Hi, David Christensen wrote: > https://en.wikipedia.org/wiki/Click_of_death
But that's a different technology (and 20 years ago). > If you cannot return the drive, I would download, install, and run > "Data Lifeguard Diagnostic for Windows": > https://support.wdc.com/downloads.aspx?DL#downloads I live in a biotope where MS-Windows is not available for tests. (Those whom i could ask have none. Those who have, i would not want to instigate to even press their Enter key.) > Does the drive make the same noise when the computer is running other > operating systems, such as Windows? It does it without any OS while showing me its mainboard firmware status page with processor temperature (31 C at 25 C room temperature), fan speeds, and the list of storage devices. > If the drive is within the warranty period and passes the diagnostic, > contact WD support and describe the symptoms. They might let you RMA it. The disk was bought as new just a week ago. I already urged my hardware provider to talk to his hardware provider. The main question is: bug or feature. (The regulars might remember that i have DVD drives with auto-pull-in feature after 200 seconds for which nobody wants to be responsible. So i am open to any insight here.) Dan Ritter wrote: > Drives should not make unexpected noises. It has much in common with the well known and hated periodic disk accesses of software watching the disk's content. But i excluded all possible watching software by umount, swapoff, shutdown. The knocking only stays away if i pull the disk's power plug (SATA). Reco wrote: > Have you tried to disable drive heads parking via hdparm? hdparm -J ? The man page says "The factory default is eight (8) seconds". That would be about twice as long as what i experience. # hdparm -J /dev/sda /dev/sda: SG_IO: bad/missing sense data, sb[]: 70 00 05 00 00 00 00 0a 04 53 00 00 21 04 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 SG_IO: bad/missing sense data, sb[]: 70 00 05 00 00 00 00 0a 04 53 00 00 21 04 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 SG_IO: bad/missing sense data, sb[]: 70 00 05 00 00 00 00 0a 04 53 00 00 21 04 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 SG_IO: bad/missing sense data, sb[]: 70 00 05 00 00 00 00 0a 04 53 00 00 21 04 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 wdidle3 = disabled The sense data bear KEY=0x5, ASC= 0x21, ASCQ=0x04. Key 5 means: "Illegal request". From MMC-5 i read for ASC=0x21 only ASCQ 0 to 3: 5 21 00 LOGICAL BLOCK ADDRESS OUT OF RANGE 5 21 01 INVALID ELEMENT ADDRESS 5 21 02 INVALID ADDRESS FOR WRITE 5 21 03 INVALID WRITE CROSSING LAYER JUMP In SPC-3 the ASC=0x21 list ends already at ASCQ=2. SBC-2 lists no own error codes. Without knowing the failed command, it is quite obscure what happened. The man page of hdparm says about -J: "WD supply a WDIDLE3.EXE DOS utility for tweaking this setting, and you should use that program instead of hdparm if at all pos‐ sible." Well, it's not possible. But "wdidle3 = disabled" does not look like i could get any larger setting for patience. So i refrain from trying to set the proposed value of 30 for now. > What about smartctl long test, does it show anything suspicious? I never used smartctl up to now. Shall i follow these instructions ? https://www.thomas-krenn.com/en/wiki/SMART_tests_with_smartctl#Test_procedure_with_smartctl Have a nice day :) Thomas