Forum Replies Created
Update 12-27-2025: All data recovered and on another server. Thanks for all the help !
Update 12-26-2025: I was able to copy the partiton table from sda to the rest of the drives and import the array. Checking a few files across the array and there seems to be no issues yet. Started a full copy.
First please don’t say restore from backup as there is no backup. This was there backup target, that then they just started writing files to and not saving them to there computer then letting the computer backup to the nas.
This nas is a family nas that was not setup by me. But as the tec guy in the family it has turned into my proublem. They were running 4 drives in a raid 5. The nas failed and would not boot. I labled the drives and pulled them out and put them into my testing server to see what was going on. I found all 4 drives showed up but 3 out of the 4 appeared to me missing the sda3 partition that I beleave to have all the data. Is there any way I can recover that sda3 partition?
root@tempnas:~# lsblk
NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINTS
sda 8:0 0 5.5T 1 disk
├─sda1 8:1 0 517.7M 1 part
│ └─md127 9:127 0 517.7M 0 raid1
├─sda2 8:2 0 517.7M 1 part
│ └─md123 9:123 0 517.7M 0 raid1
├─sda3 8:3 0 5.4T 1 part
│ └─md124 9:124 0 0B 0 md
├─sda4 8:4 0 517.7M 1 part
│ └─md126 9:126 0 448.1M 0 raid1
└─sda5 8:5 0 8G 1 part
└─md125 9:125 0 0B 0 md
sdb 8:16 0 5.5T 0 disk
└─sdb1 8:17 0 517.7M 0 part
└─md127 9:127 0 517.7M 0 raid1
sdc 8:32 0 5.5T 0 disk
└─sdc1 8:33 0 517.7M 0 part
└─md127 9:127 0 517.7M 0 raid1
sdd 8:48 0 5.5T 0 disk
└─sdd1 8:49 0 517.7M 0 part
└─md127 9:127 0 517.7M 0 raid1
RAID 6 with 5400 RPM hard drives is chosen mainly for reliability and data protection rather than speed. The dual-parity design allows the array to withstand the failure of two drives without data loss, which is important when using large drives that can take a long time to rebuild. Slower 5400 RPM disks run cooler, use less power, and experience less mechanical stress, making them a good fit for systems that stay online all the time. Although performance is lower, especially for writes, this setup works well for backups, file storage, and long-term data retention where stability and data safety matter most.
Back up all important data from the degraded RAID first, since the remaining disk has errors and the volume is now read-only. Avoid running a bad block scan, as it could worsen the disk’s condition. Once the backup is complete, power down the NAS and replace the failing disk in slot 2 with a new, healthy disk. Power the NAS back on and rebuild the RAID 1 array using QTS. Using the original disk from slot 1 is risky due to possible read errors, so a new disk is the safer option. After the rebuild, restore any missing data from the backup.
Hi,I wanted to replace the drives in my RAID 1 array with larger ones in my TS-253 Pro, using the “replace disks one by one” method. I started by removing the disk from slot 1, waited for the beeps, and then inserted the larger replacement drive. The RAID rebuild started automatically using the disk in slot 2, but it failed after a few percent with a read I/O error (“unrecovered read error”). Since then, the RAID has been degraded. QTS said i should run a bad block scan, but I’m afraid to start it as it might harm the disk even more.
Based on recommendations I found on forums, I started backing up the data over SMB using FreeFileSync. During the backup, I received another message:
RAID group “1” is degraded. Volume: DataVol1. The group has been set to read-only. Back up all data immediately.
At the same time, FreeFileSync reported that it could not access the lock file it created:
Cannot write file “\…\sync.ffs_lock” ERROR_WRITE_PROTECT: The media is write protected. [CreateFile]
Because of this, I guess the disk currently in slot 2 may no longer match the disk that was originally in slot 1 (the one I removed).
Once I have completed the backup, how should I proceed?
Should I replace the degraded slot 2 disk with the original disk from slot 1? If I do that, will the RAID rebuild automatically from that disk? The replace disks one by one option is currently disabled. What if that disk also has issues?
Thank you in advance for any tips.
Once a Synology storage pool crashes and DSM can’t repair it, the data can’t be recovered on the NAS itself. The best option is to remove the drive, connect it to another system, and use recovery software to copy the data. The process can be slow and some files may have errors, but it allows the data to be recovered and moved to a new or rebuilt NAS.
I posted a few days ago that what I hoped would be a simple process wasn’t. More has happened since then.
To recap, I upgraded my 2 disk DS218+ from 10TB to 16TB (nominal sizes) with the recommended Synology HDDs. As I also have (had) a 2 disk DS213Air so thought I upgrade that with the 10TB HDDs. I assume that, because the first drive I fitted had a recognizable file structure, the DS218 thought there was a problem and wouldn’t let me erase the contents and add it. I swapped the smaller drive back, and followed the instructions to join it to the drive that had remained installed. All seemed well, but then the drive that hadn’t been removed threw an error and another saying that the volume/pool had crashed. Nothing I did could either fix the errors or copy the data, so I’ve had to resort to data recovery software.
I tried software to access the Linux system, but no files were recognized, and after much Googling, I came across DMDE. The free version lets you check to see if it can do what you want it to (it did) and then buy a licence, which compared to some, was quite reasonable.
I already had a 3.5″ drive caddy, so loaded the drive that hadn’t shown an error, connected it to my laptop and set the software running. It managed to find (then recover) what I believe to be everything, although it showed errors on some, which I copied off. Note that, the software let me re-try the other files (which turned out to be with errors), though needed an OK press for each, so I created an AutoHotkey script for it. 2 days later (full scan and recover), I’ve got copies of what I hope is all of the data, ready to copy onto my newly bought DS223.
Hopefully, if anyone finds themselves in a similar situation, the above will help avoid problems, or fix them.
In conclusion, recovering data from a relabeled or formatted LTO-4 tape can be difficult using standard tools like dd, because they often stop at file markers and cannot interpret the tape’s structure fully. To reliably access the data, it’s best to use specialized tape recovery software that can handle LTO-4 tapes, read data independently of the original backup software, and recover untouched sectors. This approach allows selective restoration of critical files without needing to rebuild the original backup environment.
The IBM TS4300 Tape Library (Machine Type 7DMN) delivers the reliability and automation of enterprise tape systems while remaining cost-effective for open environments. It offers high-density storage, scalable capacity, and straightforward management, making it ideal for long-term data retention and helping reduce expenses related to data center space and power consumption.
This guide covers the TS4300 Tape Library available through Lenovo. It includes important technical details, pre-sales information, key features and specifications, available components and options, and guidance on configuring the system.
The guide is intended for technical experts, sales professionals, system engineers, and IT architects who want to understand the TS4300 Tape Library and evaluate its suitability for IT deployments.
The issue you’re running into is common with LTO tapes, especially when they’ve been relabeled or partially overwritten by software like Archiware P5. LTO tapes don’t fully erase old data when relabeling—only the catalog or tape header is updated—so there is often recoverable data in untouched sectors. Tools like dd can hit a file marker and stop because they don’t interpret the tape format or handle multiple filemarks correctly, which is why your attempt failed.
For this scenario, the most reliable approach is to use a dedicated tape recovery solution. Stellar Data Recovery for Tape is specifically designed for this:
It can read LTO-4 and higher tapes directly.
It bypasses the need for the original backup software or catalog.
It recovers data from untouched sectors even if the tape has been relabeled.
It allows selective recovery, so you can pull only the files you need.
Using Stellar Data Recovery for Tape, you can safely recover the data without relying on dd or rebuilding the original backup environment, which is often the only practical method when the original software isn’t an option.
In short: for a formatted or relabeled LTO-4 tape where dd fails, Stellar Data Recovery for Tape is the most effective solution.
Hello,
I’m looking for advice on software that can read data from a formatted LTO-4 tape and recover the data stored on it.
The tape was originally written using Archware P5. In our experience, relabeling a tape in P5 typically replaces the tape name but does not fully erase the existing data, which means some sectors may still be intact and readable.
I attempted to read the tape using the dd tool, but it stops after a few seconds with a message indicating that a file marker has been reached.
Can anyone suggest a way to work around this issue, or recommend software that can read and recover data from an LTO-4 tape under these circumstances?
Your disaster recovery thinking is solid—the gap is not strategy, but execution. Tape is reliable and economical for backups, but it isn’t built to deliver fast recovery on its own, especially when the original infrastructure and backup catalogs are gone. By pairing your existing LTO setup with a small off-site system and Stellar Data Recovery for Tape, you gain a practical, catalog-independent way to restore critical data within hours rather than days. This approach keeps costs under control, avoids a full environment rebuild, and gives your team timely access to the files they need to stay productive after a worst-case event.
Hello,
We’re a small business with roughly 50 knowledge workers and around 50 TB of data housed in a single building. We currently run a tape-based backup system in our server room using Backup Exec and LTO-6/LTO-7, and it works well for day-to-day protection.
Our concern is disaster recovery. If the server room were destroyed, we wouldn’t be able to restore data from tape until new hardware and a new location were in place. The resulting downtime would be unacceptable for the business.
To reduce recovery time, we’re exploring the idea of maintaining a minimal off-site system with a tape drive installed. The goal would be to restore selected data quickly so staff could continue working on reports and other tasks from home using our existing O365 environment. I asked our vendor for a quote, which led to a conversation with Quantum, and the response I received was essentially that this is “not how it’s done.”
The problem turned out to be a damaged partition table, not a failed RAID or lost data. Three of the four drives still contained the data, but the main data partition was no longer defined in the partition table, so the array could not be assembled. One drive still had the correct layout, and copying that partition table to the other drives restored the missing partitions with the correct size and alignment. Once this was done, mdadm was able to import the RAID-5 array normally, without a rebuild. Spot-checking files showed no issues, and completing a full copy confirmed the data was intact and successfully recovered.