April 1, 201412 yr Hi, I have my first red balled hard drive, I wanted to check to see if I have the right thoughts about it. The red ball came during my monthly parity check. The smart report is attached. The drive reported as "passed", but unRaid disabled the drive becuase of read errors. That's being reflected under "Raw_Read_Error_Rate" which is in a pre-fail rate? Is the next step to replace the drive? I'm running unRaid version 5.0.5 and the syslog has been attached. Smart_Report.txt syslog.txt
April 1, 201412 yr The drive reported as "passed", but unRaid disabled the drive becuase of read errors. UnRAID doesn't disable a drive due to read errors -- it corrects those by reconstructing the data from the other drives. It disables a drive whenever there is a WRITE error.
April 1, 201412 yr Author Ah okay got it. So would that imply there isn't enough power or a bad SATA cable? Would I still want to RMA the drive because of the Raw_Read_Error_Rate & Reallocated_Sector_Ct? ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE 1 Raw_Read_Error_Rate 0x002f 199 199 051 Pre-fail Always - 11333 5 Reallocated_Sector_Ct 0x0033 197 197 140 Pre-fail Always - 113
April 1, 201412 yr The SMART report indicates read issues. But if the Reallocated_Sector_Ct remains stable the disk should be fine. Run pre-clear on the disk and repeat the SMART test. Then post a new SMART report. The TYPE of those ATTRIBUTES is Pre-fail. The TYPE is a static value and is not indicative of status. Both attributes have normalized VALUEs well above the failure THRESHold. The Raw_Read_Error_Rate RAW_VALUE has meaning only to the manufacturer. The Reallocated_Sector_Ct is an actual count of reallocated sectors. The disk experienced errors during hour 70 that appear to be the cause of the reallocated sectors. The most recent SMART test failed with a read error.
April 1, 201412 yr Ah okay got it. So would that imply there isn't enough power or a bad SATA cable? Would I still want to RMA the drive because of the Raw_Read_Error_Rate & Reallocated_Sector_Ct? ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE 1 Raw_Read_Error_Rate 0x002f 199 199 051 Pre-fail Always - 11333 5 Reallocated_Sector_Ct 0x0033 197 197 140 Pre-fail Always - 113 I am not so worried with the raw read error rate, but am worried about the reallocated sectors. If you want you could run a parity check and see if it holds steady. But experience here has been that these tend to get worse - like a pothole that gets bigger every time a car goes by. If the number keeps getting bigger I would not trust it. But if it stays at 113 I think its fine.
April 1, 201412 yr (a) The drive is almost certainly fine. None of the SMART parameters are anywhere close to the failure threshold. (b) Notwithstanding that, I'd RMA the drive if it's new and can be replaced easily. I simply don't keep drives that start out with that many reallocated sectors.
April 2, 201412 yr Author Okay so now I'm very confused about what to do. I tried to preclear my disabled drive (disk 1), the server crashed during the post-read step at 5%. I went ahead and forced the server to turn off and changed out the SATA cable for the "bad" drive and also changed the port on the motherboard. After restarting I tried running smartctl on the "bad" drive and am getting the following message when I try running smartctl -a -A /dev/sdf (/dev/sdf is the new drive assignment). smartctl 6.2 2013-07-26 r3841 [i686-linux-3.9.11p-unRAID] (local build) Copyright (C) 2002-13, Bruce Allen, Christian Franke, www.smartmontools.org === START OF INFORMATION SECTION === Vendor: /8:0:0:0 Product: scsiModePageOffset: response length too short, resp_len=47 offset=50 bd_len=46 >> Terminate command early due to bad response to IEC mode page A mandatory SMART command failed: exiting. To continue, add one or more '-T permissive' options. Running smartctl -t short /dev/sdf or smartctl -T permissive /dev/sdf does not make a difference on the output of the smartctl -a -A /dev/sdf command. I also tried running preclear again using preclear_disk.sh -l and following gets returned: ====================================1.13 Disks not assigned to the unRAID array (potential candidates for clearing) ======================================== No un-assigned disks detected When in the looking at the array I can see /dev/sdf is not part of the array. What would be my next step? If it makes a difference my array has been started in maintenance mode. Thanks for your help (and the previous responses). syslog-2014-04-02.zip
Archived
This topic is now archived and is closed to further replies.