-
Title
Agent Snapshot fails with 'The specified network name is no longer available.' -
Description
The following errors appear in the agent AppRecovery or AppRecoveryPM logs:
ERROR 2014-03-19T16:17:32 [26] - Replay.Agent.Implementation.Transfer.TransferDataChannel (RemoteEndPoint=<IP>:<port>)
Sending of volume data at byte offset 0x3fbb4000, byte length 0x400000 failed
System.AggregateException: One or more errors occurred. ---> Replay.Common.Contracts.Win32Api.Win32ApiFailedException: I/O operation - The specified network name is no longer available
at Replay.Common.Implementation.Win32Api.NativeIOReader.EndTransmitFile(IAsyncResult asyncResult)
at Replay.Common.Implementation.Win32Api.NativeIOReader.<>c__DisplayClass4.b__3(IAsyncResult ar)
The transfer failed: 'Call to service method https://<agentname>:8006/apprecovery/api/agent/driverchangelogs/~|~~|~%253f~|~Volume%257b90f0cd26-f844-46c3-99d2-a948f2cbf33d%257d~|~/ GET failed: Failed to open '\\?\Volume{90f0cd26-f844-46c3-99d2-a948f2cbf33d}\System Volume Information\AALog_{5d79ce66-4016-11e3-b4a5-00155dc93c03}_000015bc.log' - The file or directory is corrupted and unreadable'
Failed to open '\\?\Volume{90f0cd26-f844-46c3-99d2-a948f2cbf33d}\System Volume Information\AALog_{5d79ce66-4016-11e3-b4a5-00155dc93c03}_000015bc.log' - The file or directory is corrupted and unreadableThe issue may also manifest itself during Nightly Jobs, Check Recovery point action. Error is: "Recovery points of Agent <agentname> contains broken volume image(s)"
-
Cause
The most likely cause of this problem is disk corruption. The next likely cause is a left over shadow copy or agent change log. It's also possible when installed antivirus solution is interfering with agent service that is attempting to read changelog files in System Volume Information folder. -
Resolution
Solution 1:
Use the Windows Check Disk utility to check the hard drive for errors and repair any it finds. Keep in mind that a read-only CHKDSK will not necessarily find all the problems that are present on the disk, which is why you should include the repair switch (/R, which implies /F).
- Open a command prompt and run a repair CHKDSK by typing CHKDSK /R.
- NOTE: If an error message stating “The type of file system is NTFS. Cannot lock current drive. Chkdsk will not run because the volume is use by another process. Would you like to schedule the volume to be checked the next time the system restarts (y/n),” press Y and then press Enter. Restart your computer.
- If the CHKDSK shows some “bad sectors”, the disk may need to be replaced.
- If the CHKDSK finds other problems, rerun it until it no longer finds any. If you’ve run it four times and it’s still finding problems then the disk may need to be replaced.
Solution 2:
Follow the steps in Knowledge Article 118402 to remove all left over change logs
Solution 3:
Open Disk Management | right click a volume | select Properties
In the Properties dialogue box, select Shadow Copies | Settings
In the Settings dialogue box, change the Maximum Size Use Limit field to 320 MB and save. Once Windows has removed all the left over shadow copies in the System Volume Information directory, select the No Limit radio button and close all dialogue boxes.
Repeat these steps for all protected volumes.
Solution 4:
Check your network connections on the agent and core. If you have physical servers, make sure they are connected to the same switch. If you agent and/or core are a virtual machine, make sure your agent is on the same vLAN as the core. Force a snapshot after correcting any network reconfigurations.
If you are still having this issue after reconfiguring your network, please change the Agent's snapshot settings and force another snapshot:
- Open a command prompt and run a repair CHKDSK by typing CHKDSK /R.