In a recent incident, an Exos X18 SATA drive was reported to fail consistently in a ZFS array setup. The server setup where this issue occurred involves a configuration of six drives, and the Exos X18 is one of them that keeps failing. These failures manifest as a large number of read and write errors, especially evident during the scrubbing and resilvering process. Unlike performance in a desktop testing environment, the server setup presents continual reliability issues, leading to the drive being ejected from the array. During testing using Seatools, the affected drive successfully passed the 'long self-test' that lasts nearly 24 hours when placed in a different desktop computer. This disparity highlights a critical concern regarding drive functionality within specific environments and raises questions on whether the testing software adequately indicates the drive's operational reliability. The user has taken several measures to troubleshoot the issue, such as changing SATA/SAS ports, swapping SATA cables, and even replacing power leads with ones from an operational drive, but the same drive consistently encounters failures. The user's attempts to seek an RMA claim have been complicated due to the insufficient evidence of failure since the Seatools test does not reflect any issues under testing conditions. The situation is exacerbated by the need to avoid invalidating the warranty while still attempting to create documentation that proves the drive's unreliability in the server environment. Overall, the concerned user is currently navigating the complexities of hardware diagnostics and warranty claims while dealing with a critical failure in their ZFS array's setup, impacting data reliability and server performance.