"Cannot open the file as archive"

[roninpawn]

I know this sounds like a 7zip issue, or a corrupt image but stay with me here, because its stranger than it sounds, and may be something to do with clonezilla-util.

Story:

• Cloned my Win10 drive (all partitions) before a wipe out and rebuild.
• Attempted to mount the partitions as directories under drive L:\ using clonezilla-util.
• The ~500MB recovery partition mounts fine as \sda2, but the \sda1 (the old C:\ drive) is just empty.
• Checked the logs and found 7z error, "Cannot open the file as archive," given after indexing.
• So I abandon indexed mounting... Used clonezilla-util to unzip the \sda1 partition fully to sda1.img.
• Attempting to open the uncompressed .img, 7zip errors: "Cannot open the file sda1.img as archive"
• Check file with hex editor: Header is okay. But end of file cuts off with 100's of pages of 00 00 00 00s, and no obvious end-header where the data cuts off. (footer)

• Abandon accessing image file... Boot to same CloneZilla live environment that made the clone.
• Restore the whole image to an empty drive.
• Boot to a Fedora KDE Plasma Live environment.
• Use Dolphin to mount and access the restored partition, offline.
• Copy data to from the old C:\ to the new C:\ -- no problem.

So, to recap: I cloned my whole drive, and when I tried to mount it using clonezilla-util's indexing, the primary partition (223Gb's unzipped; 118Gb's as .gz) became unmountable, likely due to dropping the footer, along with a big chunk of the end of the raw data. When I finally cut my losses and restored to an empty drive using CloneZilla itself, the data was all there, and accessible offline, using a Fedora Live environment.

That allowed me to go ahead and mount my new C:\ drive and my old C:\ drive next to each other and transfer some leftovers that I wanted on the new disk. So I'm all set now.

Here's some important details: The version of CloneZilla I used to make and restore the disk clone is from early 2023. It's on an old Ventoy boot disk I made to cover a wealth of needs, and I haven't had cause to swap any of the iso's out for modern counterparts. So, old CloneZilla / old linux kernel may have made a weird image? ...I wouldn't think much has changed in gzipped disk clones in the last 2 decades, honestly. But I'm mentioning it, just in case.

The other thing to note is that this Win10 NTFS environment had undergone the mbr2gpt conversion. Again, just throwing it out there as a potential oddity that might've tripped things up. I suppose I could also mention that the cloned disk was an SSD. Not an NVMe; not a platter.

And when I made the original backup, my CloneZilla/parted settings were very basic. I went through the CloneZilla UI, chose the usual options to make a full disk clone... No wacky custom command line configs, or anything. Standard.

---

I saw how useful this tool might've been, and how much time I could have saved by indexing the compressed file instead of unzipping or fully restoring it, so keep at it! And I'll be back to give it another try, next time.

[fiddyschmitt]

Thanks very much @roninpawn! Wow, the perfect level of detail. I really appreciate it :)

I'm glad you managed to restore your files with another means.

mbr2gpt - interesting! It's very possible it contributes, as I just pass the raw partition/drive image to 7-Zip and hope for the best (given it supports so many file types & systems.) So I'll keep an eye out for it.

Old versions of Clonezilla are supported (2017 onward confirmed working. Unknown if older versions work). SSDs work fine to.

Again, many thanks for taking the time to write all this up! Legend.

Cheers,
Fidel

[Roki100]

got similar issue, used old clonezilla (probably 2018-2019 build), installed new os, and now I'm with unable to open archive shit, the biggest problem I have is the fact I don't have a spare 2tb drive to clone into and I've put all my hopes into this softest that has failed me...

[fiddyschmitt]

Hi @Roki100,

Could you please upload clonezilla-img when able?

[fiddyschmitt]

Hi @roninpawn,

The issue is likely fixed in v2.6.1

Could you please give it a try?

Thanks

[roninpawn]

Can confirm.

I still had the exact same image lying around, so I was able to test it again on the new version. And it did, indeed, mount both partitions of the drive. But with some errors.

[14:05:29 INF] [sda1.img] Starting 4 instances of 7zFM.exe.
[14:08:10 WRN] Non-fatal. Error while caching File List to D:\Program Files\CloneZilla-Util-2.6.1\cache\3f707238407a0dd1\sda1.Files.json: System.OutOfMemoryException: Insufficient memory to continue the execution of the program.
   at System.Text.StringBuilder.ToString()
   at Newtonsoft.Json.JsonConvert.SerializeObjectInternal(Object value, Type type, JsonSerializer jsonSerializer)
   at Newtonsoft.Json.JsonConvert.SerializeObject(Object value, Type type, Formatting formatting, JsonSerializerSettings settings)
   at Newtonsoft.Json.JsonConvert.SerializeObject(Object value, Formatting formatting, JsonSerializerSettings settings)
   at Newtonsoft.Json.JsonConvert.SerializeObject(Object value, Formatting formatting)
   at libClonezilla.Cache.PartitionCache.SetFileList(List`1 filenames)
[14:08:10 INF] [2025-07-10-16-img] [sda1] Contains 3,907,966 files.
[14:15:52 INF] [2025-07-10-16-img] [sda2] Retrieving a list of files.
[14:15:52 INF] [sda2.img] Determining optimal way to extract files from this partition.
[14:15:55 INF] [sda2.img] 7z.exe will be used to extract files from this partition.
[14:15:55 INF] [2025-07-10-16-img] [sda2] Contains 31 files.
[14:15:55 INF] Mounting complete. Mounted to: L:\
Running. Press Enter to exit.

This was the first error, thrown after indexing, just before mounting. And the only real problem here is shown in the timestamps. I was left with a console that looked like it ended in an error, with no indication that it was actually still working in the background, for almost 8 minutes, while it mounted /sda1. Honestly, had I not gotten a phone call to distract me, I think I'd of assumed failure and shut it down manually before I found it had worked.

A little text after the "Contains" line to affirm that a potentially very long process is now taking place - "please wait" - would let the user know what to expect -- and encourage me to check the task manager processes before I assume the operation died.

But like I say it DID eventually mount, and I was able to access my directory structure. The more problematic errors came when I tried to access my 'Documents' and 'Desktop' folders. (Win10)

[14:20:21 ERR] Error while navigating to L:\fcvfzepa\4ertoysv\sda1.img\Users\Roninpawn\Documents\Downloads - Shortcut (2).lnk: System.ArgumentOutOfRangeException: length ('-1') must be a non-negative value. (Parameter 'length')
Actual value was -1.
   at System.ArgumentOutOfRangeException.ThrowNegative[T](T value, String paramName)
   at System.ArgumentOutOfRangeException.ThrowIfNegative[T](T value, String paramName)
   at System.String.ThrowSubstringArgumentOutOfRange(Int32 startIndex, Int32 length)
   at lib7Zip.UI.ListviewNative.GetRowText(Int32 rowIndex, Int32 columnIndex)
   at System.Linq.Enumerable.RangeSelectIterator`2.ToList()
   at System.Linq.Enumerable.RangeSelectIterator`2.ToList()
   at lib7Zip.UI.ListviewNative.GetRows()
   at lib7Zip.SevenZipExtractorUsing7zFM.NavigateToFilename(Int32 pid, IntPtr hWndFM, Nullable`1 desktopHandle, String fullFilename)
[14:20:21 ERR] Error while navigating to L:\fcvfzepa\4ertoysv\sda1.img\Users\Roninpawn\Documents\Downloads - Shortcut.lnk: System.ArgumentOutOfRangeException: length ('-1') must be a non-negative value. (Parameter 'length')
Actual value was -1.
   at System.ArgumentOutOfRangeException.ThrowNegative[T](T value, String paramName)
   at System.ArgumentOutOfRangeException.ThrowIfNegative[T](T value, String paramName)
   at System.String.ThrowSubstringArgumentOutOfRange(Int32 startIndex, Int32 length)
   at lib7Zip.UI.ListviewNative.GetRowText(Int32 rowIndex, Int32 columnIndex)
   at System.Linq.Enumerable.RangeSelectIterator`2.ToList()
   at System.Linq.Enumerable.RangeSelectIterator`2.ToList()
   at lib7Zip.UI.ListviewNative.GetRows()
   at lib7Zip.SevenZipExtractorUsing7zFM.NavigateToFilename(Int32 pid, IntPtr hWndFM, Nullable`1 desktopHandle, String fullFilename)

Shortcut files (with the .lnk extension mainly, but also a .url file caused the same) put the util into what seemed to be an infinite loop. My console flooded with these errors, continuously, and File Manager never managed to display the directory's contents.

So, progress definitely made. The 'cannot open the file as archive' issue looks resolved to me. But I ran into these soft-linking errors around files that have no 'length.'

To be clear, I WAS able to access other folders that didn't contain .lnk files. But if I were relying on the util to get access to my imaged files, I still would've hit a wall -- blocked from accessing the bulk of my documents by these "-1 length" linking files.

Truly sorry to ruin a perfectly good bug-resolution with another issue. 🤷
...software development, right?

[fiddyschmitt]

Haha yes. Thanks for the awesome write-up! I'll look into those other issues

[fiddyschmitt]

Hi @roninpawn,

I've just released v2.6.3 to address those issues:

1. Fixed "Error while caching File List" which happens when there are millions of files.
2. Added logging "Creating file system tree" so it doesn't look like the console has hung.
3. Fixed shortcuts & url files causing infinite loop.

Keen to know if they work for you. Thanks again for your help :)