Topic: RAID1 Problem (again)

[janet]

SchulzStefan

I suggest you perform a full backup to locally connected USB disk, BEFORE you lose data, as you are at a big risk now of losing data.

Then do a new install of sme8.0 from CD & reformat (erase) the existing drives in the process & wipe all data & partitions etc from your current system
Select software RAID1 option, although with two disks installed the installer should default to that option.
Do NOT choose the upgrade option.

You shoulc then have system with a properly (& automatically) configured & sync'd software RAID1 array

Run yum update
Then restore from USB backup
Then reinstall contribs

Your system should be fully functional with a correctly configured software RAID1 array

I advise you in future to stop playing with RAID arrays & let your system automatically configure itself.

The admin console (log in as admin) has a menu option to add a drive to an array when a drive has been replaced due to faults or errors.

REMEMBER every single time you reuse a drive, even when from the same system, & BEFORE adding it to a RAID array either automatically or manually, you MUST erase the partion & MBR information using the dd command here
http://wiki.contribs.org/Raid#Reusing_Hard_Drives
eg
dd if=/dev/zero of=/dev/sdb bs=512 count=1
(replace sdb with the location of the drive to be erased eg sda sdb sdc sdd etc)
Check drives installed & their details using
fdisk -l
 
The above step (erase MBR) is ESSSENTIAL to do whenever the drive has been already used, all drives added to arrays should (MUST) be blank

It seems to me you have been breaking some (technical) rules, & so you are having problems.

[SchulzStefan]

janet, thank you for your reply.

I suggest you perform a full backup to locally connected USB disk, BEFORE you lose data, as you are at a big risk now of losing data.

I'm doing this twice a day, one affa and one with the server-manager.

REMEMBER every single time you reuse a drive, even when from the same system, & BEFORE adding it to a RAID array either automatically or manually, you MUST erase the partion & MBR information using the dd command here
http://wiki.contribs.org/Raid#Reusing_Hard_Drives
eg
dd if=/dev/zero of=/dev/sdb bs=512 count=1
(replace sdb with the location of the drive to be erased eg sda sdb sdc sdd etc)
Check drives installed & their details using
fdisk -L
 
The above step (erase MBR) is ESSSENTIAL to do whenever the drive has been already used, all drives added to arrays should (MUST) be blank

Copying data with ddrescue on a drive says clearly "all data on the target (drive) will be erased". I assumed that ddrescue is exactly that doing. Maybe that was wrong, I don't know really.

I advise you in future to stop playing with RAID arrays & let your system automatically configure itself.

The admin console (log in as admin) has a menu option to add a drive to an array when a drive has been replaced due to faults or errors.

Well, I bought a new drive, (didn't format it, didn't partition it) plugged it in the server, and did what you suggest.

What makes me wonder is, before I made the copy on the (now) sda-disk, I run a long-test from WD. Took around 3 hours. No errors have been reported. Then I run ddrescue on this target.

I booted the server with this disk, took the other out before, and I got no errors from sda until I plugged the new disk on the board, and added sdb via admin console to the RAID. How can this be?

Maybe the Western Digital WD10EFRX Red 1TB are the problem. They are made for NAS and RAID. I don't know. Need some time for this try. If it fails, I'll buy two new, other than the WD Red, disks.

Then do a new install of sme8.0 from CD & reformat (erase) the existing drives in the process & wipe all data & partitions etc from your current system
Select software RAID1 option, although with two disks installed the installer should default to that option.

and

REMEMBER every single time you reuse a drive, even when from the same system, & BEFORE adding it to a RAID array either automatically or manually, you MUST erase the partion & MBR information using the dd command here
http://wiki.contribs.org/Raid#Reusing_Hard_Drives
eg
dd if=/dev/zero of=/dev/sdb bs=512 count=1
(replace sdb with the location of the drive to be erased eg sda sdb sdc sdd etc)
Check drives installed & their details using
fdisk -L

As both disks are used now, I'll zero them as you suggest. But I'm not really convinced. We'll see, I'll let you know.

[janet]

SchulzStefan

As both disks are used now, I'll zero them as you suggest. But I'm not really convinced. We'll see, I'll let you know.

At least you have finally heeded the advice to zero the disks using the recommended method (ie use dd command & not ddrescue).

Partitions & MBR's are not data, so while ddrescue may "delete all data on target drive", it may not be doing what is really needed. I do not know as I do not use it. The point here is that you are copying disks with ddrescue, which is not what rebuilding an array is all about.
If one disk has a problem & you copy it, then usually you will copy the problem, get it ?
When you mirror that disk in RAID1 with the (copied) problem, then you mirror the problem.

That's why I said to start afresh with a newly build RAID1 array, without any data on it, and then restore your data to that good & clean RAID1 array.
To me it is the quickest & best way to resolve your issues.

No one can easily troubleshoot your problem via Internet forum as you have made so many changes. It's really a hands on job to examine your drives etc.

Edit: To me it's prudent or wise to even zero out a new drive, at least that way you have totally ruled out the possibility of the drive being problematic for related reasons.

[SchulzStefan]

janet,

Edit: To me it's prudent or wise to even zero out a new drive, at least that way you have totally ruled out the possibility of the drive being problematic for related reasons.

I'll try your advice.

I got this from the brand new drive which I yesterday plugged in:

Feb 13 05:23:22 saturn kernel: ata2.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x0
Feb 13 05:23:22 saturn kernel: ata2.00: BMDMA stat 0x25
Feb 13 05:23:22 saturn kernel: ata2.00: cmd 35/00:08:02:59:70/00:00:74:00:00/e0 tag 0 dma 4096 out
Feb 13 05:23:22 saturn kernel:          res 51/10:08:02:59:70/10:00:74:00:00/e0 Emask 0x81 (invalid argument)
Feb 13 05:23:22 saturn kernel: ata2.00: status: { DRDY ERR }
Feb 13 05:23:22 saturn kernel: ata2.00: error: { IDNF }
Feb 13 05:23:36 saturn kernel: ata2.00: configured for UDMA/133
Feb 13 05:23:36 saturn kernel: sd 1:0:0:0: Unhandled sense code
Feb 13 05:23:36 saturn kernel: sd 1:0:0:0: SCSI error: return code = 0x08000002
Feb 13 05:23:36 saturn kernel: Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE,SUGGEST_OK
Feb 13 05:23:36 saturn kernel: sdb: Current [descriptor]: sense key: Aborted Command
Feb 13 05:23:36 saturn kernel:     Add. Sense: Recorded entity not found
Feb 13 05:23:36 saturn kernel:
Feb 13 05:23:38 saturn kernel: Descriptor sense data with sense descriptors (in hex):
Feb 13 05:23:38 saturn kernel:         72 0b 14 00 00 00 00 0c 00 0a 80 00 00 00 00 00
Feb 13 05:23:39 saturn kernel:         74 70 59 02
Feb 13 05:23:39 saturn kernel: ata2: EH complete
Feb 13 05:23:39 saturn kernel: raid1: Disk failure on sdb2, disabling device.
Feb 13 05:23:39 saturn kernel:    Operation continuing on 1 devices
Feb 13 05:23:39 saturn kernel: SCSI device sdb: 1953525168 512-byte hdwr sectors (1000205 MB)
Feb 13 05:23:39 saturn kernel: sdb: Write Protect is off
Feb 13 05:23:39 saturn kernel: sdb: Mode Sense: 00 3a 00 00
Feb 13 05:23:39 saturn kernel: SCSI device sdb: drive cache: write back

# cat /proc/mdstat
Personalities : [raid1]
md1 : active raid1 sdb1[1] sda1[0]
      104320 blocks [2/2] [UU]
     
md2 : active raid1 sdb2[2](F) sda2[0]
      976655488 blocks [2/1] [U_]
     
unused devices: <none>

smartctl -x /dev/sdb
smartctl 5.42 2011-10-20 r3458 [i686-linux-2.6.18-371.4.1.el5PAE] (local build)
Copyright (C) 2002-11 by Bruce Allen, http://smartmontools.sourceforge.net

=== START OF INFORMATION SECTION ===
Device Model:     WDC WD10EFRX-68PJCN0
Serial Number:    WD-WMC4J0189055
LU WWN Device Id: 5 0014ee 25ebeaa72
Firmware Version: 01.01A01
User Capacity:    1.000.204.886.016 bytes [1,00 TB]
Sector Sizes:     512 bytes logical, 4096 bytes physical
Device is:        Not in smartctl database [for details use: -P showall]
ATA Version is:   8
ATA Standard is:  ACS-2 (revision not indicated)
Local Time is:    Thu Feb 13 09:07:42 2014 CET
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:  (0x00)   Offline data collection activity
               was never started.
               Auto Offline Data Collection: Disabled.
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:       (13980) 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:     ( 159) minutes.
Conveyance self-test routine
recommended polling time:     (   5) minutes.
SCT capabilities:           (0x303d)   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          FLAGS    VALUE WORST THRESH FAIL RAW_VALUE
  1 Raw_Read_Error_Rate     POSR-K   100   253   051    -    0
  3 Spin_Up_Time            POS--K   100   253   021    -    0
  4 Start_Stop_Count        -O--CK   100   100   000    -    1
  5 Reallocated_Sector_Ct   PO--CK   200   200   140    -    0
  7 Seek_Error_Rate         -OSR-K   200   200   000    -    0
  9 Power_On_Hours          -O--CK   100   100   000    -    17
 10 Spin_Retry_Count        -O--CK   100   253   000    -    0
 11 Calibration_Retry_Count -O--CK   100   253   000    -    0
 12 Power_Cycle_Count       -O--CK   100   100   000    -    1
192 Power-Off_Retract_Count -O--CK   200   200   000    -    0
193 Load_Cycle_Count        -O--CK   200   200   000    -    1047
194 Temperature_Celsius     -O---K   115   109   000    -    28
196 Reallocated_Event_Count -O--CK   200   200   000    -    0
197 Current_Pending_Sector  -O--CK   200   200   000    -    0
198 Offline_Uncorrectable   ----CK   100   253   000    -    0
199 UDMA_CRC_Error_Count    -O--CK   200   200   000    -    0
200 Multi_Zone_Error_Rate   ---R--   100   253   000    -    0
                            ||||||_ K auto-keep
                            |||||__ C event count
                            ||||___ R error rate
                            |||____ S speed/performance
                            ||_____ O updated online
                            |______ P prefailure warning

General Purpose Log Directory Version 1
SMART           Log Directory Version 1 [multi-sector log support]
GP/S  Log at address 0x00 has    1 sectors [Log Directory]
SMART Log at address 0x01 has    1 sectors [Summary SMART error log]
SMART Log at address 0x02 has    5 sectors [Comprehensive SMART error log]
GP    Log at address 0x03 has    6 sectors [Ext. Comprehensive SMART error log]
SMART Log at address 0x06 has    1 sectors [SMART self-test log]
GP    Log at address 0x07 has    1 sectors [Extended self-test log]
SMART Log at address 0x09 has    1 sectors [Selective self-test log]
GP    Log at address 0x10 has    1 sectors [NCQ Command Error log]
GP    Log at address 0x11 has    1 sectors [SATA Phy Event Counters]
GP    Log at address 0x21 has    1 sectors [Write stream error log]
GP    Log at address 0x22 has    1 sectors [Read stream error log]
GP/S  Log at address 0x80 has   16 sectors [Host vendor specific log]
GP/S  Log at address 0x81 has   16 sectors [Host vendor specific log]
GP/S  Log at address 0x82 has   16 sectors [Host vendor specific log]
GP/S  Log at address 0x83 has   16 sectors [Host vendor specific log]
GP/S  Log at address 0x84 has   16 sectors [Host vendor specific log]
GP/S  Log at address 0x85 has   16 sectors [Host vendor specific log]
GP/S  Log at address 0x86 has   16 sectors [Host vendor specific log]
GP/S  Log at address 0x87 has   16 sectors [Host vendor specific log]
GP/S  Log at address 0x88 has   16 sectors [Host vendor specific log]
GP/S  Log at address 0x89 has   16 sectors [Host vendor specific log]
GP/S  Log at address 0x8a has   16 sectors [Host vendor specific log]
GP/S  Log at address 0x8b has   16 sectors [Host vendor specific log]
GP/S  Log at address 0x8c has   16 sectors [Host vendor specific log]
GP/S  Log at address 0x8d has   16 sectors [Host vendor specific log]
GP/S  Log at address 0x8e has   16 sectors [Host vendor specific log]
GP/S  Log at address 0x8f has   16 sectors [Host vendor specific log]
GP/S  Log at address 0x90 has   16 sectors [Host vendor specific log]
GP/S  Log at address 0x91 has   16 sectors [Host vendor specific log]
GP/S  Log at address 0x92 has   16 sectors [Host vendor specific log]
GP/S  Log at address 0x93 has   16 sectors [Host vendor specific log]
GP/S  Log at address 0x94 has   16 sectors [Host vendor specific log]
GP/S  Log at address 0x95 has   16 sectors [Host vendor specific log]
GP/S  Log at address 0x96 has   16 sectors [Host vendor specific log]
GP/S  Log at address 0x97 has   16 sectors [Host vendor specific log]
GP/S  Log at address 0x98 has   16 sectors [Host vendor specific log]
GP/S  Log at address 0x99 has   16 sectors [Host vendor specific log]
GP/S  Log at address 0x9a has   16 sectors [Host vendor specific log]
GP/S  Log at address 0x9b has   16 sectors [Host vendor specific log]
GP/S  Log at address 0x9c has   16 sectors [Host vendor specific log]
GP/S  Log at address 0x9d has   16 sectors [Host vendor specific log]
GP/S  Log at address 0x9e has   16 sectors [Host vendor specific log]
GP/S  Log at address 0x9f has   16 sectors [Host vendor specific log]
GP/S  Log at address 0xa0 has   16 sectors [Device vendor specific log]
GP/S  Log at address 0xa1 has   16 sectors [Device vendor specific log]
GP/S  Log at address 0xa2 has   16 sectors [Device vendor specific log]
GP/S  Log at address 0xa3 has   16 sectors [Device vendor specific log]
GP/S  Log at address 0xa4 has   16 sectors [Device vendor specific log]
GP/S  Log at address 0xa5 has   16 sectors [Device vendor specific log]
GP/S  Log at address 0xa6 has   16 sectors [Device vendor specific log]
GP/S  Log at address 0xa7 has   16 sectors [Device vendor specific log]
GP/S  Log at address 0xa8 has    1 sectors [Device vendor specific log]
GP/S  Log at address 0xa9 has    1 sectors [Device vendor specific log]
GP/S  Log at address 0xaa has    1 sectors [Device vendor specific log]
GP/S  Log at address 0xab has    1 sectors [Device vendor specific log]
GP/S  Log at address 0xac has    1 sectors [Device vendor specific log]
GP/S  Log at address 0xad has    1 sectors [Device vendor specific log]
GP/S  Log at address 0xae has    1 sectors [Device vendor specific log]
GP/S  Log at address 0xaf has    1 sectors [Device vendor specific log]
GP/S  Log at address 0xb0 has    1 sectors [Device vendor specific log]
GP/S  Log at address 0xb1 has    1 sectors [Device vendor specific log]
GP/S  Log at address 0xb2 has    1 sectors [Device vendor specific log]
GP/S  Log at address 0xb3 has    1 sectors [Device vendor specific log]
GP/S  Log at address 0xb4 has    1 sectors [Device vendor specific log]
GP/S  Log at address 0xb5 has    1 sectors [Device vendor specific log]
GP/S  Log at address 0xb6 has    1 sectors [Device vendor specific log]
GP/S  Log at address 0xb7 has    1 sectors [Device vendor specific log]
GP/S  Log at address 0xbd has    1 sectors [Device vendor specific log]
GP/S  Log at address 0xc0 has    1 sectors [Device vendor specific log]
GP    Log at address 0xc1 has   93 sectors [Device vendor specific log]
GP/S  Log at address 0xe0 has    1 sectors [SCT Command/Status]
GP/S  Log at address 0xe1 has    1 sectors [SCT Data Transfer]

SMART Extended Comprehensive Error Log Version: 1 (6 sectors)
Device Error Count: 1
   CR     = Command Register
   FEATR  = Features Register
   COUNT  = Count (was: Sector Count) Register
   LBA_48 = Upper bytes of LBA High/Mid/Low Registers ]  ATA-8
   LH     = LBA High (was: Cylinder High) Register    ]   LBA
   LM     = LBA Mid (was: Cylinder Low) Register      ] Register
   LL     = LBA Low (was: Sector Number) Register     ]
   DV     = Device (was: Device/Head) Register
   DC     = Device Control Register
   ER     = Error register
   ST     = Status register
Powered_Up_Time is measured from power on, and printed as
DDd+hh:mm:SS.sss where DD=days, hh=hours, mm=minutes,
SS=sec, and sss=millisec. It "wraps" after 49.710 days.

Error 1

• occurred at disk power-on lifetime: 13 hours (0 days + 13 hours)

  When the command that caused the error occurred, the device was active or idle.

  After command completion occurred, registers were:
  ER -- ST COUNT  LBA_48  LH LM LL DV DC
  -- -- -- == -- == == == -- -- -- -- --
  10 -- 51 00 08 00 00 74 70 59 02 e0 00  Error: IDNF 8 sectors at LBA = 0x74705902 = 1953519874

  Commands leading to the command that caused the error were:
  CR FEATR COUNT  LBA_48  LH LM LL DV DC  Powered_Up_Time  Command/Feature_Name
  -- == -- == -- == == == -- -- -- -- --  ---------------  --------------------
  35 00 00 00 08 00 00 74 70 59 02 e0 08     13:45:17.466  WRITE DMA EXT
  ea 00 00 00 00 00 00 74 70 59 09 e0 08     13:45:12.148  FLUSH CACHE EXT
  ea 00 00 00 00 00 00 74 70 59 09 e0 08     13:45:11.575  FLUSH CACHE EXT
  35 00 00 00 08 00 00 74 70 59 02 e0 08     13:45:06.056  WRITE DMA EXT
  ea 00 00 00 00 00 00 74 70 59 09 e0 08     13:45:04.380  FLUSH CACHE EXT

SMART Extended Self-test Log Version: 1 (1 sectors)
No self-tests have been logged.  [To run self-tests, use: smartctl -t]

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.

SCT Status Version:                  3
SCT Version (vendor specific):       258 (0x0102)
SCT Support Level:                   1
Device State:                        Active (0)
Current Temperature:                    28 Celsius
Power Cycle Min/Max Temperature:     25/34 Celsius
Lifetime    Min/Max Temperature:     25/34 Celsius
Under/Over Temperature Limit Count:   0/0
SCT Temperature History Version:     2
Temperature Sampling Period:         1 minute
Temperature Logging Interval:        1 minute
Min/Max recommended Temperature:      0/60 Celsius
Min/Max Temperature Limit:           -41/85 Celsius
Temperature History Size (Index):    478 (92)

Index    Estimated Time   Temperature Celsius
  93    2014-02-13 01:10    29  **********
 ...    ..(  3 skipped).    ..  **********
  97    2014-02-13 01:14    29  **********
  98    2014-02-13 01:15    30  ***********
 ...    ..(  7 skipped).    ..  ***********
 106    2014-02-13 01:23    30  ***********
 107    2014-02-13 01:24    31  ************
 108    2014-02-13 01:25    31  ************
 109    2014-02-13 01:26    30  ***********
 110    2014-02-13 01:27    31  ************
 ...    ..( 19 skipped).    ..  ************
 130    2014-02-13 01:47    31  ************
 131    2014-02-13 01:48    32  *************
 ...    ..( 34 skipped).    ..  *************
 166    2014-02-13 02:23    32  *************
 167    2014-02-13 02:24    31  ************
 ...    ..( 14 skipped).    ..  ************
 182    2014-02-13 02:39    31  ************
 183    2014-02-13 02:40    30  ***********
 ...    ..( 20 skipped).    ..  ***********
 204    2014-02-13 03:01    30  ***********
 205    2014-02-13 03:02    29  **********
 ...    ..(  4 skipped).    ..  **********
 210    2014-02-13 03:07    29  **********
 211    2014-02-13 03:08    30  ***********
 212    2014-02-13 03:09    30  ***********
 213    2014-02-13 03:10    30  ***********
 214    2014-02-13 03:11    29  **********
 215    2014-02-13 03:12    30  ***********
 ...    ..(  7 skipped).    ..  ***********
 223    2014-02-13 03:20    30  ***********
 224    2014-02-13 03:21    29  **********
 225    2014-02-13 03:22    30  ***********
 226    2014-02-13 03:23    29  **********
 ...    ..( 87 skipped).    ..  **********
 314    2014-02-13 04:51    29  **********
 315    2014-02-13 04:52    28  *********
 ...    ..( 10 skipped).    ..  *********
 326    2014-02-13 05:03    28  *********
 327    2014-02-13 05:04    29  **********
 ...    ..(239 skipped).    ..  **********
  89    2014-02-13 09:04    29  **********
  90    2014-02-13 09:05    28  *********
  91    2014-02-13 09:06    28  *********
  92    2014-02-13 09:07    28  *********

SCT Error Recovery Control:
           Read:     70 (7,0 seconds)
          Write:     70 (7,0 seconds)

SATA Phy Event Counters (GP Log 0x11)
ID      Size     Value  Description
0x0001  2            0  Command failed due to ICRC error
0x0002  2            0  R_ERR response for data FIS
0x0003  2            0  R_ERR response for device-to-host data FIS
0x0004  2            0  R_ERR response for host-to-device data FIS
0x0005  2            0  R_ERR response for non-data FIS
0x0006  2            0  R_ERR response for device-to-host non-data FIS
0x0007  2            0  R_ERR response for host-to-device non-data FIS
0x0008  2            0  Device-to-host non-data FIS retries
0x0009  2            5  Transition from drive PhyRdy to drive PhyNRdy
0x000a  2            6  Device-to-host register FISes sent due to a COMRESET
0x000b  2            0  CRC errors within host-to-device FIS
0x000f  2            0  R_ERR response for host-to-device data FIS, CRC
0x0012  2            0  R_ERR response for host-to-device non-data FIS, CRC
0x8000  4        62972  Vendor specific

It seems to me, that beside your hints, that it could also be a misconfiguration or an incompatibility.

Feb 13 05:23:22 saturn kernel: ata2.00: cmd 35/00:08:02:59:70/00:00:74:00:00/e0 tag 0 dma 4096 out
Feb 13 05:23:22 saturn kernel: res 51/10:08:02:59:70/10:00:74:00:00/e0 Emask 0x81 (invalid argument)

That's not normal to me, that a brand new disk has dma errors after a few hours running.

[SchulzStefan]

WD says, the Western-Digital-Red-WD10EFRX is linux compatible.

BUT: they are sata 6 Gb/s (SATA 3). The Board is an ASUS as P5Q SE2, which has only sata 3 Gb/s. I assume, that this maybe the reason for the errors I have with this disks. If so, I'm wondering whether the disks are not down-compatible.

Edit: WD says also, they are backward compatible.

I asked my dealer if SATA 6 Gb/s is backward compatible. He told me, there should be no problem. Perhaps it's a specific WD problem? I could buy a seagate disk and try how this works. Don't know how to proceed...

Anybody who could confirm?

[janet]

SchulzStefan

Don't know how to proceed...

You keep asking for help.
Did you follow the advice already given "ie backup, rebuild & restore" ?

[SchulzStefan]

janet,

I can't rebuild the server (with zeroing the disks) during the week. I have to wait for the weekend. I will then try what you suggested.

You keep asking for help.

Sure. If anybody could confirm a dma error is caused due a incompatibility from sata 3 Gb/s to sata 6 Gb/s with this disks, I could save time to rebuild the server twice.

[janet]

SchulzStefan

If anybody could confirm a dma error is caused due a incompatibility from sata 3 Gb/s to sata 6 Gb/s with this disks, I could save time to rebuild the server twice.

That is just a maximum rated data transfer speed. Any disk can run at a slower rate, it will in practise anyway, depending on how much data is being read or saved.
I doubt that there are any speed incompatibility issues.

[SchulzStefan]

janet,

I suggest you perform a full backup to locally connected USB disk

Question: is it o.k. to use my workstation backup (configured via server-manager), or do you suggest to do a backup to USB via the admin-console?

[janet]

SchulzStefan

is it o.k. to use my workstation backup (configured via server-manager), or do you suggest to do a backup to USB via the admin-console?

Either method should work, there are pros & cons.
In order to use the workstation backup to restore from, you will need to configure your newly rebuilt sme server (with clean OS install from CD), so that it has the same local IP & network configuration, server name & domain name etc. Then setup the identical backup job in server manager. Doing that will allow you to perform the restore ftom within the server manager panel.

Backup & restore to/from a workstation across a network connection could be considered to be slower. Restoring 500Gb could take anywhere from 4-24 hours depending on your system specs etc.

Using the same workstation backup server manager panel, but backing up to a locally connected USB drive, removes workstation & network speed issues. You still need to recreate the backup job in server manager on the new server (with identical details), in order to do the restore.

If you use the admin console "one off" backup to USB, then it is very easy to start the restore. At first boot of the new server after installing OS from CD, you will be asked once if you want to restore from local USB drive. That is when you connect the USB drive you performed the console backup to. It will start the restore immediately then without needing any further configuration of the server. So the admin console backup has an easier restore procedure.

This feature ideally suits rebuilding SME on newer or different hardware.

The method you use is your choice. I personally pfefer one of the backup to locally connected USB methods, less hassle with networks & workstations then.

[SchulzStefan]

janet,

I included several directories onto the standard backup directories. Will they be saved via USB-Backup also? And what about the compression?

[janet]

SchulzStefan

I included several directories onto the standard backup directories. Will they be saved via USB-Backup also? And what about the compression?

If you followed this
http://wiki.contribs.org/Backup_with_dar
& particularly this
http://wiki.contribs.org/Backup_with_dar#Adding_Files_and_Directories
then additional folders should be included in your backup, made using one of the standard default SME backup methods.

You can easily check a backup to see what is included.
IIRC dar files can be viewed in midnight commander & other ways too, Google for it.

Read this document for useful info about backup & restore, and alternative approaches.
It may help you to better understand the backup & restore process.
http://wiki.contribs.org/Backup_server_config

[SchulzStefan]

janet,

Run yum update
Then restore from USB backup
Then reinstall contribs

Edit:
Is a reboot required after the restore? Or should I install the contribs before rebooting?

Are you sure about the order? Why not installing first the contribs and then restore?

[janet]

SchulzStefan

Are you sure about the order? Why not installing first the contribs and then restore?

Why?
Do you have better knowledge of SME server than the developers ?

The procedure order:
Install clean OS
Restore from backup
Reinstall contribs
is and always has been the recommended procedure by devs for performing a SME server full backup & restore.

Note too that a full restore should NEVER be performed to a SME server that is configured with users, ibays & data & so on, other than basic network setup required to allow the restore to be performed (as that basic setup will be overwritten by the restored data).

A full restore expects the SME server file system to be in a certain pre defined  "clean" state, as exists after a fesh install from a CD.
Other non default settings & configuration (eg ibays that are not in the backup) are not overwritten or written to the databases, refer to the standard backup inclusion to see what is written back to the server.
You can end up with a server in an indeterminate state if you do a full restore to an existing (not clean) highly configured installation.

Here is an example to prove this design concept:
if you use the admin console backup, you are then given an opportunity to restore from USB on first boot of the server immediately after you install the fresh OS from CD, clearly well before any contribs have been or can be reinstalled.

Obviously you did not read the Backup Server Config wiki Howto I pointed you to, as it is described there also, and in many posts in these forums by key developers.

IIUC the rpm contrib package will reconfigure itself with new default data if no restored data exists.
If restore is done first, then when the contrib rpm is installed, the rpm will see the existing config & data for the contrib and use that instead, so no need to reconfigure the contrib.

[SchulzStefan]

janet,

Do you have better knowledge of SME server than the developers ?

Where did I say that?

I couldn't find an answer to my question:

Edit:
Is a reboot required after the restore? Or should I install the contribs before rebooting?

Obviously you did not read the Backup Server Config wiki Howto I pointed you to, as it is described there also, and in many posts in these forums by key developers.

Where has this question been answered?

It was a technical question. After restoring the server form the backup, there have also settings been restored from several contribs. While installing the contribs the restored settings might be overwritten. And how do I get back this settings?

Example: Installing firebird will overwrite the restored settings. How about openvpn bridge, zarafa, phpki and so on.