Troubleshooting issues with file hash calculation
Log files:
: Linux:
/usr/openv/logs/nbwebservice: Linux:
/usr/openv/netbackup/logs/bpdbm: Linux:
/usr/openv/netbackup/logs/bprd: Linux:
/usr/openv/netbackup/logs/admin/:
/<file hash storage path>/hashdb/log
The following error is encountered when performing a file hash search.
network connection broken
File hash search results in a network timeout (initial failure error code 4106), and subsequent searches continue to fail with error code 40.
This issue typically occurs when the file hash server services are not functioning correctly or become unresponsive.
:
Restart the file hash server services to restore connectivity and allow hash search operations to complete successfully.
The following warning message appears in the job activity monitor:
File hash transfer queued for image
The above issue may be due to one of the following reasons:
File hash server is not configured correctly.
An earlier hash-only image transfer did not complete successfully.
Hash-only images exist but were not transferred to the file hash server.
Check the following logs on the primary server:
bprd logs for entries similar to:
immediate transfer of hash-only image <backup-id> to file hash server failed
bpdbm logs for entries similar to:
failed to create hash-only bid info file
:
Verify hash-only image creation:
Check for all hash-only images: /usr/openv/netbackup/bin/admincmd/bpimagelist -hash_only
Check hash-only images for a specific parent backup: /usr/openv/netbackup/bin/admincmd/bpimagelist -hash_only -parent <parent_backup_id>
If no hash-only images are listed, then configure the file hash server by following the procedure described in the following section:
Retrigger the hash calculation job.
If hash-only images are listed but not available on the file hash server, then manually initiate the transfer from the primary server using the following command:
/usr/openv/netbackup/bin/goodies/nbfhsmgr -transfer
After the transfer completes, start hash database processing on the file hash server using the following command:
/usr/openv/pdde/pdcr/bin/fhdbctl --op db_process start
The following warning message appers in the job activity monitor:
File hash transfer queued for image
The above issue may be due to one of the following reasons:
A previous hash ingestion process is still in progress or did not complete successfully.
The file hash server database is locked by another ingestion job.
Check the file hash logs for the following message:
failed to acquire lock within 40s for systemtest5230h_1777303980.f (waited 40s), another ingestion in progress
:
Manually initiate the hash-only image transfer from the Primary Server using the following command:
/usr/openv/netbackup/bin/goodies/nbfhsmgr -transfer
Start database processing on the file hash server using the following command:
/usr/openv/pdde/pdcr/bin/fhdbctl --op db_process start
Monitor the file hash server logs to ensure that the ingestion completes successfully before retriggering the hash calculation job.
Failed to validate the compute host. The following error message is displayed on the NetBackup Web UI:
Failed to validate the compute host.
The above issue may be due to one of the following reasons:
The compute host was added and deleted multiple times.
The compute host type is not correctly configured for malware scanning.
Check the nbwebservice logs on the Primary Server for entries similar to:
Exception occurred while binding the queue, closing the connection if opened. java.io.IOException at com.netbackup.messagequeue.service.MQServiceUtil.bindQueue
:
On the compute host, edit the
/usr/openv/netbackup/bp.conffile.Add the following entry:
MACHINE_NBU_TYPE = NB_SCAN_HOST
Restart NetBackup services on the compute host:
Stop: /usr/openv/netbackup/bin/bp.kill_all Start: /usr/openv/netbackup/bin/bp.start_all
Re-validate the compute host from the Web UI.
More Information