Another case of 'Device(s) not found'

Question

Fujitsu 2540 server with Emulex HBA's. Linux RHEL7 with PowerPath 6.4

Two luns attached, each with four paths. Both luns can be used directly, I created a partition and created an xfs filesystem, could mount /dev/sdb1 and such with no issues at all.

But all powermt commands come back with "Device(s) not found", including powermt check and powermt config.

Any suggestions?

DELL-Sam L · Answer

Hello Karel_A_J, Which emc storage system do you have as you didn’t state it in your post?

Karel_A_J · Answer

Apologies for not mentioning that. The trouble is that I don't really know, those cabinets are managed by another team. It is certainly EMC, I think it must be VNX. Below is some output from another server (which works fine), and that I believe to be connected to the same storage. I hope this helps?

# powermt display dev=all
Pseudo name=emcpowera
VPLEX ID=CKM00170502039
Logical device ID=6000144000000010E00EA1C07BEFC835 [device_TUJL_CSTORDBC_S004_1_vol]
state=alive; policy=ADaptive; queued-IOs=0; protocol=SCSI; size=3.00G

Karel_A_J · Answer

Both links fail for me, with a message 'permission required'. Sigh

DELL-Sam L · Answer

Here is the second link.

powermt display dev=all "Device(s) not found"

Audience Level: Customer

Article Content

Symptoms

powermt display dev=all
Device(s) not found
Host sees devices at the OS level:
Solaris = " format"
HPUX = " ioscan -fnC disk"
AIX = " lsdev -Cc disk"
Linux = " /proc/scsi/scsi"
Windows = " device manager"
.....or simply an " INQ"

Newly assigned devices not visible in powermt display dev=all

powermt display unmanaged gives the following output:

root@ukmsrv051 # powermt display unmanaged
CLARiiON ID=CK200072800447
Logical device ID=6006016094111E000A4454F6A8EEDC11
Vendor ID=DGC ; Product ID=RAID 5
===============================================
-------------------- Host ---------------------
HW Path I/O Paths
===============================================
ssm@0,0/pci@18,600000/QLGC,qla@1/sd@1,5 c7t1d5s0
ssm@0,0/pci@18,600000/QLGC,qla@1/sd@2,5 c7t2d5s0

Resolution

Make sure PowerPath array status is set for "managed" either by checking the " powermt display options" file in the emcgrab or by running the same "powermt display options" command at the host prompt.

If the specified array is set for "unmanaged":

powermt manage class=symm (or class=clariion, etc.)
powermt display dev=all
powermt save.

Additional Information

Starting with PowerPath 5.0, arrays can be managed or unmanaged. The intention of this solution is to help make sure that the easiest and sometimes most overlooked issue is checked first, which is an "unmanaged class" situation. This can be determined very quickly.
There are other OS variants which have their own methods to determine device visibility. This solution covers the common platforms.

Karel_A_J · Answer

Hm, does that mean that the powermt output that I contributed was not (or not sufficiently) helpful? What can I do to provide helpful information?

DELL-Sam L · Answer

Hello Karel_A_J, Here are the links to a couple of kb’s that maybe of assistance. https://dell.to/3NpzuN7 https://dell.to/3GU1oyc

DELL-Sam L · Answer

Hello Karel_A_J,

Both links are customer facing and you should be able to access them if login with your EMC account.

Here is what the links state.

PowerPath for Linux: "devices not found"

Audience Level: Customer

Article Content

Symptoms

Error message: " Devices not found"

Cause

PowerPath is configured not to manage CLARiiON or Symmetrix devices.

Resolution

Use "powermt display options" to show which classes of array are flagged for PowerPath control. If your array shows up with an "unmanaged" status, use "powermt manage class= xxx" to bring that array under PowerPath, followed by "powermt config" and "powermt save".

Karel_A_J · Answer

However, the 'status = unmanaged' seems to be quite general, could it be due to an upgrade or reinstallation of powerpath software? # powermt display options | grep statusstatus = unmanagedstatus = unmanagedstatus = unmanagedstatus = unmanagedstatus = unmanagedstatus = unmanagedstatus = unmanagedstatus = unmanagedstatus = unmanagedstatus = unmanagedstatus = unmanagedstatus = unmanaged

DELL-Sam L · Answer

Hello Karel_A_J, It is best to open a support case for this as we will need to review some logs and some other things to see what is going on.

Karel_A_J · Answer

# powermt display options | grep -A10 Invista
Invista periodic autorestore = on
reactive autorestore = on
status = managed
proximity based autostandby = N/A
path retry timeout = 5
all paths dead retry timeout = 45
auto host registration = N/A
app finger printing = N/A
device to array performance report = N/A
device in use to array report = N/A

Karel_A_J · Answer

# powermt display unmanagedNo unmanaged devices found.

Karel_A_J · Answer

Yes, no surprise there, I will see into it with my team mate who handles the EMC support contract. This is not an easy one, apparently Thanks for rapid follow-up!

Karel_A_J · Answer

cat /proc/scsi/scsiAttached devices:Host: scsi0 Channel: 02 Id: 00 Lun: 00Vendor: FTS Model: PRAID EP400i Rev: 4.68Type: Direct-Access ANSI SCSI revision: 05Host: scsi1 Channel: 00 Id: 00 Lun: 00Vendor: EMC Model: Invista Rev: 6200Type: Direct-Access ANSI SCSI revision: 04Host: scsi1 Channel: 00 Id: 00 Lun: 01Vendor: EMC Model: Invista Rev: 6200Type: Direct-Access ANSI SCSI revision: 04Host: scsi1 Channel: 00 Id: 01 Lun: 00Vendor: EMC Model: Invista Rev: 6200Type: Direct-Access ANSI SCSI revision: 04Host: scsi1 Channel: 00 Id: 01 Lun: 01Vendor: EMC Model: Invista Rev: 6200Type: Direct-Access ANSI SCSI revision: 04Host: scsi2 Channel: 00 Id: 00 Lun: 00Vendor: HP Model: Ultrium 6-SCSI Rev: 258SType: Sequential-Access ANSI SCSI revision: 06Host: scsi2 Channel: 00 Id: 00 Lun: 01Vendor: STK Model: SL150 Rev: 0260Type: Medium Changer ANSI SCSI revision: 05Host: scsi2 Channel: 00 Id: 01 Lun: 00Vendor: HP Model: Ultrium 7-SCSI Rev: M571Type: Sequential-Access ANSI SCSI revision: 06Host: scsi2 Channel: 00 Id: 01 Lun: 01Vendor: HP Model: MSL6480 Rev: 5.20Type: Medium Changer ANSI SCSI revision: 05Host: scsi3 Channel: 00 Id: 00 Lun: 00Vendor: EMC Model: Invista Rev: 6200Type: Direct-Access ANSI SCSI revision: 04Host: scsi3 Channel: 00 Id: 00 Lun: 01Vendor: EMC Model: Invista Rev: 6200Type: Direct-Access ANSI SCSI revision: 04Host: scsi3 Channel: 00 Id: 01 Lun: 00Vendor: EMC Model: Invista Rev: 6200Type: Direct-Access ANSI SCSI revision: 04Host: scsi3 Channel: 00 Id: 01 Lun: 01Vendor: EMC Model: Invista Rev: 6200Type: Direct-Access ANSI SCSI revision: 04Host: scsi4 Channel: 00 Id: 00 Lun: 00Vendor: HP Model: Ultrium 6-SCSI Rev: 25FSType: Sequential-Access ANSI SCSI revision: 06Host: scsi4 Channel: 00 Id: 01 Lun: 00Vendor: HP Model: Ultrium 7-SCSI Rev: M571Type: Sequential-Access ANSI SCSI revision: 06

Karel_A_J · Answer

And still no good, after more thorough reading....

# powermt display options | grep -A10 Invista
Invista periodic autorestore = on
reactive autorestore = on
status = managed
proximity based autostandby = N/A
path retry timeout = 5
all paths dead retry timeout = 45
auto host registration = N/A
app finger printing = N/A
device to array performance report = N/A
device in use to array report = N/A

# powermt config
# powermt save
# powermt display
Device(s) not found.

PowerPath

Another case of "Device(s) not found"

powermt display dev=all "Device(s) not found"

Article Content

Symptoms

Resolution

Additional Information

PowerPath for Linux: "devices not found"

Article Content

Symptoms

Cause

Resolution

Was this post helpful?