErrorCode ::

Bermeote debug log from the Enterprise Vault Server

[ndmp\ndmpcomm]      - ndmpConnectEx: unable to connect using NetworkOptions to 10.x.x.X
[BESocket]           - BETCPConnection::CreateConnectionFromHostAndPort: Remote Host: "10.x.x.X": There were no addresses returned, belonging to family: IPv6.
[ndmp\ndmpcomm]      - ERROR: ndmp_process_messages: No handler for message 0x502
[ndmp\ndmpcomm]      - NrdsAdvertiserThread: negative (purge) advertisement cycle started.
[ndmp\ndmpcomm]      - NrdsAdvertiserThread: no purge is pending.
[ndmp\ndmpcomm]      - NrdsAdvertiserThread: negative (purge) advertisement cycle complete.  Waiting 5 minutes before advertising again.
[ndmp]               + ndmp_comm.cpp (2172):
[ndmp]               | NDMPAgentConnector::GetNotifyConnectedStatus : Wait for notify_connected timed out after 30 seconds
[ndmp]               + ndmp_comm.cpp (3064):
[ndmp]               | ndmpConnectEx: Error 8 validating TCP Connection
[BESocket]           - @@@@@@@MyCloseSocket called with sockfd = 1528(0x5f8)     retval = 0
[ndmp\ndmpcomm]      - GetClientCertificateByIPReverseLookup: Can't connect to ndmp connection
[ndmp\ndmpcomm]      - ndmp_writeit: NDMP Socket Error when trying to write message. Socket 0x5f8 len 0x1c
[ndmp\ndmpcomm]      - ndmp_writeit: ErrorCode :: 10038 : An operation was attempted on something that is not a socket.
[BESocket]           - @@@@@@@MyCloseSocket called with sockfd = 1528(0x5f8)     retval = -1
[ndmp\ndmpcomm]      - ConnectToNameServer: failed to get active client certificate for '10.x.x.X'

 

 

Cause

If there are multiple restore jobs queued in Backup Exec awaiting a device to be available then it can result in a Timeout from Enterprise Vault as it does not get the requested data.

Resolution

Increase the Timeout Value for the Backup Exec Agent in the migrator settings of the partition. Check the following document

https://www.veritas.com/docs/HOWTO23550

 

 

 

Applies To

Veritas Backup Exec 2014 and above

Veritas Enterprise Vault 10.3 and above

Issue/Introduction

Veritas Enterprise Vault is configured to migrate data to Backup Exec Tape location. On recalling multiple items or running a DA Export job, Enterprise vault may send multiple restore requests to Backup Exec to retrieve the data from the Tape drive. This creates multiple restore jobs in Backup Exec. If the jobs have to wait for longer period for the Tape Device to be available then some of the restores may fail reporting the error Final error: 0xe000150a - The operation was terminated by the database application. There is no specific trend as when this failure may occur and is intermittent.

Additional Information

ETrack: 3683357