unknown partition table dell 720 centOS 7 after kernel upgrade

MightyU · 02-07-2016, 10:32 AM

Dear all,
I am managing a dell r720 server, which is configured with 2 RAID partitions (see below) via the dell H710 raid manager.
after the latest kernel update of centos 7, my server floods the error log with the following error :
"/dev/sdb unknown partition table"

the report of the following command:

[CODE]
omreport storage vdisk
[CODE]

gives

Code:

List of Virtual Disks in the System

Controller PERC H710P Mini (Embedded)
ID                                : 0
Status                            : Ok
Name                              : VD-1
State                             : Ready
Hot Spare Policy violated         : Not Assigned
Encrypted                         : No
Layout                            : RAID-1
Size                              : 837.75 GB (899527213056 bytes)
T10 Protection Information Status : No
Associated Fluid Cache State      : Not Applicable
Device Name                       : /dev/sda
Bus Protocol                      : SAS
Media                             : HDD
Read Policy                       : Adaptive Read Ahead
Write Policy                      : Write Back
Cache Policy                      : Not Applicable
Stripe Element Size               : 64 KB
Disk Cache Policy                 : Disabled

ID                                : 1
Status                            : Ok
Name                              : VD-2
State                             : Ready
Hot Spare Policy violated         : Not Assigned
Encrypted                         : No
Layout                            : RAID-5
Size                              : 18,627.50 GB (20001125826560 bytes)
T10 Protection Information Status : No
Associated Fluid Cache State      : Not Applicable
Device Name                       : /dev/sdb
Bus Protocol                      : SAS
Media                             : HDD
Read Policy                       : Adaptive Read Ahead
Write Policy                      : Write Back
Cache Policy                      : Not Applicable
Stripe Element Size               : 64 KB
Disk Cache Policy                 : Disabled

the output of the lsscsi command gives:

Code:

[0:2:0:0]    disk    DELL     PERC H710P       3.13  /dev/sda
[0:2:1:0]    disk    DELL     PERC H710P       3.13  /dev/sdb

dmesg gives:

Code:

[    2.625903] scsi 0:2:0:0: Direct-Access     DELL     PERC H710P       3.13 PQ: 0 ANSI: 5
[    2.626089] scsi 0:2:1:0: Direct-Access     DELL     PERC H710P       3.13 PQ: 0 ANSI: 5
[    2.626892] mpt2sas0: MSI-X vectors supported: 1, no of cores: 8, max_msix_vectors: 8
[    2.626943] mpt2sas 0000:03:00.0: irq 69 for MSI/MSI-X
[    2.626975] mpt2sas0-msix0: PCI-MSI-X enabled: IRQ 69
[    2.626978] mpt2sas0: iomem(0x00000000d9ff0000), mapped(0xffffc90016980000), size(65536)
[    2.626979] mpt2sas0: ioport(0x000000000000dc00), size(256)
[    2.632453] fbcon: mgadrmfb (fb0) is primary device
[    2.645253] sd 0:2:0:0: [sda] 1756889088 512-byte logical blocks: (899 GB/837 GiB)
[    2.645273] sd 0:2:1:0: [sdb] 39064698880 512-byte logical blocks: (20.0 TB/18.1 TiB)
[    2.645343] sd 0:2:0:0: [sda] Write Protect is off
[    2.645346] sd 0:2:0:0: [sda] Mode Sense: 1f 00 00 08
[    2.645361] sd 0:2:1:0: [sdb] Write Protect is off
[    2.645364] sd 0:2:1:0: [sdb] Mode Sense: 1f 00 00 08
[    2.645404] sd 0:2:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
[    2.645426] sd 0:2:1:0: [sdb] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
[    2.646054]  sda: sda1 sda2
[    2.646214]  sdb: unknown partition table
[    2.646429] sd 0:2:0:0: [sda] Attached SCSI disk
[    2.646499] sd 0:2:1:0: [sdb] Attached SCSI disk

it seems like now the 2 raid systems are read as sda1 (containing /dev/sda) and sda2 (containing /dev/sdb) and sdb is ignored.

Keruskerfuerst · 02-08-2016, 09:30 AM

Then try to use testdisk.

MightyU · 02-08-2016, 09:35 AM

More details:

Code:

partx -l /dev/sda
# 1:      2048-  2054143 (  2052096 sectors,   1050 MB)
# 2:   2054144-1755152383 (1753098240 sectors, 897586 MB)
partx -l /dev/sdb
partx: /dev/sdb: failed to read partition table

Code:

Filesystem                1K-blocks       Used   Available Use Mounted on
/dev/mapper/centos-root    81880000    6619196    75260804   9 /
devtmpfs                   32863676          0    32863676   0 /dev
tmpfs                      32879540          0    32879540   0% /dev/shm
tmpfs                      32879540      17660    32861880   1% /run
tmpfs                      32879540          0    32879540   0% /sys/fs/cgroup
/dev/sda1                    993528     291420      634424  32% /boot
/dev/dm-6               19454492920 7017457584 11460401584  38% /scrap
/dev/dm-7                  24056928     118052    22693796   1% /home
/dev/dm-9                 695977956   33686768   662291188   5% /var
/dev/dm-8                  24557860      33020    24524840   1% /tmp

/scrap should be assigned to sdb but it is not: it is assigned to /dev/dm-6 .

Code:

uname -a : Linux chef 3.10.0-327.4.5.el7.x86_64 #1 SMP Mon Jan 25 22:07:14 UTC 2016 x86_64 x86_64 x86_64 GNU/Linux

I guess that there is a conflict between modules in the kernel?

Code:

Disk /dev/sdb - 20 TB / 18 TiB - CHS 2431665 255 63
Current partition structure:
     Partition                  Start        End    Size in sectors

   P Linux LUKS               0   0  1 2431665  10 25 39064698880

MightyU · 02-08-2016, 09:45 AM

Quote:

Originally Posted by Keruskerfuerst

Then try to use testdisk.

I will run the analysis now, it is going to take a while.

Keruskerfuerst · 02-08-2016, 12:03 PM

/scrap should be assigned to sdb but it is not: it is assigned to /dev/dm-6 .

But this cant be true.

MightyU · 02-09-2016, 02:27 AM

Yes, I agree. The system was OK until last kernel update, I just do not how understand what happened after I switched to 3.10.0-327.4.5.el7. The disk analysis is still at 68%, when this ends I will load an older kernel and see if the problem persists. And file a report to centOS in case.

MightyU · 02-10-2016, 08:45 AM

disk analysis does not reveal errors in the disks

Further details:

Code:

# lsblk -a
sda                                                 8:0    0 837.8G  0 disk  
├─sda1                                              8:1    0  1002M  0 part  /boot
└─sda2                                              8:2    0   836G  0 part  
  └─luks-CODE1111                                 253:0    0   836G  0 crypt 
    ├─centos-swap                                 253:1    0  46.9G  0 lvm   [SWAP]
    ├─centos-root                                 253:2    0  78.1G  0 lvm   /
    ├─centos-01                                   253:3    0 664.1G  0 lvm   
    │ └─luks-CODE2222                             253:8    0 664.1G  0 crypt /var
    ├─centos-00                                   253:4    0  23.4G  0 lvm   
    │ └─luks-CODE3333                             253:7    0  23.4G  0 crypt /tmp
    └─centos-home                                 253:5    0  23.4G  0 lvm   
      └─luks-CODE4444                             253:9    0  23.4G  0 crypt /home
sdb                                                 8:16   0  18.2T  0 disk  
└─CODE5555                                        253:6    0  18.2T  0 crypt

I see that in "/sys/block/dm-6/"
the name is defined as CODE5555(see sdb above)
a folder for sdb is given in /sys/block/dm-6/slaves

So I do not understand what happened. A remapping of /dev/sdb in dm-6 ? I am not an expert in storage, any help would be appreciated!

Keruskerfuerst · 02-10-2016, 09:00 AM

Where did you get the Image of Centos?

MightyU · 02-10-2016, 09:14 AM

If you mean the image of centos I used to set up the server, I just downloaded the everything iso.

Keruskerfuerst · 02-10-2016, 03:06 PM

You have to buy the distro.

MightyU · 02-11-2016, 04:50 AM

?
From https://wiki.centos.org/FrontPage

Quote:

What is CentOS Linux?
CentOS Linux is a community-supported distribution derived from sources freely provided to the public by Red Hat for Red Hat Enterprise Linux (RHEL). As such, CentOS Linux aims to be functionally compatible with RHEL. The CentOS Project mainly changes packages to remove upstream vendor branding and artwork. CentOS Linux is no-cost and free to redistribute. Each CentOS version is maintained for up to 10 years (by means of security updates -- the duration of the support interval by Red Hat has varied over time with respect to Sources released). A new CentOS version is released approximately every 2 years and each CentOS version is periodically updated (roughly every 6 months) to support newer hardware. This results in a secure, low-maintenance, reliable, predictable and reproducible Linux environment.