What follows is a list of Yosemite Server Backup errors that I have encountered while using the software, along with a few notes or comments from me. Maybe they will be useful to you, and maybe not. I'm not intending on being your tech support, but was always frustrated that there wasn't a whole lot on the Internet talking about these.
Note that I was told once by Barracuda's tech support that my virtual libraries should only have 1 or 2 devices (the "drives" setting directly above the number of storage slots). Even though it's more convienent to have more in many cases. (For example, deleting some old media while your "device" is in use by a backup job that is complaining it is out of disk space.) The reason he stated this was because he said it was faster (which I could argue has some merit). The real reason, I believe, is because they have some programming bugs when it comes to more than one device. See error 4020 as one example where it will show up only with multiple devices. So, if you are just constantly getting some random errors with a virtual library, see if using one "drive" for your virtual library helps.
In the future, I hope to get more of a page set up where others can add errors and the fixes or workarounds that they know.
"An invalid or unsupported command has been received. Please make a note of this error. To locate additional information about this error, you may search our Knowledge Base at http://www.barracuda.com/support."
Seen when pressing Run on a restore job I had just made. Trying again allowed it to run without issue.
From the error description, "This error occurs during failure modes when a handle value is incorrectly used." Seen on a virtual library that was stopped and had error 38 when I tried to access the properties of it.
Seen again on a virtual library when I tried to stop it after I had already stopped it. Probably just a small UI bug that shows the Stop button when it shouldn't. If I click away and then back, it changes to Run.
Also once when trying to see the properties of a Virtual Library Device when it had error 4013 and that was offline.
The file probably existed while creating the backup list, and then was deleted before being backed up. Linux filesystems seem to give error 3020. Can also be seen in the Media Maintenance of a virtual library if a media file is missing.
A file that was required could not be opened. Seen during a backup. Not sure what caused it.
Unknown if this is talking about a particular file, or the virtual media files.
Seems to be talking about where the virtual media is saved (not the source, but the destination of the backup).
There was a problem writing a to the backup device for some reason. Once seen when the virtual library file had error 208 during a backup.
Seen on a virtual library after disassociating and deleting some invalid media and also going from 2 "devices" to 1. I didn't try too hard to fix it since I didn't need to (I was able to just delete and recreate the virtual library since I didn't need any of the information in it.)
Seen on a virtual library. The virtual library device was stopped. Start it.
I rebooted a server while it was doing a backup job and it gave me this message.
Seen when trying to log in, but don't know why. I could ping the machine and TCP port 3817 was accessible on the master server from the machine I was trying to log in from. Simply quitting the GUI and relaunching it allowed me to log in without error.
Also saw this once after a power outage where it was likely the backup target wasn't fully booted yet.
The server or some device closed a connection.
"Yosemite Server Backup is unable to connect to a remote host."
Seen when some part of the program can not connect to another machine, or the connection was lost in the middle.
"Yosemite Server backup was unable to detect a TCP/IP address on this system." Client trying to find a server by name, but name doesn't exist in DNS.
Seen once when I rebooted a server for Windows Updates while it was in the middle of a backup.
Also seen when trying to log in to a Yosemite server that was not the database/domain controller.
Seen "when trying to rename, or create" a media with the same name as one that already existing in the folder.
Something could not be located. This is a generic error returned whenever Yosemite Server Backup fails to locate an object.
In my case, I had just moved the database over to a new computer, and the virtual library's paths were now incorrect. I ended up having to recreate the virtual library and associate all the foreign media.
I also saw it once after a catalog restore and trying to connect to a Virtual Library Device.
Seen when you cancel a job. Also sometimes happens automatically per server in conjunction with error 1040 and error 4067.
Might show up for several different reasons. I've seen this whenever I complete fill up a removable hard disk with a copy job (doesn't matter if the hard disk is connected to a Windows or Linux machine).
Seen on a Linux machine when I tried to stop the YTBackup service while it was still starting up (start pending).
This is the error that is given when I am trying to backup a computer that normally gives error 5093 or error 5094, but I am backing it up to another device besides the attached USB hard drive. See error 5094 for more information.
Probably similar to error 16, although the whole directory the file used to be in was gone instead.
This computer cannot contact the backup domain server. Perhaps port 3817 is being blocked, or the computer is offline. Check your install path\config\ytconfig.ini for the default server (although you can manually type it in before logging in as well).
Unknown what caused this error.
Saw this error once when I rebooted a computer that was about to be backed up. Or maybe it had a virtual library on it that was about to be used.
I have no idea what caused this error. Seen a few times during backup jobs of a Linux host. Restarting the host seemed to fix the error, although restarting the Yosemite service would probably be sufficient.
"Mail server refused connection. An invalid hostname was received. Please check the hostname and try again."
The mail server is responding, but refusing to send the email. Depending on how your mail server is set up, it can be picky about whether or not it's being sent from the master server or not, since the test email will be sent from whatever computer you are logged on to.
The Yosemite user that you are logged in as does not have permissions to this object.
Seen when having troubles using a virtual library. I had just changed the path that the virtual library was using on the backend and got this message when trying to run a backup job. Running the backup job a second time (no changes) seemed to fix the problem.
The user name you entered doesn't exist.
"The password you entered is invalid." Type in the right password.
Unknown.
Seen when an alert showed that it could not find the media (since the tape was not in the drive). Upon hitting cancel on that alert, the job failed with this error. Can also see this error with virtual media if the virtual library is offline.
This user exists and has no password, but has "Require password" checked. You must either log on as a user that can set the password for this user and do so, or log on to the master server directly.
You need to insert a tape into your tape drive. If you are dealing with virtual libraries, then this can be seen sometimes if the device has gone offline since the job started (although sometimes other error messages are shown instead depending on exactly when it went offline).
The other time you will see it with virtual libraries is when you are all out of Storage Slots for your virtual media. Go increase the number of storage slots on your virtual device.
My master server had a bad hard drive where the catalog was stored and I received this error. I fixed the hard drive error, and then restarted the master server. The first time I logged on to Yosemite Server Backup, it ran through a repair process (took about an hour) and it seems to have worked. I suppose it depends on how corrupted the database became as to whether or not it will be able to repair.
An incorrect tape is in your tape drive. If using virtual media, then an incorrect media is "loaded" in the virtual "device". For removeable media, it could be set to have multiple sets and does not want to overwrite the prior media.
Probably seen when the server was told to reboot, but a job was in progress. Also seen when a running job is stopped or stopped due to an error.
Seen on a virtual library. Not sure what caused it since there were plenty of free storage slots. Ended up having to delete the virtual library and then recreate it (pointing to the same storage locations) in order to fix it. Obviously had to reassociate all of the "foreign" media.
This file was not backed up since it was locked by another process.
Seen during backups if the file existed when the backup list was created, but did not exist by the time the process got around to trying to back it up. The file then gets counted as "Not attempted". See also error 16, which seems to be the same type of error for Windows filesystems.
This can also be seen during a restore, but is not common and usually involves a little known bug. The restore operation basically creates an empty file with the correct name, and then fills up the file with the restored data. If it's not able to create the empty file, but thinks that it did, and then tries to write to the file, then you'll see this error during a restore.
Seen during a restore where the target destination did not have space to restore the file.
An error seen while backing up the catalog. Don't know what caused it.
Seen when trying to backup WMI Repository (part of the System State). Never saw it again.
If a file to be restored is locked then this message will be seen.
As part of Restore Catalog, you'll need to restart the Yosemite Server Backup service. Alternatively, you can restart the entire system.
Can see this message if you try to view the properties of a virtual library while the virtual library is in use by a backup or restore job. From the description: "The current task cannot use the device because the device is in use by another task."
Seen on a virtual library device that is having some trouble, like error 4428. In the end, I had to delete the virtual library and recreate it. Luckily I was in a situation where I could do that. Perhaps support would have a better solution.
I saw this again when selecting a Virtual Library Device that had Status: Offline, this was the "Initialized state". I didn't have to delete the virtual library though, just starting it allowed this error to go away.
Seen while identifying virtual media. Seems pretty random. You can run it again and it can be different storage slots that have the error. I have never seen it while identifying a single storage slot, only with groups of storage slots. Similarly, if you change your virtual library to have only 1 "drive" or "device" then it won't show. Seems like a parallel thread programming bug to me.
Seen on a USB device when trying to identify the media. In this case (on a Linux machine), the USB device was plugged in, but not mounted and had been manually programmed in to ytconfig.ini.
Seen on a tape drive where the tape or drive was going bad. Eventually turned into error 4075.
The tape is write protected, or in my case, the USB drive was having a little trouble and thought it was write protected.
Seen during a restore. The virtual media may be corrupt. Looking at the file identifiers for the virtual media, I saw that I actually had about 70 files (1 GB each) but the library was listing it as a 2 GB media in the media maintenance. I disassociated the media and then associated it and it correctly showed the larger size. Running the restore again worked without a problem.
Seen when identifying a removeable device that was used with a different catalog/installation of Yosemite Server Backup. Also sometimes seen after deleting media when there were multiple virtual library devices.
Seen after erasing media that previously had error 4054. Also seen once after erasing a freshly formatted media.
Once seen when trying to import media in a Virtual Library Device but the storage slot is "Invalid - blank media." Must first Associate Foreign media in the Media Maintenance.
Seen when the virtual library was out of available disk space.
Usually accompanied with another error as well. From the description: "The operation has stopped because an error has occurred while reading or writing data which the drive cannot correct."
Your tape is bad. From the description: "Your data is at risk: 1. Copy any data you require from this tape. 2. Do not use this tape again. 3. Restart the operation with a different tape."
Seen in conjunction with errors 4074 and 4075. From the description: "The tape is from a faulty batch or the tape drive is faulty: 1. Use a good tape to test the drive. 2. If the problem persists, call the tape drive supplier helpline."
The manual eject button on the tape drive was pushed while this job was still in progress. From the description: "You cannot eject the cartridge because the tape drive is in use. Wait until the operation is complete before ejecting the cartridge."
Seen on a file when the job returned error 4409. Also seen on a job with an overall error 18. Also seen on a job with an overall error 4436.
This one is no fun. Probably means that your backup is corrupted. You could maybe try a repair in the devices area if it's a virtual media, not sure how it works for actual tapes.
Either the object changed between being backed up and being verified, or your media may be going bad.
Saw this error while trying to repair a virtual media that had error 16. From the description: "A read or open operation has failed on the selected virtual media extent file."
Seen in the virtual library maintenance area. Running a repair appeared to fix this error several times (although I have not tested it beyond that).
Additionally, I saw this once where a repair did not fix it (it did not give an error either, just went back to the repair screen). Looking at the file identifiers and the actual files, there was the {file-identifier}-00000000 and then there was {file-identifier}-99999999 with a very small 1 Kb filesize. It wasn't actually all 9s, it was just a high number that was out of sequence. Deleting the high numbered, out-of-order file seemed to fix the problem.
From the description: "The virtual library device is reporting that a selected virtual media does not exist. Refresh the virtual media list, and try the operation again." Seen while trying to view the Media Maintenance section of a virtual library which had a storage slot that was "Empty" and wouldn't let me use the library since it claimed it was not initialized. I had previously had backup jobs failing claiming error 1037 and tried to then delete some older media.
In the end, I had to delete the virtual library and then recreate it, but luckily I could do that in my situation. Perhaps support can offer a better solution.
Another time, disassociating the media and then deleting it caused the GUI to crash, and then this error just moved to another media.
I have seen this when trying to repair a virtual media that had error 16. From the description: "A failure occurred during a read operation of the virtual library device inventory control file."
Seen during a repair of a virtual media that had error 4434. You also may see this if your Yosemite user doesn't have Write rights and you are trying a repair.
Seen while trying to repair some virtual media that had error 16. From the description: "The repair process failed to repair the virtual media because there are no extent files present to operate on."
A repair did not seem to work at first (failed with error 4431), but did on another media with the same error. Have also seen error 4435 when trying to repair this error.
After disassociating some virtual media with this (4434) error, it changed to error 4426 (and the size it listed for the virtual media changed). A repair then appeared to work and I could reassociate the media. Also, the fix detailed for error 4426 worked for this error once as well.
Seen when trying to repair some degraded virtual media with error 4434. From the description: "The virtual library device is reporting a internal error. Please contact support for assistance in resolving this error."
Seen during backing up to virtual media. Unknown was causes it.
You cannot associate foreign media that is in a degraded state. It must be healthy. From the description: "Import of degraded media is not allowed. Only virtual media marked as healthy can be imported. To set a virtual media status to healthy, all the virtual media extent files must be seen by the virtual library device. In order for a virtual library device to see virtual media extent files, each path where these extent files exist must be added to the storage folder path list. If you are sure that all the proper storage folders have been added, and you want [to] attempt to use this virtual media, run a repair operation, and try again. Be warned however that a repair operation is a one way process, and can lead to data loss."
Yosemite Server Backup crashed after showing this message once.
Seen while trying to delete some "Foreign, Degraded" virtual media with error 16. From the description: "It is likely that this file is in use by another process, or the file no longer exists." You may have to go delete it manually based on the file identifier. I saw this where the media was missing all of the low numbered files, but had a small high numbered file left over.
Seen with a virtual library, although I cannot remember the circumstances of making this happen.
This was also seen once in the log files while the virtual library itself was stuck with error 1616.
Pretty self explanatory. The virtual library is not able to write to the storage paths that it was configured with.
Yosemite Server Backup was unable to create a Disaster Recovery boot image because the WinPE files have not been installed on this machine.
Other files are still backed up with this error message, just not a Disaster Recovery boot image.
This error kept occuring on a particular server each time I backed it up. Restarting the server seemed to fix it. Unknown was caused it initially.
Unknown.
Happens on my really big server sometimes. I'm not sure if it's the VSS writers timing out, or Barracuda with some sort of timeout. All my VSS writers seem to indicate no problem, even right after getting this error, so I'm not sure.
Probably talking about the VSS writers on this Windows XP machine. I seem to recall reading somewhere that Windows XP only allows one snapshot to be created at a time, where the Windows Server versions allow more. I could be remembering wrong though. A reboot of the machine throwing the error seems to fix this problem, although perhaps you could just check and make sure two backup jobs (that use VSS, like System State) aren't trying to run at the same time.
I see this error on Windows Server 2003, and error 5094 on Windows Server 2008 and 2008 R2.
On Windows, the VSS services returned an error when asked to make a snapshot. Also, oddly enough, VSS will throw back errors consistently if you have a large USB hard drive plugged into the computer.* You don't even have to be using it for your backup, just being plugged in is enough.
*Maybe someday I will go dig up the link, but people were talking about how Windows expects a default sector size of 512 bytes, but some of these larger hard drives don't use that and it causes VSS to fail -- and therefore Yosemite won't back up that computer. Seen on Windows Server 2003, 2008, and 2008 R2. I haven't tested Windows Server Backup (the built-in backup program with Server 2008) to see if it would also have troubles trying to use the VSS writers. According to Wikipedia however, it would fail.
To see if you have a hard drive with a sector size of 4096 bytes instead of the old default of 512 bytes, you can run the following command on Windows: wmic DISKDRIVE get bytespersector, caption
To my knowledge, Barracuda has not addressed this issue, instead saying it is a Microsoft problem (which is really kind of is, since it's the VSS writers that are failing.) "vssadmin list writers" could help you narrow down which VSS writer(s) are failing.
This can happen if two backups jobs are running on the same server at the same time and they are both asking the computer to create a VSS snapshot simultaneously. Also Shadow Copies uses VSS.
Last updated: 2021-08-11