March 6, 20179 yr Hi Guys, Had a drive fail on me tonight. Tried running the SMART test but it would not run. I've ordered a replacement but it won't be here until Tuesday. My question is should I power off the server until the replacement drive gets here to avoid the unlikely possibility that I have another drive fail? This is the first drive I've had fail on me since setting up this server so I'm not sure what best practice is for this situation. I definitely want to avoid any loss of data. I read through the wiki and couldn't find anything about leaving the server running with a bad drive. Thanks in advance for the help. Added the diag and smart test files just in case. tower-diagnostics-20170305-1658.zip tower-smart-20170305-1749.zip Edited March 6, 20179 yr by nickro8303
March 6, 20179 yr Author Was able to run a short SMART test after rebooting the server but it doesn't seem to have much info. Attached it anyway just in case. tower-smart-20170305-1825.zip
March 9, 20179 yr Your SMART report says your device has SMART capability, but SMART support is disabled; probably because the drive failed? Not sure what to make of that. Hopefully you've received your replacement disk and are well on your way to rebuilding it. I don't think it matters much what you decided to do. I tend to believe electronic equipment has a greater probability to fail during power cycles, but you can't really avoid it unless your hardware supports hot plugging of sata ports. But even then, you run the risk of bumping another disk while trying to hot install the new one. Oh, the irony if that happened... I would just shut down the server, replace the disk, and let it rebuild.
March 9, 20179 yr Community Expert 3 hours ago, dboonthego said: Your SMART report says your device has SMART capability, but SMART support is disabled; probably because the drive failed? Not sure what to make of that. Some drives come with it disabled, and you have to turn it on. Search google for smartctl enable.
Archived
This topic is now archived and is closed to further replies.