John_M Posted January 9, 2016 Share Posted January 9, 2016 Hi. I've got the opportunity to acquire a WD6001FXYZ (6 TB Western Digital Re) in exchange for a WD60EFRX (6 TB Western Digital Red) and I wondered if it would make a good parity disk alongside regular Reds as data disks. It's a 7200 RPM enterprise SATA III disk and it uses the 4Kn advanced format, not 512e. It's brand new, was bought in error by a colleague and it normally costs almost double what I'd have to "pay" for it. So, is it a bargain or should I avoid it? Quote Link to comment
John_M Posted January 11, 2016 Author Share Posted January 11, 2016 No comments? Is nobody interested? I've decided to get the drive and experiment with it. It might prove to be an expensive mistake or it might simply just work. I'm more hopeful of success than I would be using Windows. Either way, it should be educational. I've done a search of the boards but I haven't found anything about 4k native disks and unRAID. Quote Link to comment
HellDiverUK Posted January 11, 2016 Share Posted January 11, 2016 Linux has supported 4K native since kernel version 2.6.31. UnRAID is currently on 4.1.13. Unless someone knows differently, I don't see any reason why it wouldn't work. Quote Link to comment
SSD Posted January 11, 2016 Share Posted January 11, 2016 Linux has supported 4K native since kernel version 2.6.31. UnRAID is currently on 4.1.13. Unless someone knows differently, I don't see any reason why it wouldn't work. I tend to agree. With the GPT style partitions used today, I think it won't be an issue. But let us know of your experiences. Quote Link to comment
John_M Posted January 12, 2016 Author Share Posted January 12, 2016 Well, I traded hard disks with my colleague. I felt a bit bad at the cost difference and offered him some cash in addition to meet him half way, but he's happy that he got what he needs and he realises I'm taking something of a risk, so he refused the money. I put the 6 TB Re disk and three 6 TB WD Reds in a new HP Microserver Gen8 that I'd bought recently to build a small unRAID server for my brother but the thing is so cute I'm keeping it for myself and I'll buy another one for him! It's straight out of the box with Celeron 1610T processor and 4 GB of ECC RAM. These beauties are so inexpensive (about the same price as one of the disks) and so full of possibilities I might yet get a third one to experiment with! Anyway, I know from other threads that they'll run unRAID just fine provided you set the embedded storage controller to AHCI mode. I also enabled the drive write cache option, which is off by default. Here's a snippet from the syslog where the first two disks are detected, showing the 4Kn nature of /dev/sdb and the 512e nature of /dev/sdc for comparison. Note that the latter has exactly eight times as many 512-byte logical sectors as the former has 4096-byte sectors: Jan 12 18:23:14 Northolt kernel: ata1: SATA link up 6.0 Gbps (SStatus 133 SControl 300) Jan 12 18:23:14 Northolt kernel: ata1.00: ATA-10: WDC WD6001FXYZ-01SUYB1, WD-WX61D65744ES, ER43, max UDMA/133 Jan 12 18:23:14 Northolt kernel: ata1.00: 1465130646 sectors, multi 0: LBA48 NCQ (depth 31/32), AA Jan 12 18:23:14 Northolt kernel: ata1.00: configured for UDMA/133 Jan 12 18:23:14 Northolt kernel: scsi 1:0:0:0: Direct-Access ATA WDC WD6001FXYZ-0 ER43 PQ: 0 ANSI: 5 Jan 12 18:23:14 Northolt kernel: sd 1:0:0:0: Attached scsi generic sg1 type 0 Jan 12 18:23:14 Northolt kernel: sd 1:0:0:0: [sdb] 1465130646 4096-byte logical blocks: (5.86 TB/5.45 TiB) Jan 12 18:23:14 Northolt kernel: sd 1:0:0:0: [sdb] Write Protect is off Jan 12 18:23:14 Northolt kernel: sd 1:0:0:0: [sdb] Mode Sense: 00 3a 00 00 Jan 12 18:23:14 Northolt kernel: sd 1:0:0:0: [sdb] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA Jan 12 18:23:14 Northolt kernel: sd 1:0:0:0: [sdb] 1465130646 4096-byte logical blocks: (5.86 TB/5.45 TiB) Jan 12 18:23:14 Northolt kernel: sd 1:0:0:0: [sdb] 1465130646 4096-byte logical blocks: (5.86 TB/5.45 TiB) Jan 12 18:23:14 Northolt kernel: sd 1:0:0:0: [sdb] Attached SCSI disk Jan 12 18:23:14 Northolt kernel: ata2: SATA link up 6.0 Gbps (SStatus 133 SControl 300) Jan 12 18:23:14 Northolt kernel: ata2.00: ATA-9: WDC WD60EFRX-68MYMN1, WD-WX21D45C7NKU, 82.00A82, max UDMA/133 Jan 12 18:23:14 Northolt kernel: ata2.00: 11721045168 sectors, multi 0: LBA48 NCQ (depth 31/32), AA Jan 12 18:23:14 Northolt kernel: ata2.00: configured for UDMA/133 Jan 12 18:23:14 Northolt kernel: scsi 2:0:0:0: Direct-Access ATA WDC WD60EFRX-68M 0A82 PQ: 0 ANSI: 5 Jan 12 18:23:14 Northolt kernel: sd 2:0:0:0: [sdc] 11721045168 512-byte logical blocks: (6.00 TB/5.45 TiB) Jan 12 18:23:14 Northolt kernel: sd 2:0:0:0: Attached scsi generic sg2 type 0 Jan 12 18:23:14 Northolt kernel: sd 2:0:0:0: [sdc] 4096-byte physical blocks Jan 12 18:23:14 Northolt kernel: sd 2:0:0:0: [sdc] Write Protect is off Jan 12 18:23:14 Northolt kernel: sd 2:0:0:0: [sdc] Mode Sense: 00 3a 00 00 Jan 12 18:23:14 Northolt kernel: sd 2:0:0:0: [sdc] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA Jan 12 18:23:14 Northolt kernel: sd 2:0:0:0: [sdc] Attached SCSI disk After successfully booting, I checked the SMART status of all drives and ran the short self-test. Here's the initial SMART report for the Re disc. Does anyone know what attribute 16 refers to? smartctl 6.2 2013-07-26 r3841 [x86_64-linux-4.1.13-unRAID] (local build) Copyright (C) 2002-13, Bruce Allen, Christian Franke, www.smartmontools.org === START OF INFORMATION SECTION === Device Model: WDC WD6001FXYZ-01SUYB1 Serial Number: WD-WX61D65744ES LU WWN Device Id: 5 0014ee 004180f60 Firmware Version: ER43 User Capacity: 6,001,175,126,016 bytes [6.00 TB] Sector Size: 4096 bytes logical/physical Rotation Rate: 7200 rpm Device is: Not in smartctl database [for details use: -P showall] ATA Version is: ACS-3 (unknown minor revision code: 0x006d) SATA Version is: SATA 3.1, 6.0 Gb/s (current: 6.0 Gb/s) Local Time is: Tue Jan 12 20:41:03 2016 GMT SMART support is: Available - device has SMART capability. SMART support is: Enabled === START OF READ SMART DATA SECTION === SMART overall-health self-assessment test result: PASSED General SMART Values: Offline data collection status: (0x80) Offline data collection activity was never started. Auto Offline Data Collection: Enabled. Self-test execution status: ( 0) The previous self-test routine completed without error or no self-test has ever been run. Total time to complete Offline data collection: ( 0) seconds. Offline data collection capabilities: (0x7b) SMART execute Offline immediate. Auto Offline data collection on/off support. Suspend Offline collection upon new command. Offline surface scan supported. Self-test supported. Conveyance Self-test supported. Selective Self-test supported. SMART capabilities: (0x0003) Saves SMART data before entering power-saving mode. Supports SMART auto save timer. Error logging capability: (0x01) Error logging supported. General Purpose Logging supported. Short self-test routine recommended polling time: ( 2) minutes. Extended self-test routine recommended polling time: ( 575) minutes. Conveyance self-test routine recommended polling time: ( 6) minutes. SCT capabilities: (0x003d) SCT Status supported. SCT Error Recovery Control supported. SCT Feature Control supported. SCT Data Table supported. SMART Attributes Data Structure revision number: 16 Vendor Specific SMART Attributes with Thresholds: ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE 1 Raw_Read_Error_Rate 0x002f 200 100 051 Pre-fail Always - 0 3 Spin_Up_Time 0x0027 100 100 021 Pre-fail Always - 0 4 Start_Stop_Count 0x0032 100 100 000 Old_age Always - 3 5 Reallocated_Sector_Ct 0x0033 200 200 140 Pre-fail Always - 0 7 Seek_Error_Rate 0x002e 100 100 000 Old_age Always - 0 9 Power_On_Hours 0x0032 100 100 000 Old_age Always - 2 10 Spin_Retry_Count 0x0032 100 100 000 Old_age Always - 0 11 Calibration_Retry_Count 0x0032 100 100 000 Old_age Always - 0 12 Power_Cycle_Count 0x0032 100 100 000 Old_age Always - 3 16 Unknown_Attribute 0x0022 000 200 000 Old_age Always - 309738010 183 Runtime_Bad_Block 0x0032 100 100 000 Old_age Always - 0 192 Power-Off_Retract_Count 0x0032 200 100 000 Old_age Always - 1 193 Load_Cycle_Count 0x0032 200 100 000 Old_age Always - 1 194 Temperature_Celsius 0x0022 113 113 000 Old_age Always - 39 196 Reallocated_Event_Count 0x0032 200 100 000 Old_age Always - 0 197 Current_Pending_Sector 0x0032 200 100 000 Old_age Always - 0 198 Offline_Uncorrectable 0x0030 100 100 000 Old_age Offline - 0 199 UDMA_CRC_Error_Count 0x0032 200 200 000 Old_age Always - 0 200 Multi_Zone_Error_Rate 0x0008 100 100 000 Old_age Offline - 0 SMART Error Log Version: 1 No Errors Logged SMART Self-test log structure revision number 1 Num Test_Description Status Remaining LifeTime(hours) LBA_of_first_error # 1 Short offline Completed without error 00% 0 - SMART Selective self-test log data structure revision number 1 SPAN MIN_LBA MAX_LBA CURRENT_TEST_STATUS 1 0 0 Not_testing 2 0 0 Not_testing 3 0 0 Not_testing 4 0 0 Not_testing 5 0 0 Not_testing Selective self-test flags (0x0): After scanning selected spans, do NOT read-scan remainder of disk. If Selective self-test is pending on power-up, resume after 0 minute delay. It all looks very good so far. My next step with new disks is to pre-clear them. I know that opinions differ, but I'm not one who pre-clears a disk to death. I do (1) a SMART short self-test, followed by (2) a single pre-clear pass, followed by (3) a SMART long self-test. If it fails (1) I reject the disk. If it fails (2) I repeat once and reject if it fails on the second pass. If it passes (3) I put the disk into service. All four disks are currently on the pre-read phase of pre-clear. After 3.5 hours the Re is 34% through the pre-read, while the Red next to it is at 28%. So the Re is fast, having the advantage of high density platters similar to those used in the Red plus a faster spin speed. I'm also happy to say that it passed the 12.5% (one-eighth) point without smashing the actuator into its end-stop, so I'm confident that the pre-clear script handles 4Kn disks properly. It's also running a little warmer than the Reds, but not worryingly so. I'll report back when the pre-clears have finished. If all is still good I'll run the SMART long self-test on all four drives (polling time is 575 minutes for the Re compared with 692 for the Red, which is nearly two hours faster) and then I'll run the diskspeed script before assigning the disks to an array. Quote Link to comment
SSD Posted January 13, 2016 Share Posted January 13, 2016 Attribute 16 looks very odd. With a value of 0 and threshold of 0, based on the Smart rules I understand, the Smart report should report the drive as failing now. But it isn't. Also, "worst" should never be higher than "value". (0 is much worse than 200). I could find no reference to attribute 16 in several sources that document known attributes. I think it unlikely that this attribute is anything to be overly concerned about, but I have interacted with the smartctl author on another issue, and may refer him to your smart report so he is aware. Quote Link to comment
John_M Posted January 13, 2016 Author Share Posted January 13, 2016 Thanks Brian. I've noticed that smartctl returns a lot of unknown attributes when used with SSDs but that for HDDs they are generally well defined. It would be useful to fill in this gap. I've asked WD Support for information and I'll report back with anything interesting. Meanwhile, I'm still pre-clearing. It's 72% of the way through the zeroing phase and it's fast: the Reds (no slouches themselves) are a long way behind at 35%, 33% and 33%, respectively. I'm using your version of the script (many thanks for that, btw) to speed up the verifying phase so perhaps by Friday it will have finished. I'll be back. Quote Link to comment
John_M Posted January 14, 2016 Author Share Posted January 14, 2016 When I woke up this morning the Re had completed its pre-clear successfully (but I've received no response yet from WD regarding that unknown attribute 16 and its strange behaviour): ========================================================================1.15b == invoked as: /boot/config/plugins/preclear.disk/preclear_disk.sh -c 1 -f -J /dev/sdb == WDCWD6001FXYZ-01SUYB1 WD-WX61D65744ES == Disk /dev/sdb has been successfully precleared == with a starting sector of 1 == Ran 1 cycle == == Using :Read block size = 1003520 Bytes == Last Cycle's Pre Read Time : 12:11:48 (136 MB/s) == Last Cycle's Zeroing time : 9:35:37 (173 MB/s) == Last Cycle's Post Read Time : 12:24:13 (134 MB/s) == Last Cycle's Total Time : 34:12:37 == == Total Elapsed Time 34:12:37 == == Disk Start Temperature: 28C == == Current Disk Temperature: 37C, == ============================================================================ ** Changed attributes in files: /tmp/smart_start_sdb /tmp/smart_finish_sdb ATTRIBUTE NEW_VAL OLD_VAL FAILURE_THRESHOLD STATUS RAW_VALUE Unknown_Attribute = 0 0 0 near_thresh 4479750604 Temperature_Celsius = 115 124 0 ok 37 No SMART attributes are FAILING_NOW 0 sectors were pending re-allocation before the start of the preclear. 0 sectors were pending re-allocation after pre-read in cycle 1 of 1. 0 sectors were pending re-allocation after zero of disk in cycle 1 of 1. 0 sectors are pending re-allocation at the end of the preclear, the number of sectors pending re-allocation did not change. 0 sectors had been re-allocated before the start of the preclear. 0 sectors are re-allocated at the end of the preclear, the number of sectors re-allocated did not change. ============================================================================ so I started a long SMART self-test and went to work. This evening the self-test completed without error and in the meantime the three Reds completed their pre-clears successfully. Here's the report of one of them for comparison: ========================================================================1.15b == invoked as: /boot/config/plugins/preclear.disk/preclear_disk.sh -c 1 -f -J /dev/sdc == WDCWD60EFRX-68MYMN1 WD-WX21D45C7NKU == Disk /dev/sdc has been successfully precleared == with a starting sector of 1 == Ran 1 cycle == == Using :Read block size = 1000448 Bytes == Last Cycle's Pre Read Time : 14:45:39 (112 MB/s) == Last Cycle's Zeroing time : 12:10:31 (136 MB/s) == Last Cycle's Post Read Time : 14:41:01 (113 MB/s) == Last Cycle's Total Time : 41:38:10 == == Total Elapsed Time 41:38:10 == == Disk Start Temperature: 26C == == Current Disk Temperature: 31C, == ============================================================================ ** Changed attributes in files: /tmp/smart_start_sdc /tmp/smart_finish_sdc ATTRIBUTE NEW_VAL OLD_VAL FAILURE_THRESHOLD STATUS RAW_VALUE Temperature_Celsius = 121 126 0 ok 31 No SMART attributes are FAILING_NOW 0 sectors were pending re-allocation before the start of the preclear. 0 sectors were pending re-allocation after pre-read in cycle 1 of 1. 0 sectors were pending re-allocation after zero of disk in cycle 1 of 1. 0 sectors are pending re-allocation at the end of the preclear, the number of sectors pending re-allocation did not change. 0 sectors had been re-allocated before the start of the preclear. 0 sectors are re-allocated at the end of the preclear, the number of sectors re-allocated did not change. ============================================================================ As predicted, the 7200 RPM disk is both faster and warmer than the Intellipower disk (smartctl reports it as 5700 RPM). One thing I don't understand is the "Using :Read block size =" line. In the case of the 4Kn disk it has the value 1003520, which is a multiple of 4096. In the case of the 512e disk it has the value of 1000448, which isn't a multiple of 4096, but it is a multiple of 1024, and therefore also of 512. Why are they different? I should have thought that it is more efficient to read any Advanced Format disk in 4096-byte aligned blocks. The Reds are now SMART long self-testing. Quote Link to comment
John_M Posted January 14, 2016 Author Share Posted January 14, 2016 Here's the result of the speed test. sdb is the WD6001FXYZ, the others are WD60EFRX. Quote Link to comment
John_M Posted January 15, 2016 Author Share Posted January 15, 2016 Here's WD's reply to my query about the FXYZ's SMART attribute 16: Dear John, this is Janek from Western Digital Customer Support. Sadly there's no information available of the purpose of this particular attribute. However I've escalated your case to our higher technical level to see if can get this info for you. Alternatively, while we wait and if you have the possibility, there are a couple of things that can be tried: - if you can upgrade your smartmontool package to a newer version (especially the drivedb.h file, which gets constantly upgraded with new info about drives) then this attribute might not be unknown to the program anymore. - attach the drive to a Windows PC and check it with our Data LifeGuard Diagnostics tool. The software will consider all relevant attributes for the drive health, so if it doesn't flag the 16th one as problematic, then there's no reason to worry. Sincerely, Janek Western Digital Service and Support :sigh: I suppose it's only to be expected from first level support, these days. Hopefully the next level will be able to provide a more useful answer. The SMART long self-tests are at 90%. If they all pass I'll have a go at allocating disks to create an array next. Quote Link to comment
John_M Posted January 15, 2016 Author Share Posted January 15, 2016 Oh, this is disappointing! The Reds all passed their long self-tests shortly after my last posting, so I allocated the disks as follows: parity - WD6001FXYZ; disks 1 to 3 - WD60EFRX but the array wouldn't start. No error displayed in the GUI, no error in the syslog. I tried again. Nothing. No message about parity not being the largest disk - that isn't the case anyway as the four disks are shown as having identical capacities. However it does have fewer sectors, of course - only one eighth as many, but exactly one eighth as many. I unassigned parity, leaving just the four Reds as data disks, and was able to start the array and format them. I stopped the array, did a New Config, rebooted the server to start with a nice clean syslog and worked methodically, making notes as I went. First, I downloaded the diagnostics (attached). Next, I tried the same allocation as before: parity - sdb, disk1 - sdc, disk2 - sdd, disk3 - sde (where sdb is the FXYZ and the others are EFRX). The array wouldn't start. I unassigned parity and the array started. 18 TB of unprotected storage. I stopped the array and reassigned the FXYZ as parity. The array wouldn't start. I did a New Config and assigned sdb as disk1, and no others. The array wouldn't start. I did a New Config and downloaded the syslog (too big, so zipped and attached). My conclusions: the bells and whistles seem to have no problem with 4Kn Advanced Format disks but the unRAID driver itself silently rejects them. The Linux kernel has no problem, nor the embedded controller in the HP Microserver Gen8, nor the pre-clear script. I don't see any way forward from here without help from LimeTech. I'll make a feature request. Surprisingly, I can find no other reference to 4Kn anywhere, in any thread on this board, except this one. The encouraging thing is that I got as far as I did. I'm sure it would not be difficult to add support (it may already be inherently there, but disabled), and doing so can only help compatibility with 512e Advanced Format disks. northolt-diagnostics-20160115-0857.zip syslog.txt.zip Quote Link to comment
John_M Posted February 15, 2016 Author Share Posted February 15, 2016 I finally received a reply to my query from WD: Dear John, this is Janek from Western Digital Customer Support. My apology for the delay. According to our higher technical level this attribute shows is the raw value of drive's workload indication and doesn't play any role in the health assessment of the unit. Sincerely, Janek Western Digital Service and Support So SMART parameter 16 is a measure of the disk's workload and can be ignored. Quote Link to comment
John_M Posted September 20, 2017 Author Share Posted September 20, 2017 It works! 4Kn support was added in unRAID 6.4.0-rc8q. Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.