11-22-2019 05:31 AM
Backup fails with this error "The directory or file was not found, or could not ba accessed"
This is happening for multiple directories.
Directories exist and are accessible via Windows Explorer and the selection screen in Backup Exec.
Veritas Support have no idea and just keep asking for new job logs and VQA reports.
Any ideas?
11-22-2019 06:55 PM
It could be that your volume is corrupt.
Do a "chkdsk /f" on the volume to correct any errors.
11-25-2019 06:19 AM - edited 11-25-2019 06:20 AM
Hi pkh, thanks for your reply.
The volume contains CAD-specific data which is being accessed on a daily basis and none of the relevant users are reporting problems of any sort so it's doubtful the volume is the cause of the problem.
Also, when the job fails, there are always differing directories which are reported as being the problem. I have run this job both in office hours and in the middle of the night to rule out file locks although the job is confgured to use VSS to avoid this regardless of the time it is run.
I have also tried using modified time vs archive bit and LTO tape vs disk storage, all with the same results. Other applications have no problems accessing the directories or the data they contain so this points to the problem lying with Backup Exec.
11-25-2019 09:35 AM
0xE00084AF The directory or file was not found, or could not be accessed.
You most likely have VSS problems.
Check the Application and System Logs right around the time you see the job log starting skipping files. Compare job log to Event Logs on the system being backed up.
Specifically VolSnap or VSS events.
11-26-2019 12:20 AM
Hi Steve, thanks for you reply.
I have searched the Event Viewer as you have suggested but have found no entries in the Application or System logs for VolSnap or VSS during the times of the job (screenshot attached - unfortunately I seem to only be able to upload one screenshot). Other jobs backing up from the same volume using VSS are having no problems so I would think it doubtful VSS is the cause.
I have run the job again and this time the error is different: 0x000fe36 - Corrupt data encountered.
Again, I must reiterate that we are as certain as we can be that there is no corrupt data in this volume. Users accessing this data on a daily basis have reported no problems and the data is accessible through Windows Explorer and the Backup Exec selection window in the job definition.
All evidence so far is pointing to Backup Exec I'm afraid.
11-26-2019 09:20 AM
Can you upload a few job logs (One from the old error and one from the new error)?
11-27-2019 12:33 AM
Here's two logs, the first from 27/09/19
11-27-2019 12:34 AM
The second from the 27/11/19
11-27-2019 12:43 AM
I'd really like to know what Backup Exec considers to be "corrupt data".
11-27-2019 04:33 PM
These snips in the log are a good indication something has happened to the VSS snapshot or there is a VSS/VOLSnap issue on the server.
V-79-57344-34108 - An unexpected error occurred when cleaning up snapshot volumes. Confirm that all snapped volumes are correctly resynchronized with the original volumes.
6 November 2019 20:01:06 - V-79-57344-34108 - An unexpected error occurred when cleaning up snapshot volumes. Confirm that all snapped volumes are correctly resynchronized with the original volumes.
I noticed the job log indicates a remote agent server backed up by IP. This is where you want to look in the Event Logs if you havent already around 6 November 2019 20:01:06 time frame.
If you like DM me a zipped up copy of App and System logs in evtx format from that remote server.
Also I notiec the folder it was on was called Vaults - Is this Enterprise Vault Folders?
11-28-2019 01:17 AM - edited 11-28-2019 04:24 AM
Hi Steve,
No this is not Enterprise Vaults, it's data for a CAD storage and versioning application called DDM. The vault is the location for CAD part files.
I've tried backing up this location using Windows Server Backup on the SQL server and all the jobs so far have failed with VSS errors although everything I've investigated so far for VSS has checked out and looks okay.
Strangely though, I'm backing up the 1000 folders in the \Vault\Parts directory in batches of 100 in Backup Exec to see if there is any particular point of failure. So far they've all been successful despite the WSB failures and they're set to use the Microsoft Shadow Copy Provider in the job settings.
In addition (and relating to the error message about cleaning up and synchronising snapped volumes) another directory (\Vault\Drawings) on the same volume which was also failing with the same errors is now working.
Go figure...
I've looked through the Application and System logs on both servers on the 6th Nov from 19:00:00 to 21:00:00 for Warnings, Critical and Error level events but there is only a single hit and this is from CAPI2, ID 513 on the backup server.