I wanted to migrate to Raspberry Pi 5 and take all my data from the old Pi 4. The restore failed and I was able to isolate the issue: When trying to restore my MariaDB and/or InfluxDB add-ons the supervisor reports:
2025-05-13 07:19:15.295 ERROR (MainThread) [supervisor.jobs] Unhandled exception: Error -1 Invalid deflate block found
Traceback (most recent call last):
File “/usr/src/supervisor/supervisor/jobs/decorator.py”, line 304, in wrapper
return await self._method(obj, *args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File “/usr/src/supervisor/supervisor/addons/addon.py”, line 1404, in restore
tmp, data = await self.sys_run_in_executor(_extract_tarfile)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File “/usr/local/lib/python3.13/concurrent/futures/thread.py”, line 59, in run
result = self.fn(*self.args, **self.kwargs)
File “/usr/src/supervisor/supervisor/addons/addon.py”, line 1390, in _extract_tarfile
backup.extractall(
~~~~~~~~~~~~~~~~~^
path=tmp.name,
^^^^^^^^^^^^^^
members=secure_path(backup),
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
filter=“fully_trusted”,
^^^^^^^^^^^^^^^^^^^^^^^
)
^
File “/usr/local/lib/python3.13/tarfile.py”, line 2334, in extractall
self._extract_one(tarinfo, path, set_attrs=not tarinfo.isdir(),
~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
numeric_owner=numeric_owner)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File “/usr/local/lib/python3.13/tarfile.py”, line 2397, in _extract_one
self._extract_member(tarinfo, os.path.join(path, tarinfo.name),
~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
set_attrs=set_attrs,
^^^^^^^^^^^^^^^^^^^^
numeric_owner=numeric_owner)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File “/usr/local/lib/python3.13/tarfile.py”, line 2480, in _extract_member
self.makefile(tarinfo, targetpath)
~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^
File “/usr/local/lib/python3.13/tarfile.py”, line 2534, in makefile
copyfileobj(source, target, tarinfo.size, ReadError, bufsize)
~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File “/usr/local/lib/python3.13/tarfile.py”, line 250, in copyfileobj
buf = src.read(bufsize)
File “/usr/local/lib/python3.13/gzip.py”, line 340, in read
return self._buffer.read(size)
~~~~~~~~~~~~~~~~~^^^^^^
igzip_lib.IsalError: Error -1 Invalid deflate block found
2025-05-13 07:19:15.309 WARNING (MainThread) [supervisor.backups.backup] Can’t restore Add-on core_mariadb:
I tried to extract the contents of the two add-ons from the backup archive but I only received an “unknown error”. So indeed the archive is corrupted. But why?
What I tried to identify the root cause:
- Repeat backup → still corrupted → no random event
- Store on SD card and download via browser as well as on my NAS and downloaded via samba → still corrupted → most probably not caused by conversion while transmitting data via HTTP
I want to highlight the following for everyone who suffers from this issue too:
ha backup new --addons core_mariadb --uncompressed=true
→ Works!! This produces an uncompressed backup of only MariaDB. It can be restored without any problem.
From my point of view, there is something wrong with the compression in HA. Now I have questions:
- Did I do something wrong?
- The regular backups of MariaDB and InfluxDB are (at least for me) useless because I will not be able to restore them. Is there a way to schedule uncompressed backups? (I could do it using automations and fork the backup process so that it will not be subject to timeout - but this is not what I mean here
) - Should I create a bug report? (I guess so, because this behavior destroys trust in HA. However, maybe someone knows what the odds are that it will get implemented at all…)