Failed backup but why?

Example output of a log-file is a little more readable, but also the warning amount isn’t capped.

2024-06-26 07:13:46 -04 - [Warning-Duplicati.Library.Main.Operation.Backup.FileEnumerationProcess-FileAccessError]: Error reported while accessing file: C:\backup source\no_administrators\
System.UnauthorizedAccessException: Access to the path '\\?\C:\backup source\no_administrators' is denied.
   at System.IO.__Error.WinIOError(Int32 errorCode, String maybeFullPath)
   at System.IO.FileSystemEnumerableIterator`1.CommonInit()
   at System.IO.FileSystemEnumerableIterator`1..ctor(String path, String originalUserPath, String searchPattern, SearchOption searchOption, SearchResultHandler`1 resultHandler, Boolean checkHost)
   at System.IO.Directory.GetDirectories(String path)
   at Duplicati.Library.Snapshots.NoSnapshotWindows.ListFolders(String localFolderPath)
   at Duplicati.Library.Utility.Utility.<EnumerateFileSystemEntries>d__23.MoveNext()

How many warnings did you have? It would have been in a yellow popup at the screen bottom.
Is everything in the backup from Netware drive? Can you backup a single file without warning?

What do you mean by “extended file open error messages”? I showed the code, if that helped.
Duplicati is an application program now running on .NET Framework running on Windows API.
IOException means Windows perceived an I/O error on the Netware access, and passed it up.
You can probably see the driver level activity in Process Monitor, to help with Netware support.

EDIT:

System Error Codes (0-499) (WinError.h) - Win32 apps | Microsoft Learn

suggests Netware caused Windows to return an ERROR_NOT_SUPPORTED, numerically 50.

Thanks I will look at sysinternals to see what is happening.

The netware/OES driver is simply a network shell that provides extended permissions, rights, etc to that server.

As for the decompression, files are compressed on the netware servers. It is a process that runs in the background to save space for infrequently used files. When a backup program tries to use a file that has not been opened in a while, the netware server needs to decompress the file. It may be that if you call the file open and the open times out, this may be the cause.

Should we test this as a CFS share versus a mapped drive?

Thanks. Please see my other email reply. .Net does not handle extended file ops from other drivers well. We have seen this before. Let me see the details in procman and go from there.

As you can see from the screen shot below, these two folders have over 35,000 files in 3000 folders. Not sure why we would have ~500 warnings but will look at procman for details. Thanks for your help! Cool program if we can tweak it a bit for our needs. Thank you.

I don’t know what you mean by a shell. Are you saying it doesn’t actually access remote files?
From what I can find, you have at least the option to not use Windows protocols, but run NCP.
That’s how access gets done, right? It’s not like there’s SMB or CIFS that it’s “just” extending?
OES 2023: NCP Server for Linux Administration Guide
NetWare Core Protocols Documentation

This is why I suggested testing a single file to be nice and easy. Can it handle that file?

If you mean CIFS the SMB dialect with direct Windows support, you could certainly test:

Microsoft SMB Protocol and CIFS Protocol Overview

I’m glad you know that. You could probably get PowerShell confused then, as it’s .NET.
If there’s a problem at the .NET level, that’s below Duplicati, so maybe can’t be solved.

Neither the size nor the file counts seem high. I’m not sure where you get 500 warnings.
That’s less than all the files, so maybe there’s some pattern, or just temporary overload.
Either getting a log file and looking, or testing different areas and sizes might be helpful.

The screenshot does explain how OES appears to be an extension, but I think the tabs
that don’t say OES are actually provided by NCP, though tab doesn’t specifically say so.
Windows has some sort of adaptability where its view changes based on what’s around.

Channel Pipeline is complex, but the 20 warnings in the default log are not opening files.
Enumeration is primarily just a walk of the source trees, so I’m not sure how that can fail.
One possible guess is that the file attribute filter got upset, but there are few clues so far.
Process Monitor might have some, and (less likely) a stack trace might give a little more.

How do I specify just ONE file. The program will only let me get to one FOLDER. Please advise! Thanks!

You expand folders and check it. If you can’t expand a folder, something else may be wrong.

image

is how I just selected a single file for a test for a support issue about backing up locked files.

EDIT:

Click on the triangle pointing at the folder icon to expand a folder. Triangle then points down.

Thanks…but this version of the code does not allow me to select individual files. See the screenshot below. All I see are folders and subfolders. There are no individual files listed even though I know there are files in these folders!

Also, thanks for working with me on this. I appreciate your efforts! Rich DeVito

What is the version, .e.g. in About screen?

image

You are on the Destination screen. Try the Source screen to select things to backup.
If you still can’t find individual files, see if something local to the PC works differently.

Good Grief… Source → Destination!

I don’t use a reverse polish notation calculator any more! :blush:

Sorry. Will look closer next time…but consider a swap in source and destination.

How do I stop a job that continues to run?

As you can see from this screen shot, the system claims to be in verify mode…but it never finishes. If I cancel the job, nothing changes on the screen.

I want to test the single file per your request but I never get to that point of being able to run it.

As a followup, I used wireshark to look at what was going out of the ethernet card and no traffic is coming from the machine. The only traffic is the ping to the storage server. See below.

C:\Users\radjr>ping storage2200.is.cc

Pinging storage2200.is.cc [67.217.57.194] with 32 bytes of data:
Reply from 67.217.57.194: bytes=32 time=13ms TTL=55
Reply from 67.217.57.194: bytes=32 time=10ms TTL=55
Reply from 67.217.57.194: bytes=32 time=9ms TTL=55
Reply from 67.217.57.194: bytes=32 time=12ms TTL=55

Ping statistics for 67.217.57.194:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 9ms, Maximum = 13ms, Average = 11ms

There is no screen shot. Are you sending posts by email (looking at top right of post)?
Maybe that can’t handle whatever you are sending, and you might have to do via web.
If you mean status bar says Verifying backend data, it’s before and after backups.
Verifying backend files is possibly where things are. Does it seem after some backup?

Perhaps you can still find a clue in About → Show log → Live → Information or above.
Although there is some SQL, perhaps more likely is the FTP destination has got stuck.
You were having destination troubles, maybe due to huge files. Any update on that fix?
I’ve never even been able to get any answer on files on the destination. Please check.