Recurring database corruption on Roon Nucleus One during library build (ref#A6SMOW)

Support
1

Hi! What’s not quite right with Roon?

· Music won’t play or issues with my library

Music won’t play or issues with my library

· Trouble creating or restoring backups

Tell us what's going on

· Hi Roon Support,
I have a recurring database corruption problem with my Roon Nucleus One and would appreciate advice on whether the Nucleus hardware, particularly the system M.2 SSD or RAM, needs investigation/replacement.
System
Roon Nucleus One
Internal music storage: 4 TB Samsung SSD 870
Library: approximately 2,500 albums / 44,000 tracks
Roon database/system storage has plenty of free space (Web Administration previously showed approximately 99% of 86 GB available)
Roon Server: 2.73 (build 1694), production
RoonOS: 2.1 (build 271)
History of the problem
I have repeatedly received the Roon message:
“There was an issue loading your library. To avoid data loss, restore a backup now.”
Initially, restoring a database backup would get Roon running again, but after a few days the same problem would return. On some earlier occasions a complete power cycle of the Nucleus also appeared to help temporarily.
Most recently I restored a backup dated 18 September 2026, but the database corruption subsequently returned.
On 30 September 2026, restarting Roon Server did not resolve the problem. A complete Nucleus power cycle also did not resolve it.
The logs from the failed database contained database corruption errors including:
Error: [broker/database] corruption detected
Critical: [loadstatus] database is corrupt
IsDatabaseCorrupt False => True
The corruption was detected while Roon was carrying out a large metadata refresh.
Clean-database test
To determine whether the corruption was already contained in my backups, I then:
Stopped Roon Server.
Renamed the existing RoonServer directory so that it was preserved.
Restarted Roon Server, allowing Roon to create a completely new database.
Connected to the newly created Roon Server.
Did not restore any database backup.
Allowed Roon to start building a new library from the music files on the internal 4 TB SSD.
Unfortunately, the brand-new database also became corrupt during this initial library build.
The new logs show at approximately 12:56 on 30 September:
LevelDb.Exception: Corruption: corrupted compressed block contents
followed by:
[broker] detected corrupt database
IsDatabaseCorrupt False => True
Critical: [loadstatus] database is corrupt
This occurred in a newly created Roon database into which no previous backup had been restored.
I also noticed that during the import Roon identified at least one apparently corrupt WAV file (Arean, Jenny/Portret (Disc 2)/03 - Some People.WAV, plus another WAV with invalid stream parameters). I mention this for completeness, although I would not expect an unreadable audio file to cause physical LevelDB block corruption.
Logs
I have retained the logs from both:
the original recurring database corruption; and
the subsequent corruption of the completely fresh database.
I can attach both ZIP archives to this support request.
Given that a completely new database, with no restored backup, developed LevelDb.Exception: Corruption: corrupted compressed block contents during its initial library build, could you please advise whether this points to a problem with the Nucleus One system M.2 SSD, RAM, or another hardware component?
I have stopped further restore/reset attempts so that I don't destroy potentially useful diagnostic information.
Please let me know if you would like the complete logs or any additional diagnostic information from the Nucleus.
Thank you.

Tell us about your home network

· My Roon Nucleus One is connected with Cat5 cable to a Cisco switch which is hard wired to the internet modem.

2

Hi @Thiem_Schoonderbeek,

We’re terribly sorry to hear you’re experiencing this with the Nucleus One and we’re eager to investigate to pin down a cause.

For context, Roon checksums the entire database during any indexing to scan for unexpected changes. Corruption detection is granular and pretty consistent, including on-the-fly while you’re browsing or playing music. In most cases, users with corruption will restore an early Backup to try to precede the corrupting event. When the database again encounters corruption, it usually means that the actual corrupting event preceded Roon’s first logged warning of corruption.

If a completely fresh database corrupts, as in your case, that generally points to a cause external to the software, either with the unit itself or in the available power source. Recurring outages or voltage fluctuations can be a cause. Or, as you’ve mentioned, a failing SSD or internal component can be a cause.

All this is to say, you’re right to anticipate that the drive might need replacing. It’s an unfortunate situation.

We’ll closely inspect logs to confirm the cause and a next step. If you’re able to provide the entire zipped Logs folder from within Roon Server, that will include the RoonOS-level logging we’ll specifically need. Here’s a link you can use for uploading.

We’ll watch for your reply and proceed from there. Thank you.

3

Hey @Thiem_Schoonderbeek,

Dropping in to see if there are any updates on this. Were you able to send over the zipped Logs folder from Roon Server so we can look at the RoonOS-level details, and if needed check the Backup guide before any next step? If you have anything new from the Nucleus One side, please reply here and we’ll pick it up from there.

Please note, if we don’t hear back from you this thread may close automatically soon. If the thread auto-closes and you need further assistance, please submit a reopen support request via the technical support help form below and specify that the issue should be reopened. Thank you.

4

Hi Support Team,

I have uploaded the log files twice and I will try again via the link that you have provided. Can you please have a look what went wrong?

Thank you