NETGEAR is aware of a growing number of phone and online scams. To learn how to stay safe click here.
Forum Discussion
RN104 data-0 inactive since April, still stuck after PSU fix - need assemble advice
Hi all,
I have a ReadyNAS 104 (firmware 6.10.10) with a 4-bay X-RAID2 setup: two Seagate IronWolf 4TB drives and two WD Red 3TB drives. The 3TB baseline across all four disks forms a RAID5 volume called data-0 (8.17 TB), and the extra capacity on the two 4TB drives forms a separate RAID1 volume (931 GB).
What happened, as best I can reconstruct from the logs:
- On April 27, 2026 at 11:01 AM, the system log shows "Disk in channel 4 (Internal) changed state from ONLINE to FAILED." That drive is one of the two WD Reds. SMART data shows a burst of about 878 read errors on that drive's data-0 partition earlier that same morning, right at the tail end of the partition. No reallocated, pending, or uncorrectable sectors before, during, or since, on any drive.
- - Alert emails have apparently never worked on this box ("Alert message failed to send" recurs in the logs going back to 2016), so nobody knew about the April failure. The volume was presumably running degraded on 3 of 4 drives from that point on.
- - On September 13, 2026, the external power adapter started failing, causing all four drives to drop and reconnect erratically, which pushed the NAS into booting its factory-default recovery image.
- - I replaced the power adapter (correct NU90-9120700-I2, 12V/7A). The NAS now boots its normal OS again, all four physical drives are detected correctly, and a separate small mirror volume auto-repaired and finished resyncing on its own.
- - data-0, the 8.17 TB volume with almost all my data, has never been auto-assembled by mdadm across two separate boots since the PSU was fixed, even though the smaller pools repair themselves automatically. The admin page shows "Remove inactive volumes to use the disk. Disk #3,4."
What I've verified so far:
- All four drives pass SMART self-tests clean, zero reallocated, pending, or uncorrectable sectors.
- - I have not clicked "Remove inactive volumes," Format, New Volume, or X-RAID anywhere.
- - SSH itself fails to start on this box right now (error code 15002030001, and "Connection refused" on an actual attempt).
- - I have full log bundles from before and after the PSU replacement if useful.
What I'm hoping for:
I found a very similar solved thread here on this forum (RN104, disk kicked as non-fresh after a power problem, fixed via tech support mode and mdadm --assemble --really-force --readonly). Before I touch anything, I'd appreciate help confirming whether that's the right first step for my logs, and the correct device list for my array specifically. Since SSH won't start, I understand the path in is tech support mode via the boot menu and Telnet, unless there's a better route.
Thanks so much in advance.
Link to that similar solved thread: https://community.netgear.com/discussions/readynas-use/readynas-104---cannot-access-data---remove-inactive-volumes-to-use-the-disk/2454089
4 Replies
OK, just to make sure what's going on is understood:
Is data-0 something you actually created, or something that showed up after the failure? It's highly unusual that you would have a volume called data-0 that was ever usable. Normally, data-0 shows up as the name for a part of your data volume when the NAS cannot properly assemble data since two volumes cannot have the same name. When this happens, both data and data-0 are unusable, but both do contain your files -- just not fully assembled in a usable form. Is data showing up in addition to data-0? If not, did you do anything to data, like delete it?
This is not unusual when a power supply fails, though having one dead drive at the time may have made it more likely and/or affect recovery likelihood. Note that by not unusual, I don't mean it happens every time; but it happens enough that it's one of the reasons real backup on a separate device or the web is necessary to fully protect your data.
Letting StephenB analyze your logs is your best next step, but this information may help him.
BTW, SSH may not be working because it's not enabled. But volume recovery is best done from tech support mode anyway, even if SSH works. In tech support mode, the OS has made no attempt to mount the volume.
Thanks for checking, that's a great question. I can answer it directly from the NAS's own logs: "data" carries its own RAID metadata showing a creation date of April 20, 2016, so it's a genuinely separate volume that has existed for about a decade, not something that appeared this month as a fallback naming workaround. Both "data" (931 GB, healthy after a recent resync) and "data-0" (8.17 TB, still inactive) are showing side by side in the interface right now. I haven't deleted anything called "data" or touched it in any way. Getting the full logs ready to send over to StephenB now, thank you both for the help.
Both "data" (931 GB, healthy after a recent resync) and "data-0" (8.17 TB, still inactive) are showing side by side in the interface right now
Both RAID groups are normally concatenated into a single BTRFS filesystem. It sounds like something failed with the 4x3TB group, so it no longer assembles.
If you are lucky, the group is simply out of sync. You will be able to force the re-assembly, as was done in the thread you reference. But the precise commands needed to need to be tailored to your system (particularly here, when you have multiple RAID groups in the volume).
And of course the cause in your particular case might be different.
If you like, I can analyze the logs for you. Upload them to cloud storage. Then send me the links via private message (PM) using the envelope icon in the upper right of the forum page. Make sure the permissions are set so anyone with the links can download.
NETGEAR Academy
Boost your skills with the Netgear Academy - Get trained, certified and stay ahead with the latest Netgear technology!
Join Us!
