A newly unboxed G5500 storage device still can’t be used just by plugging in drives—the same logic as a server’s RAID card: bare disks must first be assembled into an array in the device’s management UI before the system recognizes anything. This post records the process and key points of configuring RAID on the G5500 management UI. The original article consisted mainly of 21 operation screenshots that didn’t migrate with the text; this version keeps the confirmed device facts and organizes the things you must think through before configuring RAID, with UI entry points depending on your actual firmware version. This is a graphical-interface operation record; nothing was run from the command line.

What the Management UI Solves

A storage device works on the same principle as a server’s RAID card: “managing disks” is done at the device firmware layer. Creating arrays, deleting arrays, and rebuilding after a disk swap all happen in the management UI; the upper-level system sees only a ready-made logical disk and never needs to care how many physical disks sit underneath or at what level. The G5500 provides a web management UI—reach the device’s management IP from a browser and you’re set, no monitor or keyboard attached.

Logging In to the Management UI

Open the device management IP in a browser and log in with the default account and password.

Key fact: the management UI’s default password is Admin@9000. Log in the first time with the default credentials.

The first thing to do after logging in is change that password—the default is public knowledge, and leaving it unchanged long-term amounts to hanging the device’s admin rights on the door. The change-password entry usually lives under system settings or user management.

Note: the original text was 21 operation screenshots, and the specific click paths did not migrate with the article. For step-by-step operations, consult the device’s actual management UI or the official configuration guide. What follows are the general decision points to settle before configuring RAID, applicable to the management UI of the vast majority of storage devices.

Three Decisions to Make Before Configuring RAID

Before entering the RAID configuration menu, these three decisions are irreversible once made (rebuilding an array = total data loss):

1. Pick the Level: How to Trade Capacity, Safety, and Performance

RAID Level Usable Capacity Fault Tolerance Typical Use
RAID 0 Sum of all disks None (one disk fails, everything is lost) Cache disks, temporary data
RAID 1 Single-disk capacity (two-disk mirror) Survives 1 disk failure System disks, small-capacity high-reliability scenarios
RAID 5 (N-1) disks of capacity Survives 1 disk failure The balance of capacity and safety; the most common
RAID 6 (N-2) disks of capacity Survives 2 disk failures Large arrays (guards against a second failure during rebuild)
RAID 10 Half the disks’ capacity Each mirror group survives 1 failure IO-sensitive scenarios such as databases

The more disks, the higher the probability of a second failure during a RAID 5 rebuild—past 4 disks, go straight to RAID 6 or RAID 10.

Run the numbers on six 4TB disks and the differences between three mainstream combinations are plain:

Combination (6×4TB) Usable Capacity Fault Tolerance When to Choose
RAID 5 About 20 TB Survives 1 failure Few disks, tight budget, rebuild risk acceptable
RAID 6 About 16 TB Survives 2 failures The default when there are many disks or large per-disk capacity
RAID 10 12 TB 1 failure per mirror group Write-intensive, sensitive to rebuild windows

The capacity cost sits right in the table: RAID 5/6 each surrender 1-2 disks’ worth, RAID 10 cuts straight in half. The other hidden cost is rebuild time—the bigger the disks, the more rebuilds are counted in days, and during the migration of those terabytes the array runs in a non-redundant or semi-redundant state; one more failure in that window is a data incident. That’s the logic behind “prefer RAID 6 for large-capacity arrays.”

2. Pick the Member Disks: Every Status Must Be Healthy

Before checking physical disks into the array, confirm each disk shows a normal/online status (Normal/Online) in the UI. Exclude or rebuild any disk in an abnormal state (Failed/Unconfigured Bad) first—build an array with problem disks, and the rebuild will kick off immediately or fail outright.

3. Plan the Hot Spare

If you have a spare bay, designate a hot spare disk. There are two kinds: a global hot spare watches all arrays in the device and covers whichever group has trouble; a dedicated hot spare is bound to one specific array. When a disk in an array fails, the hot spare steps in automatically and the rebuild starts without waiting for a human to reach the machine room—storage devices usually run unattended, so a hot spare’s value is even higher than on a server. The hot spare carries no business data in normal operation; its capacity just needs to be no smaller than the largest member disk in the array.

Configuration and Verification

After completing “pick disks → pick level → confirm creation” in the RAID management menu, verify three things:

  1. Array status: the new array should show Optimal/Normal; Degraded means a member disk isn’t working properly—inspect the disks immediately.
  2. Capacity check: the logical disk’s capacity should match the formula for the chosen level (e.g. 3 disks of 4TB in RAID 5 should yield about 8TB).
  3. Visibility upstream: the attached system should see a logical disk of the corresponding capacity (lsblk on Linux, Disk Management on Windows).

Note: the verification methods above are general practice, not run on this device; the exact status wording in the G5500 UI depends on the actual firmware.

Things to Watch Out For

  • Change the default password Admin@9000 as soon as you first log in to the management UI; a long-unchanged default credential is a common security hazard.
  • Before configuring RAID, confirm the disks in the enclosure are all healthy, and back up before touching anything involving existing data—deleting an array and rebuilding one are both total-data-loss operations.
  • RAID is not backup: it protects “the service keeps running when a disk dies,” not against accidental deletion, ransomware, or RAID controller failure. Important data still needs independent backups.
  • Put access controls on the network the management IP lives in, and never expose the web management UI directly to the public internet.

From bare disks to a logical disk is a step no storage device lets you skip. Operating the G5500 itself isn’t complicated; what actually goes wrong is the up-front decisions—a wrong level means a rebuild, and a bad-status disk slipping into the array means redoing the work. Settle those three things before clicking confirm, and the rest is just walking through the UI’s flow.

This article was rewritten from the author’s CSDN blog posts originally published between 2020 and 2024.