HELP Backup to Azure fails at "Waiting for upload"

Support
1

I have a backup configure to azure blob and it used to work well, but some days ago, i received this error on email

Failed: The request was aborted: The request was canceled.
Details: Microsoft.WindowsAzure.Storage.StorageException: The request was aborted: The request was canceled. ---> System.Net.WebException: The request was aborted: The request was canceled.
   at System.Net.ConnectStream.InternalWrite(Boolean async, Byte[] buffer, Int32 offset, Int32 size, AsyncCallback callback, Object state)
   at System.Net.ConnectStream.Write(Byte[] buffer, Int32 offset, Int32 size)
   at Microsoft.WindowsAzure.Storage.Core.ByteCountingStream.Write(Byte[] buffer, Int32 offset, Int32 count)
   at Microsoft.WindowsAzure.Storage.Core.Util.StreamExtensions.WriteToSync[T](Stream stream, Stream toStream, Nullable`1 copyLength, Nullable`1 maxLength, Boolean calculateMd5, Boolean syncRead, ExecutionState`1 executionState, StreamDescriptor streamCopyState)
   at Microsoft.WindowsAzure.Storage.Core.Executor.Executor.ExecuteSync[T](RESTCommand`1 cmd, IRetryPolicy policy, OperationContext operationContext)
   --- End of inner exception stack trace ---
   at Duplicati.Library.Main.Operation.BackupHandler.HandleFilesystemEntry(ISnapshotService snapshot, BackendManager backend, String path, FileAttributes attributes)
   at Duplicati.Library.Main.Operation.BackupHandler.RunMainOperation(ISnapshotService snapshot, BackendManager backend)
   at Duplicati.Library.Main.Operation.BackupHandler.Run(String[] sources, IFilter filter)
   at Duplicati.Library.Main.Controller.<>c__DisplayClass16_0.<Backup>b__0(BackupResults result)
   at Duplicati.Library.Main.Controller.RunAction[T](T result, String[]& paths, IFilter& filter, Action`1 method)
Request Information
RequestID:
RequestDate:
StatusMessage:

I then deleted the backup and start it from scratch. But again, after uploading 40GB (full backup size), i have the same error on “Waiting for upload” stage.

I tested with a new backup and it works well with small files. When i put some “big” file (100mb or more) it gave the same error. On this “test backup”, i solved the issue changing the “block size” from 50 to 20mb. But in the “production backup” of 40GB this trick doesn’t work.

I noticed that in the end of backup, duplicati tries to upload a 80mb file, and then it gave the error message, tries to rename the “block” and the backup fails.

Questions:

Why are duplicati uploading a 84MB file, if my settings are configured to 20MB block size?

How can i solve it or give more informations?

Screenshots:

duplicati
duplicati.png1335×579 14.9 KB

2

The final file is the “file list”, and if you have a lot of files, this can be large. There is no support for splitting this list over multiple zip files, unfortunately.

3

Thanks for the answer!

Theres a workaround to this? I think it occurs because my slow internet connection. The strange part is: i have this backup configured for months and always works great.

If we solve the problem with “block size” larger than 20MB, i think the backup will work again.

Can i give you more information or something?

Thanks!

4

Not really, but if you split your backup into multiple backups, each backup will have a smaller file list.

5

I split the backup and it works! Thanks.

6

Hello,

I have backup of size only about 1GB and dblock-size set to 20MB and I am getting the exactly same exception in 3/5 tries of backup. I think it is caused by slow upload speed of ADSL line and most probably some timeout of Azure SDK is causing this.

I have read somewhere on the Internet that this can be tuned by changing BlobRequestOptions (MaximumExecutionTime, ServerTimeout) of Microsoft.WindowsAzure.Storage.Blob, but this is not made available to change from Duplicati.

I am thinking of adding this option to the Duplicati myself and perform some tests, but I wanted to check here if somebody maybe already found another solution to this error?

Thank you

7

To verify your suggestion, you could setup rclone to your backend, and change duplicati to use rclone. I believe rclone has the advanced settings made available.

8

If that’s the case (and you have the issue proposed above) then somebody with an Azure account should be able to replicate the issue be setting a dblock size to your final “file list” size then use the built in throttle feature (or external tool) to simulate a slow connection.

If you’re using a newer canary or experimental version, I believe the are now http-timeout and http-retry Advanced parameters that might help get around this issue.