headers
Christian Franke | 1 Oct 2011 14:09
Picon
Favicon

Re: MegaRAID SAT layer is reportedly buggy

Alex Samorukov wrote:
>
> But it did not resolve the problem. Do you know any other OUT command 
> to check this?
>

The only other DATA-OUT command that makes sense in the scope of 
smartctl is WRITE LOG EXT. For SCT commands it can be used as an 
alternative to SMART WRITE LOG. I did not implement it because it is a 
48-bit ATA command.

Recent ATA ACS revisions list the following (PIO) DATA-OUT commands:

WRITE MULTIPLE - C5h
WRITE MULTIPLE EXT - 39h
WRITE MULTIPLE FUA EXT
WRITE SECTOR(S) - 30h
WRITE SECTOR(S) EXT - 34h

DCO feature set:
DEVICE CONFIGURATION SET - B1h/C3h

HPA security extensions:
SET MAX SET PASSWORD - F9h/01
SET MAX UNLOCK - F9h/03h

Security feature set:
SECURITY DISABLE PASSWORD - F6h
SECURITY ERASE UNIT - F4h
SECURITY SET PASSWORD - F1h
(Continue reading)

Permalink | Reply | 0 comments |
headers
Artem Bokhan | 3 Oct 2011 13:51
Picon
Favicon

Re: MegaRAID SAT layer is reportedly buggy

On 30.09.2011 05:33, Alex Samorukov wrote:
> "-l xerror -l xselftest -l directory -l scttemp -l scterc -l sataphy"
>> None of above. Actually the -x adds one more than that: -l scttemp,
>> which increases the Other Error Count by 4 each run.
>>
> Please try latest SVN - i submitted patch to resolve this bug (for
> SAT12/SAT16 protocols). It works for me, at least.

1. Does this break scterc set?

Device: /dev/sg0 [megaraid_disk_57] [SAT], set of SCT Error Recovery Control failed

2. I have new error messages with new version, are they connected with your patch?

  Device: /dev/sg0 [megaraid_disk_13] [SAT], offline data collection was 
suspended by an interrupting command from host (auto:on)

------------------------------------------------------------------------------
All the data continuously generated in your IT infrastructure contains a
definitive record of customers, application performance, security
threats, fraudulent activity and more. Splunk takes this data and makes
sense of it. Business sense. IT sense. Common sense.
http://p.sf.net/sfu/splunk-d2dcopy1
Permalink | Reply | 5 comments |
headers
Alex Samorukov | 3 Oct 2011 13:53
Picon

Re: MegaRAID SAT layer is reportedly buggy

On 10/03/2011 01:51 PM, Artem Bokhan wrote:
> On 30.09.2011 05:33, Alex Samorukov wrote:
>> "-l xerror -l xselftest -l directory -l scttemp -l scterc -l sataphy"
>>> None of above. Actually the -x adds one more than that: -l scttemp,
>>> which increases the Other Error Count by 4 each run.
>>>
>> Please try latest SVN - i submitted patch to resolve this bug (for
>> SAT12/SAT16 protocols). It works for me, at least.
>
> 1. Does this break scterc set?
>
> Device: /dev/sg0 [megaraid_disk_57] [SAT], set of SCT Error Recovery 
> Control failed
I don`t think that scterc set is possible with or without this patch. 
Also i don`t think that there is any practical sense to change it in hw 
raid.
>
> 2. I have new error messages with new version, are they connected with 
> your patch?
>
>  Device: /dev/sg0 [megaraid_disk_13] [SAT], offline data collection 
> was suspended by an interrupting command from host (auto:on)
Very unlikely

------------------------------------------------------------------------------
All the data continuously generated in your IT infrastructure contains a
definitive record of customers, application performance, security
threats, fraudulent activity and more. Splunk takes this data and makes
sense of it. Business sense. IT sense. Common sense.
http://p.sf.net/sfu/splunk-d2dcopy1
(Continue reading)

Permalink | Reply | 4 comments |
headers
Artem Bokhan | 3 Oct 2011 14:15
Picon
Favicon

Re: MegaRAID SAT layer is reportedly buggy

On 03.10.2011 18:53, Alex Samorukov wrote:
> On 10/03/2011 01:51 PM, Artem Bokhan wrote:
>> On 30.09.2011 05:33, Alex Samorukov wrote:
>>> "-l xerror -l xselftest -l directory -l scttemp -l scterc -l sataphy"
>>>> None of above. Actually the -x adds one more than that: -l scttemp,
>>>> which increases the Other Error Count by 4 each run.
>>>>
>>> Please try latest SVN - i submitted patch to resolve this bug (for
>>> SAT12/SAT16 protocols). It works for me, at least.
>>
>> 1. Does this break scterc set?
>>
>> Device: /dev/sg0 [megaraid_disk_57] [SAT], set of SCT Error Recovery Control 
>> failed
> I don`t think that scterc set is possible with or without this patch. Also i 
> don`t think that there is any practical sense to change it in hw raid.

It was possible to set before, but was not possible to check ;) The only 
practical sense is with hw raid)

------------------------------------------------------------------------------
All the data continuously generated in your IT infrastructure contains a
definitive record of customers, application performance, security
threats, fraudulent activity and more. Splunk takes this data and makes
sense of it. Business sense. IT sense. Common sense.
http://p.sf.net/sfu/splunk-d2dcopy1
Permalink | Reply | 3 comments |
headers
Alex Samorukov | 3 Oct 2011 14:11
Picon

Re: MegaRAID SAT layer is reportedly buggy

On 10/03/2011 02:15 PM, Artem Bokhan wrote:
>>>> "-l xerror -l xselftest -l directory -l scttemp -l scterc -l sataphy"
>>>>> None of above. Actually the -x adds one more than that: -l scttemp,
>>>>> which increases the Other Error Count by 4 each run.
>>>>>
>>>> Please try latest SVN - i submitted patch to resolve this bug (for
>>>> SAT12/SAT16 protocols). It works for me, at least.
>>>
>>> 1. Does this break scterc set?
>>>
>>> Device: /dev/sg0 [megaraid_disk_57] [SAT], set of SCT Error Recovery 
>>> Control failed
>> I don`t think that scterc set is possible with or without this patch. 
>> Also i don`t think that there is any practical sense to change it in 
>> hw raid.
> orukov wrote:
>
> It was possible to set before, but was not possible to check ;) The 
> only practical sense is with hw raid)
>
Why do you think that it was set then?

------------------------------------------------------------------------------
All the data continuously generated in your IT infrastructure contains a
definitive record of customers, application performance, security
threats, fraudulent activity and more. Splunk takes this data and makes
sense of it. Business sense. IT sense. Common sense.
http://p.sf.net/sfu/splunk-d2dcopy1
Permalink | Reply | 2 comments |
headers
Artem Bokhan | 3 Oct 2011 14:37
Picon
Favicon

Re: MegaRAID SAT layer is reportedly buggy

On 03.10.2011 19:11, Alex Samorukov wrote:
>>
>> It was possible to set before, but was not possible to check ;) The only 
>> practical sense is with hw raid)
>>
>
> Why do you think that it was set then?
>

The command passed to FW without errors. It is really better to ask Christian.

------------------------------------------------------------------------------
All the data continuously generated in your IT infrastructure contains a
definitive record of customers, application performance, security
threats, fraudulent activity and more. Splunk takes this data and makes
sense of it. Business sense. IT sense. Common sense.
http://p.sf.net/sfu/splunk-d2dcopy1
Permalink | Reply | 1 comment |
headers
Artem Bokhan | 3 Oct 2011 14:58
Picon
Favicon

Re: MegaRAID SAT layer is reportedly buggy

On 03.10.2011 19:37, Artem Bokhan wrote:
> On 03.10.2011 19:11, Alex Samorukov wrote:
>>> It was possible to set before, but was not possible to check ;) The only
>>> practical sense is with hw raid)
>>>
>> Why do you think that it was set then?
>>
> The command passed to FW without errors. It is really better to ask Christian.

Drives that do not support scterc caused an error with set.

> ------------------------------------------------------------------------------
> All the data continuously generated in your IT infrastructure contains a
> definitive record of customers, application performance, security
> threats, fraudulent activity and more. Splunk takes this data and makes
> sense of it. Business sense. IT sense. Common sense.
> http://p.sf.net/sfu/splunk-d2dcopy1
> _______________________________________________
> Smartmontools-support mailing list
> Smartmontools-support <at> lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/smartmontools-support

------------------------------------------------------------------------------
All the data continuously generated in your IT infrastructure contains a
definitive record of customers, application performance, security
threats, fraudulent activity and more. Splunk takes this data and makes
sense of it. Business sense. IT sense. Common sense.
http://p.sf.net/sfu/splunk-d2dcopy1
Permalink | Reply | 0 comments |
headers
Thomas Mieslinger | 4 Oct 2011 08:09
Picon
Favicon

Re: Please add support for Renice Z2 SSDs

> -----Original Message-----
> From: Christian Franke [mailto:Christian.Franke <at> t-online.de]
> Sent: Thursday, September 29, 2011 11:17 PM
> To: Thomas Mieslinger
> [..]
> Please provide a smartctl -A output. Otherwise we cannot check whether
> this drive uses the same attribute format as the indilinx devices
> already known.

Smartctl -A /dev/hda

smartctl 5.41 2011-06-09 r3365 [i686-w64-mingw32-win7(64)-sp1] (sf-win32-5.41-1)

Copyright (C) 2002-11 by Bruce Allen, http://smartmontools.sourceforge.net

=== START OF READ SMART DATA SECTION ===
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     0x0000   006   000   000    Old_age   Offline      -
       0
  9 Power_On_Hours          0x0000   117   006   000    Old_age   Offline      -
       0
 12 Power_Cycle_Count       0x0000   067   003   000    Old_age   Offline      -
       0
184 End-to-End_Error        0x0000   072   000   000    Old_age   Offline      -
       0
195 Hardware_ECC_Recovered  0x0000   000   000   000    Old_age   Offline      -
       0
(Continue reading)

Permalink | Reply | 3 comments |
headers
Alex Samorukov | 6 Oct 2011 18:58
Picon

Re: RocketRAID 2340 Devices 9-16

On 09/25/2011 10:18 AM, sourceforge <at> photosurrealism.com wrote:
> I installed smartmontools on FreeBSD and was unable to access drives
> above 8 on my RocketRAID 2340 (it has 16).   Increasing the limit in
> the argument parser seems to solve the problem.  I can now see them all.
> They all report different serial numbers, so I'm guessing it's working
> correctly
>
> There appears to be code to find "maxchannel" for the controller, so
> is the argument check really necessary?  I didn't dip in deep enough to
> see really.
>
> Changing the if on line 1768 of os_freebsd.cpp seems sufficient for
> function, but obviously the warning messages and documentation would
> need to be brought to match.  Patch below.
I increased channel number to 16 in os_linux/os_freebsd and corrected 
documentation.

Please confirm if SVN version works for you.

P.S. I found that hpt driver is using old interface. If you are 
interested to test i can try to convert it to the  new code which is 
better and more advanced.
P.P.S. maxchannel is used only for old ata driver. I found no hard 
limits for channel number in driver sources, so let it be 16 for now.

------------------------------------------------------------------------------
All the data continuously generated in your IT infrastructure contains a
definitive record of customers, application performance, security
threats, fraudulent activity and more. Splunk takes this data and makes
sense of it. Business sense. IT sense. Common sense.
(Continue reading)

Permalink | Reply | 0 comments |
headers
Christian Franke | 6 Oct 2011 23:59
Picon
Favicon

Re: Please add support for Renice Z2 SSDs

Thomas Mieslinger wrote:
>> -----Original Message-----
>> From: Christian Franke [mailto:Christian.Franke <at> t-online.de]
>> Sent: Thursday, September 29, 2011 11:17 PM
>> To: Thomas Mieslinger
>> [..]
>> Please provide a smartctl -A output. Otherwise we cannot check whether
>> this drive uses the same attribute format as the indilinx devices
>> already known.
> Smartctl -A /dev/hda
>
> smartctl 5.41 2011-06-09 r3365 [i686-w64-mingw32-win7(64)-sp1] (sf-win32-5.41-1)

Patch included in r3427. Run update-smart-drivedb.exe to test.

Thanks,
Christian

------------------------------------------------------------------------------
All the data continuously generated in your IT infrastructure contains a
definitive record of customers, application performance, security
threats, fraudulent activity and more. Splunk takes this data and makes
sense of it. Business sense. IT sense. Common sense.
http://p.sf.net/sfu/splunk-d2dcopy1
Permalink | Reply | 2 comments |

Gmane