Bananian

Using SATA Multiplier

15 5221
Hello,

I have a Banana Pro running bananina and i would like to connect a Linn Li Ex-503, but I have the problem with the SATA Multiplier.
Can someone explain how to enable it, or post a .img with this function enabled?
I'm a new in using linux, and I don't really understand how to fix the problem with the pmp.
Thanks in advance

Kind Regards
Panda_92
tkaiser  
Your enclosure features a JMB393 and as far as i know noone reported that this multiplier works with the A20 SoC so far (for details have a look at this thread).

So even if you patch and recompile the kernel as outlined in the aforementioned thread it might not work. And if you're an inexperienced user you will have no clue whether it's the PM's or your fault.

Why don't you use the controller's RAID features if you want to connect large amounts of storage to a Banana Pi (which is an insane idea in my eyes ). In this mode the external enclosure should behave like a single large disk and you don't have to fiddle around with kernel patches/compiling?

Thanks tkaiser,

I was reconsidering this option, but testing different OS for Banana Pro I found out that the OMV (BananaNAS) recognizes all 5 HDD's. And the main use will be a NAS-Server. So i think this thread can be closed. Thanks for your Help ;)

Kind Regards
Panda_92

tkaiser  
Edited by tkaiser at Mon Feb 16, 2015 12:35
Panda_92 replied at Mon Feb 16, 2015 09:36
I found out that the OMV (BananaNAS) recognizes all 5 HDD's


Can you please provide the output of
  1. dmesg | grep ata
Copy the Code
so that we can add the PCI vendor/device id of the Port Multiplier your enclosure contains to the list of known PMs?

Have you had a look at the throughput possible with this PM? I would chdir into the directory in question and then enter
  1. iozone -a -g 2000m -s 2000m -i 0 -i 1 -r4k
Copy the Code
to test with 2GB (for details please have a look at e.g. http://forum.lemaker.org/forum.p ... 2&fromuid=33332). Maybe you would've to install iozone before (apt-get install iozone3)

Edited by Panda_92 at Tue Feb 17, 2015 06:45

Hey tkaiser,

Here you have your info
  1. dmesg | grep ata
Copy the Code
  1. [    0.000000] CPU: PIPT / VIPT nonaliasing data cache, VIPT aliasing instruction cache
  2. [    0.000000] Memory policy: ECC disabled, Data cache writealloc
  3. [    0.000000]       .data : 0xc08a0000 - 0xc09147f8   ( 466 kB)
  4. [    0.387795] libata version 3.00 loaded.
  5. [    1.607125] ata1: SATA max UDMA/133 mmio [mem 0x01c18000-0x01c18fff] port 0x100 irq 88
  6. [    2.166149] ata1: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
  7. [    2.203989] ata1.15: Port Multiplier 1.2, 0x197b:0x0325 r0, 5 ports, feat 0x5/0xf
  8. [    2.233902] ata1.00: hard resetting link
  9. [    2.594685] ata1.00: SATA link up 3.0 Gbps (SStatus 123 SControl 320)
  10. [    2.604100] ata1.01: hard resetting link
  11. [    3.054022] ata1.01: link resume succeeded after 1 retries
  12. [    3.254851] ata1.01: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
  13. [    3.278201] ata1.02: hard resetting link
  14. [    4.361823] ata1.02: link resume succeeded after 4 retries
  15. [    4.490963] ata1.02: SATA link up 1.5 Gbps (SStatus 113 SControl 300)
  16. [    4.583395] ata1.03: hard resetting link
  17. [    4.964613] ata1.03: SATA link up 3.0 Gbps (SStatus 123 SControl 320)
  18. [    4.974010] ata1.04: hard resetting link
  19. [    5.324628] ata1.04: SATA link up 3.0 Gbps (SStatus 123 SControl 320)
  20. [    5.336753] ata1.00: ATA-7: ST2000DL003-9VT166, 0957, max UDMA/133
  21. [    5.347211] ata1.00: 3907029168 sectors, multi 1: LBA48
  22. [    5.356300] ata1.00: configured for UDMA/133
  23. [    5.395022] ata1.01: ATA-8: ST31500343AS, SD35, max UDMA/133
  24. [    5.406432] ata1.01: 2930277168 sectors, multi 0: LBA48 NCQ (depth 31/32)
  25. [    5.458385] ata1.01: configured for UDMA/133
  26. [    5.468498] ata1.02: ATA-8: ST2000DL003-9VT166, CC32, max UDMA/133
  27. [    5.480460] ata1.02: 3907029168 sectors, multi 0: LBA48 NCQ (depth 31/32)
  28. [    5.491291] ata1.02: configured for UDMA/133
  29. [    5.927881] ata1.03: ATA-8: WDC WD20EARS-00MVWB0, 51.0AB51, max UDMA/133
  30. [    5.940333] ata1.03: 3907029168 sectors, multi 0: LBA48 NCQ (depth 31/32)
  31. [    5.960425] ata1.03: configured for UDMA/133
  32. [    6.391276] ata1.04: ATA-8: WDC WD5000AAVS-00ZTB0, 01.01B01, max UDMA/133
  33. [    6.403729] ata1.04: 976773168 sectors, multi 0: LBA48 NCQ (depth 31/32)
  34. [    6.414420] ata1.04: configured for UDMA/133
  35. [    6.420735] ata1: EH complete
  36. [   14.101889] EXT4-fs (mmcblk0p2): mounted filesystem with ordered data mode. Opts: (null)
Copy the Code
I have some problems with iozone not having enough space. I will try to fix that on my 16 GB SD should be enough space.

Edit:
I will make the iozone test on wednesday, due lack of time, I made a normal write and read test:
  1. root@bananas:/media/C212DC4112DC3BDB# dd if=/dev/zero of=tempfile bs=1M count=10                                                                                        24 conv=fdatasync,notrunc
  2. 1024+0 records in
  3. 1024+0 records out
  4. 1073741824 bytes (1.1 GB) copied, 127.119 s, 8.4 MB/s
  5. root@bananas:/media/C212DC4112DC3BDB# echo 3 | sudo tee /proc/sys/vm/drop_caches                                                                                       
  6. 3
  7. root@bananas:/media/C212DC4112DC3BDB# dd if=tempfile of=/dev/null bs=1M count=10                                                                                        24
  8. 1024+0 records in
  9. 1024+0 records out
  10. 1073741824 bytes (1.1 GB) copied, 17.7399 s, 60.5 MB/s
  11. root@bananas:/media/C212DC4112DC3BDB# dd if=tempfile of=/dev/null bs=1M count=10                                                                                        24
  12. 1024+0 records in
  13. 1024+0 records out
  14. 1073741824 bytes (1.1 GB) copied, 17.4411 s, 61.6 MB/s
Copy the Code
Hope it helps

Edit2:

Any suggestion how to increase speed?

Kind Regards
Panda_92



tkaiser  
Edited by tkaiser at Tue Feb 17, 2015 10:26
Panda_92 replied at Mon Feb 16, 2015 14:48
  1. [    2.203989] ata1.15: Port Multiplier 1.2, 0x197b:0x0325 r0, 5 ports, feat 0x5/0xf
Copy the Code


Interesting... "0x197b:0x0325" is the same PCI vendor/device ID the JMB321 uses. So it seems that a JMB393 (also RAID capable) acts like a JMB321 when configured as PM.

Regarding speed: Using dd without the appropriate flags will always measure something (buffers, caches, whatever) but will not tell you how fast disk I/O is. All you would need for the recommended test with iozone is 2 GB free space on any of the HDs. But please keep in mind that using a PM under high load for longer periods of time might lead to malfunctions if the PM overheats. Please read from here on: http://forum.lemaker.org/forum.p ... 7&fromuid=33332

tkaiser  
Edited by tkaiser at Tue Feb 17, 2015 10:32
Panda_92 replied at Mon Feb 16, 2015 14:48
  1. [    4.361823] ata1.02: link resume succeeded after 4 retries
  2. [    4.490963] ata1.02: SATA link up 1.5 Gbps (SStatus 113 SControl 300)
Copy the Code


I would urgently check the CRC error counters with 'smartctl -a /dev/sdc' (maybe you've to 'apt-get install smartmontools' before) since the message above might indicate a cable/connection problem. If SMART attribute 199 increases over time you've a problem.

Edited by Panda_92 at Tue Feb 17, 2015 12:40
tkaiser replied at Tue Feb 17, 2015 10:31
I would urgently check the CRC error counters with 'smartctl -a /dev/sdc' (maybe you've to 'apt-ge ...


Here is the smart code:
  1. root@bananas:~# smartctl -a /dev/sdc
  2. smartctl 5.41 2011-06-09 r3365 [armv7l-linux-3.4.103+] (local build)
  3. Copyright (C) 2002-11 by Bruce Allen, http://smartmontools.sourceforge.net

  4. === START OF INFORMATION SECTION ===
  5. Model Family:     Seagate Barracuda Green (AF)
  6. Device Model:     ST2000DL003-9VT166
  7. Serial Number:    00000000
  8. LU WWN Device Id: 5 000c50 0490eb740
  9. Firmware Version: CC32
  10. User Capacity:    2,000,398,934,016 bytes [2.00 TB]
  11. Sector Size:      512 bytes logical/physical
  12. Device is:        In smartctl database [for details use: -P show]
  13. ATA Version is:   8
  14. ATA Standard is:  ATA-8-ACS revision 4
  15. Local Time is:    Fri Aug  7 04:58:06 2015 UTC
  16. SMART support is: Available - device has SMART capability.
  17. SMART support is: Enabled

  18. === START OF READ SMART DATA SECTION ===
  19. SMART overall-health self-assessment test result: PASSED

  20. General SMART Values:
  21. Offline data collection status:  (0x82) Offline data collection activity
  22.                                         was completed without error.
  23.                                         Auto Offline Data Collection: Enabled.
  24. Self-test execution status:      (   0) The previous self-test routine completed
  25.                                         without error or no self-test has ever
  26.                                         been run.
  27. Total time to complete Offline
  28. data collection:                (  612) seconds.
  29. Offline data collection
  30. capabilities:                    (0x7b) SMART execute Offline immediate.
  31.                                         Auto Offline data collection on/off supp
  32.                                         Suspend Offline collection upon new
  33.                                         command.
  34.                                         Offline surface scan supported.
  35.                                         Self-test supported.
  36.                                         Conveyance Self-test supported.
  37.                                         Selective Self-test supported.
  38. SMART capabilities:            (0x0003) Saves SMART data before entering
  39.                                         power-saving mode.
  40.                                         Supports SMART auto save timer.
  41. Error logging capability:        (0x01) Error logging supported.
  42.                                         General Purpose Logging supported.
  43. Short self-test routine
  44. recommended polling time:        (   1) minutes.
  45. Extended self-test routine
  46. recommended polling time:        ( 255) minutes.
  47. Conveyance self-test routine
  48. recommended polling time:        (   2) minutes.
  49. SCT capabilities:              (0x30b7) SCT Status supported.
  50.                                         SCT Feature Control supported.
  51.                                         SCT Data Table supported.

  52. SMART Attributes Data Structure revision number: 10
  53. Vendor Specific SMART Attributes with Thresholds:
  54. ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  WHEN_
  55.   1 Raw_Read_Error_Rate     0x000f   113   099   006    Pre-fail  Always       -
  56.   3 Spin_Up_Time            0x0003   093   092   000    Pre-fail  Always       -
  57.   4 Start_Stop_Count        0x0032   100   100   020    Old_age   Always       -
  58.   5 Reallocated_Sector_Ct   0x0033   100   100   036    Pre-fail  Always       -
  59.   7 Seek_Error_Rate         0x000f   100   253   030    Pre-fail  Always       -
  60.   9 Power_On_Hours          0x0032   096   096   000    Old_age   Always       -
  61. 10 Spin_Retry_Count        0x0013   100   100   097    Pre-fail  Always       -
  62. 12 Power_Cycle_Count       0x0032   100   100   020    Old_age   Always       -
  63. 183 Runtime_Bad_Block       0x0032   100   100   000    Old_age   Always       -
  64. 184 End-to-End_Error        0x0032   100   100   099    Old_age   Always       -
  65. 187 Reported_Uncorrect      0x0032   100   100   000    Old_age   Always       -
  66. 188 Command_Timeout         0x0032   100   099   000    Old_age   Always       -
  67. 189 High_Fly_Writes         0x003a   100   100   000    Old_age   Always       -
  68. 190 Airflow_Temperature_Cel 0x0022   073   063   045    Old_age   Always       -
  69. 191 G-Sense_Error_Rate      0x0032   100   100   000    Old_age   Always       -
  70. 192 Power-Off_Retract_Count 0x0032   100   100   000    Old_age   Always       -
  71. 193 Load_Cycle_Count        0x0032   100   100   000    Old_age   Always       -
  72. 194 Temperature_Celsius     0x0022   027   040   000    Old_age   Always       -
  73. 195 Hardware_ECC_Recovered  0x001a   017   003   000    Old_age   Always       -
  74. 197 Current_Pending_Sector  0x0012   100   100   000    Old_age   Always       -
  75. 198 Offline_Uncorrectable   0x0010   100   100   000    Old_age   Offline      -
  76. 199 UDMA_CRC_Error_Count    0x003e   200   200   000    Old_age   Always       -
  77. 240 Head_Flying_Hours       0x0000   100   253   000    Old_age   Offline      -
  78. 241 Total_LBAs_Written      0x0000   100   253   000    Old_age   Offline      -
  79. 242 Total_LBAs_Read         0x0000   100   253   000    Old_age   Offline      -

  80. SMART Error Log Version: 1
  81. No Errors Logged

  82. SMART Self-test log structure revision number 1
  83. No self-tests have been logged.  [To run self-tests, use: smartctl -t]


  84. SMART Selective self-test log data structure revision number 1
  85. SPAN  MIN_LBA  MAX_LBA  CURRENT_TEST_STATUS
  86.     1        0        0  Not_testing
  87.     2        0        0  Not_testing
  88.     3        0        0  Not_testing
  89.     4        0        0  Not_testing
  90.     5        0        0  Not_testing
  91. Selective self-test flags (0x0):
  92.   After scanning selected spans, do NOT read-scan remainder of disk.
  93. If Selective self-test is pending on power-up, resume after 0 minute delay.
Copy the Code
I'm trying to run a iozone test, hope it works this time. Regarding overheat, it's ventilaed by a fan and i'm not that hardcore Hdd writer/reader.


Kind Regards
Panda_92

tkaiser  
Edited by tkaiser at Tue Feb 17, 2015 12:36
Panda_92 replied at Tue Feb 17, 2015 12:21
Here is the smart code


Everything's cut off after the 80th column so the only interesting information is missing

This should look like:
  1. ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  WHEN_FAILED RAW_VALUE
  2. ...
  3. 199 UDMA_CRC_Error_Count    0x0032   200   200   000    Old_age   Always       -       2
Copy the Code
In this case the disk reports 2 checksum errors that indicate cable, connector or port multiplier problems. When I did my tests with an overheating PM this counter increased permanently when the PM started to fail.

Edited by Panda_92 at Tue Feb 17, 2015 12:45

I get this error message running iozone:

  1. root@bananas:~#     iozone -a -g 2000m -s 2000m -i 0 -i 1 -r4k
  2.         Iozone: Performance Test of File I/O
  3.                 Version $Revision: 3.397 $
  4.                 Compiled for 32 bit mode.
  5.                 Build: linux

  6.         Contributors:William Norcott, Don Capps, Isom Crawford, Kirby Collins
  7.                      Al Slater, Scott Rhine, Mike Wisner, Ken Goss
  8.                      Steve Landherr, Brad Smith, Mark Kelly, Dr. Alain CYR,
  9.                      Randy Dunlap, Mark Montague, Dan Million, Gavin Brebner,
  10.                      Jean-Marc Zucconi, Jeff Blomberg, Benny Halevy, Dave Boone,
  11.                      Erik Habbinga, Kris Strecker, Walter Wong, Joshua Root,
  12.                      Fabrice Bacchella, Zhenghua Xue, Qin Li, Darren Sawyer.
  13.                      Ben England.

  14.         Run began: Fri Aug  7 05:04:41 2015

  15.         Auto Mode
  16.         Using maximum file size of 2048000 kilobytes.
  17.         File size set to 2048000 KB
  18.         Record Size 4 KB
  19.         Command line used: iozone -a -g 2000m -s 2000m -i 0 -i 1 -r4k
  20.         Output is in Kbytes/sec
  21.         Time Resolution = 0.000001 seconds.
  22.         Processor cache size set to 1024 Kbytes.
  23.         Processor cache line size set to 32 bytes.
  24.         File stride size set to 17 * record size.
  25.                                                             random  random    bkwd   record   stride
  26.               KB  reclen   write rewrite    read    reread    read   write    read  rewrite     read   fwrite frewrite   fread  freread
  27.          2048000       4
  28. Error writing block 434594, fd= 3
  29. write: No space left on device

  30. iozone: interrupted

  31. exiting iozone
Copy the Code
Do I need to expand the SD Card?

Edit: On the HDD is enough space, Hdd with less free space has 40 GB free.

Btw I think the smart test is fine:
  1. 199 UDMA_CRC_Error_Count    0x003e   200   200   000    Old_age   Always  -   0
Copy the Code

You have to log in before you can reply Login | Sign Up

Points Rules