Synology NAS: The peculiarities of data organization and recovery
Synology is one of the most widely known brands in the segment of Network Attached Storages for homes and businesses. The manufacturer offers a vast array of NAS units that range from small desktop boxes to enormous machines that comprise scores of drives. Yet, all of them share the same qualities that are well-liked by users – reliability, simplicity of management and high-end features. Read on to learn more about these devices and their typical characteristics that are likely to matter in the event of a data loss emergency.
General overview of Synology NAS products
The company has two major product lines: DiskStation and RackStation. DiskStation models are primarily designed for households and up to medium-sized offices. Such devices have pedestal or tower-like enclosures and may contain from one to as many as 12 bays. RackStation models prevail in big enterprises. These are rack-mounted industrial solutions that typically have much more capacity and computing power than their counterparts. The drive bays of both device types can hold hard disk drives or solid state drives. The company has also started offering all-Flash rack-mounted arrays with the FlashStation series, highly scalable NAS appliances for extended storage needs with the SA series, and high-availability SAN solutions for iSCSI/FC environments with the UC series.
Some devices within each line can be paired with physical expansion units for additional storage space. The storages themselves don’t provide NAS functions, they have to be connected to the corresponding DiskStation, RackStation or FlashStation appliances, depending on the first letter in their names.
As a rule, the model names of Synology NAS can be interpreted as follows: “DS” stands for DiskStation, “RS” is used for RackStation, and “FS” – for FlashStation. “DX”, “RX”, “FX” refer to expansion units they can be paired with. The first digit or two determine the total number of drives that can be inserted, whether via the internal bays or external expansion units. The last two identify the model or the year of manufacture. In some cases, an additional suffix may be used. “Slim” denotes compact models that employ laptop class 2.5" drives. “J” is applied for low-end budget series geared for personal home use. “Play” can be found in the names of devices that are focused on tasks related to multimedia. The plus sign at the end (“+”) indicates boxes that offer the best performance on a consumer level and have hardware-accelerated video transcoding capabilities. The “XS” NAS boxes are designed for data-intensive business applications. Such models with “RP” come with additional power supplies, whereas ones with a “+” – with built-in 10 gigabit Ethernet (GbE) as well.
All Synology NAS appliances run a proprietary Linux-based operating system called DiskStation Manager or simply DSM. It has an intuitive web interface and covers most vital NAS functions, such as file sharing, data management and protection, backups, cloud services, virtualization and others.
Depending on the model, Ext4 and the newer Btrfs file system of Linux can be used for creating Synology NAS volumes. If both are supported by the device, users can choose either of these file systems, depending on their preferences.
Furthermore, all Synology NAS boxes support the 256-bit AES data encryption. When enabled for a shared folder, the data within it is stored in an encrypted format with a set of encryption keys.
Data organization on Synology NAS
Synology NAS units support multiple RAID options that may vary depending on the model. Some of them may not be configured as RAID and thus lack redundancy. The possible variants include:
One drive without any redundancy can be found in single-bay Synology NAS appliances that often serve as backup solutions for home users or small offices.
RAID 0 may also be used in low-end NAS boxes. It combines two or more drives and offers higher access speeds than individual drives. Yet, the data is not protected against disk failures.
RAID 1 is applied on two-drive NAS units and provides a full backup in case of breakdown of a single drive, since the drives are identical clones of each other.
RAID 5 requires minimum 3 drives, offers faster access to the data and allows the system to keep operating if one of them fails. This is the most popular configuration, especially in the business environment.
RAID 6 needs at least 4 drives, accelerates access to the data, and makes it possible to recover the system even when two drives fail at a time. This is a very reliable, yet, not particularly cost-efficient configuration.
JBOD can be used on any Synology NAS with multiple bays, but this setup is not recommended, as it has neither redundancy nor increased performance. Several drives simply act as a single volume.
RAID 10 requires at least 4 drives, offers faster access and clones the data like in RAID 1, so that a backup would be present when one drive in any or both mirror sets fail.
Synology Hybrid RAID (SHR) is the specific technology of the brand that allows creating RAID from drives of different capacities and add new ones to it anytime. This is basically achieved by establishing several RAID layouts over the disk set. The first RAID is created from the “portions” of the drives whose sizes equal to the capacity of the smallest drive. The remaining “tails” of the drives with larger capacities are organized in another RAID. After that, both sets are spanned together to constitute a single storage unit. The inner mechanism relies on the combination of Linux mdadm and LVM. Yet, in general, when the drives of the same sizes are used in the array, the technology works the same way as the traditional mdadm RAID 5 (or RAID 6).
SHR can protect the array from a single drive failure and SHR 2 – from two concurrent disk failures.
Hint: The peculiarities of data organization on Synology Hybrid RAID are described in the corresponding article.
Common data loss problems of Synology NAS
On the whole, Synology NAS unit are solid and dependable devices. However, they may also develop a variety of issues that impede access to the information stored on the drives or even result in total data loss. Among them are:
The user may make an error when attempting to expand the storage or replace component drives with larger ones. Although Synology devices allow changing from one RAID type to another via the Storage Manager without losing the existing data, replacement of a wrong drive, improper rebuild or other mishaps may result in the corruption of a new RAID setup and make the whole storage pool inaccessible. Accidental deletion of files may also take place, while the Recycle Bin option may appear to be not enabled for the affected shared folder.
The configuration of NAS or RAID metadata can be lost or damaged due to power outages and other unforeseen incidents. This problem also leads to the loss of user files, they can no longer be accessed over the network.
Crash during RAID rebuild
After a failed drive in RAID is replaced with a functioning one, the appliance automatically starts rebuilding the array. But even everything is done correctly, the process may get interrupted by a sudden power loss and not be finished, resulting in total corruption.
Hard drive failure
An abrupt shutdown, a power failure or voltage fluctuations may result in the mechanical failure of a drive. Problems may also occur due to the impact of various natural factors, such as overheating and degradation of the drive’s moving parts, or the breakdown may happen even without any obvious reasons. In some cases, a buzzing or ticking noise may come from the box, and the Assistant will be unable to connect to the NAS. The light bulbs on the device may also flash unusually.
Most users perform firmware upgrades to resolve any issues with the Synology NAS or to benefit from new features of DSM. Yet, after the operation is completed, the storage pool may become unavailable for access and the repair function in the Storage Manager may not help to overcome the problem. Also, unexpected power cuts or a drive failure during the update procedure may make the OS inaccessible. In this case, the NAS itself won’t start, and reinstallation of DSM will format the drives, causing the loss of data.
Failure of the NAS box itself
A NAS appliance is a miniature sever that has many component parts, such a CPU, memory, power supply and others. Though they are less prone to failures than hard drives, if an issue arises with any of them, the box itself may also stop operating, preventing access to the entire content of the storage.
The process of data recovery on Synology NAS: recommendations and precautions
To recover the data lost from your Synology NAS, first of all, it is required to prepare the following:
- A host computer. Its resources are used during the whole data recovery procedure, so it should be powerful enough to make the process faster. Moreover, its connectivity capabilities have to be sufficient for hosting the necessary number of drives. It is advisable to use a PC running Windows. Unlike Linux, this operating system doesn’t support Ext4 and Btrfs, therefore, it cannot silently overwrite any structures on the original drives.
- Enough connectors and power supplies. The computer needs to have a sufficient number of ports to host all the drives at a time. It's recommended to connect the drives directly to the motherboard using SATA cables. Please note that 3.5" drives need to be powered from the internal power supply or from a dedicated external power source. If the motherboard doesn’t have enough slots, multi-port USB-to-SATA adapters or external enclosures can be used. The USB 3.0 and higher variants are preferred, while old adapters may not work properly with drives over 2 TB. Also, the adapter has to be powered appropriately, and the USB cable should be connected to the computer only after the drive is powered. As an alternative, you may use expansion rack boxes. Another possible option is to create bit-to-bit images of the drives, but for this you will need enough space to save files this large.
- Adequate conditions for work with the drives. Traditional hard drives require very careful handling. They need to be kept horizontally on a flat surface. Sufficient cooling should be also ensured. Static electricity during the work with internal components may cause the damage to electronics, that is why it is important to take the necessary precautions: to wear an anti-static wrist strap or touch an unpainted metal surface on the case of a grounded computer during manipulations.
- A target storage. The attempt to copy the restored data back to the NAS drives will lead to permanent data loss. You will have to prepare an additional storage device capable of placing the recovered files – capacious external disks, another network storage, or anything else, depending on the amount of the data to be copied.
To shut the NAS down, you can press and hold the power button until you hear a beep sound, and then disconnect the power cord from the power source. After the device is turned off, the drives can be removed from its enclosure. Some Synology models require a key for access to the drives, whereas with others you will just need to remove the front panel. If an expansion unit was applied in your configuration, its drives have to be extracted as well.
When all the drives are out, their order should be marked, for instance, using paper stickers. This is because they will have to be returned to the box in that very order, if you decide to continue using the same arrangement after data recovery. Moreover, this may aid you in assembling the RAID or SHR configuration of your NAS (when applicable).
After that, all of them, excluding the failed one(s) from RAID or SHR, can be connected to the computer, and the storage layout should be reassembled in data recovery software. UFS Explorer of the RAID, Network RAID and Professional editions serves as an ultimate assistant when it comes to handling both RAID- or SHR-based Synology NAS units. The software is capable of reading out the metadata present on the drives and, as long as this information isn’t damaged, it automatically builds the array in a safe virtual mode. The assembled RAID sets can be found at the end of the list of attached storages, after individual drives, and are labeled with a complex storage icon.
Hint: If you want to learn more about how different drives and volumes are labeled in the interface of UFS Explorer, please refer to the Identification of different storages and technologies.
If the RAID is not present in the list of connected storages in the program, this may mean serious damage to its metadata and the need for manual reconstruction of the array. UFS Explorer provides a convenient instrument for easy definition of RAID parameters and its assembly in a virtual mode.
Hint: The detailed instructions on the assembly of the array can de found in Data recovery on RAID.
An SHR setup usually causes issues for users in this respect. If one or more drives are of larger capacity than the smallest one in this set, SHR creates another RAID with their “tails”. After that, this RAID is spanned with the main one. The major problem here is to assemble the "tail" components correctly (using the accurate offsets) and then span them with the main RAID. However, if all the NAS drives have equal capacities, SHR can be handled like any conventional RAID. In this case, an SHR on two drives corresponds to RAID 1, an SHR on three or more drives – to RAID 5, and SHR 2 – to RAID 6.
For instance, the layout of our two-drive Synology NAS is composed of multiple RAID sets. The ones that keep service information, including DSM, user settings, and system logs, have much smaller sizes of several gigabytes. The data on them is stored using the mirroring technique. The RAID holding the data of our users and installed package configurations has the largest size (1.82 TB) and two partitions. The largest of them, “vg1-volume_1”, contains the actual user files:
The names of system catalogs generally start with the “at” symbol (@). The shared folders created in the Control Panel of DSM typically preserve their original names, like the folder called “Shared” in our example, which stores our documents, music, pictures and videos.
Another instance we have is an SHR-based Synology NAS with 4 drives. The 1st drive is of 500 GB, while the rest three drives are of 1 TB each. Consequently, our SHR configuration includes three RAID sets: The 1st RAID 5 consists of 4 partitions on 4 drives, each partition is of 500 GB in size minus the size of the system and swap partitions; The 2nd RAID 5 consists of 3 partitions of 500 GB each on 3 drives of 1 TB; The 3rd RAID is the SPAN (JBOD) of the 1st and 2nd RAID 5 sets:
After successful reconstruction of RAID, you will be able to choose the volume on it and check whether the needed content is present.
On this stage, you may also face the following difficulties:
Encrypted shared folder. If a shared folder was encrypted upon creation, you won’t be able to see any valid items within it: such a folder has to be decrypted first. As a rule, the names of encrypted folders start and end with the “at” sings (@). The files in them have also unusual names, starting with “ENCRYPTFS”:
The Professional edition of UFS Explorer allows unlocking such folders directly in the software interface, provided that the correct password is available. For this, you can select the file system object located at the top (not the folder itself) and choose the “Transform file system files” option in its context menu:
In the opened dialog, you will have to provide the decryption key you’ve specified when the folder was being encrypted:
After you hit “OK”, the folder will be deciphered and become available for access in the software. The files in it will be displayed with their initial names:
The data was located on an iSCSI target. The iSCSI target service allows creating LUNs and targets on Synology NAS. Such a storage is treated by the computer’s operating system as a conventional locally attached drive. However, in reality, it is a virtual disk located on Synology NAS.
To access the content of an iSCSI target via UFS Explorer, you will need to find the virtual disk file that corresponds to it and mount it in the software. For this, you can go to the “data volume” on the assembled RAID and find a folder called @SCSI. It should contain a subfolder called “LUN”.
The “LUN” folder will have the “FILE” folder within it. The latter will contain a folder with a long unreadable name (it actually corresponds to the hexadecimal name of the logical unit):
You can select the LUN-file located in that folder and open it as a disk image using its context menu:
This option is provided only by the Professional edition of UFS Explorer. Yet, the users who employ UFS Explorer RAID Recovery or UFS Explorer Network RAID may save this file to the computer and then open it in the program via the main menu. The mounted file of the iSCSI target will contain a partition that may have any format, depending on the operating system it was used for. All the user data of this target can be found in this partition:
The data was stored on a virtual disk. Work with virtual drives of a virtual machine implemented on Synology NAS is similar to handling iSCSI targets. However, the virtual disk file to be mounted is located in a folder with an unreadable name that, unlike those of other service folders, doesn’t start with the “at” symbol (@):
If everything is done correctly, but the needed folders and files are missing from the volume, you can scan it for lost data. After the scan, the software presents its result for you to revise and find the required items. The information available in Evaluation and saving the results of data recovery may facilitate your work with the obtained data. When the files and folders are found, you can select and save them to the target storage. If you are going to save the recovered data to another network storage, please check the provided guide.
Last update: September 10, 2021