Renaming emcpower pseudo-devices on two Linux cluster nodes

Question

Hi again, I need to 'rename' the emcpower* pseudo-devices, but I don't know how to do. I'm reading some how-to on poerlink site, but they're not for my current PowerPath version. I've installed this version: EMCpower.LINUX-5.1.0-194.rhel5.i386.rpm From my first node: powermt display dev=all Pseudo name=emcpowera CLARiiON ID=CK200062401xxx [Cluster Oracle DWH] Logical device ID=6006016040301900xxxxxxxxxxxxxxxx [DWH OCFS2] state=alive; policy=BasicFailover; priority=0; queued-IOs=0 Owner: default=SP B, current=SP A Array failover mode: 1 ============================================================================== ---------------- Host --------------- - Stor - -- I/O Path - -- Stats --- ### HW Path I/O Paths Interf. Mode State Q-IOs Errors ============================================================================== 0 qla2xxx sdc SP A1 unlic alive 0 0 0 qla2xxx sdf SP B1 unlic alive 0 0 1 qla2xxx sdi SP A0 active alive 0 0 1 qla2xxx sdl SP B0 active alive 0 0 Pseudo name=emcpowerb CLARiiON ID=CK200062401xxx [Cluster Oracle DWH] Logical device ID=6006016040301900xxxxxxxxxxxxxxxx [DWH DG_DATA] state=alive; policy=BasicFailover; priority=0; queued-IOs=0 Owner: default=SP B, current=SP A Array failover mode: 1 ============================================================================== ---------------- Host --------------- - Stor - -- I/O Path - -- Stats --- ### HW Path I/O Paths Interf. Mode State Q-IOs Errors ============================================================================== 0 qla2xxx sda SP A1 unlic alive 0 0 0 qla2xxx sdd SP B1 unlic alive 0 0 1 qla2xxx sdg SP A0 active alive 0 0 1 qla2xxx sdj SP B0 active alive 0 0 Pseudo name=emcpowerd CLARiiON ID=CK200062401xxx [Cluster Oracle DWH] Logical device ID=6006016040301900xxxxxxxxxxxxxxxx [DWH DG_FRA] state=alive; policy=BasicFailover; priority=0; queued-IOs=0 Owner: default=SP B, current=SP A Array failover mode: 1 ============================================================================== ---------------- Host --------------- - Stor - -- I/O Path - -- Stats --- ### HW Path I/O Paths Interf. Mode State Q-IOs Errors ============================================================================== 0 qla2xxx sdb SP A1 unlic alive 0 0 0 qla2xxx sde SP B1 unlic alive 0 0 1 qla2xxx sdh SP A0 active alive 0 0 1 qla2xxx sdk SP B0 active alive 0 0 From my second node: powermt display dev=all Pseudo name=emcpowerb CLARiiON ID=CK200062401xxx [Cluster Oracle DWH] Logical device ID=6006016040301900xxxxxxxxxxxxxxxx [DWH OCFS2] state=alive; policy=BasicFailover; priority=0; queued-IOs=0 Owner: default=SP B, current=SP A Array failover mode: 1 ============================================================================== ---------------- Host --------------- - Stor - -- I/O Path - -- Stats --- ### HW Path I/O Paths Interf. Mode State Q-IOs Errors ============================================================================== 0 qla2xxx sdc SP A1 unlic alive 0 0 0 qla2xxx sdf SP B1 unlic alive 0 0 1 qla2xxx sdi SP A0 active alive 0 0 1 qla2xxx sdl SP B0 active alive 0 0 Pseudo name=emcpowera CLARiiON ID=CK200062401xxx [Cluster Oracle DWH] Logical device ID=6006016040301900xxxxxxxxxxxxxxxx [DWH DG_DATA] state=alive; policy=BasicFailover; priority=0; queued-IOs=0 Owner: default=SP B, current=SP A Array failover mode: 1 ============================================================================== ---------------- Host --------------- - Stor - -- I/O Path - -- Stats --- ### HW Path I/O Paths Interf. Mode State Q-IOs Errors ============================================================================== 0 qla2xxx sda SP A1 unlic alive 0 0 0 qla2xxx sdd SP B1 unlic alive 0 0 1 qla2xxx sdg SP A0 active alive 0 0 1 qla2xxx sdj SP B0 active alive 0 0 Pseudo name=emcpowerc CLARiiON ID=CK200062401xxx [Cluster Oracle DWH] Logical device ID=6006016040301900xxxxxxxxxxxxxxxx [DWH DG_FRA] state=alive; policy=BasicFailover; priority=0; queued-IOs=0 Owner: default=SP B, current=SP A Array failover mode: 1 ============================================================================== ---------------- Host --------------- - Stor - -- I/O Path - -- Stats --- ### HW Path I/O Paths Interf. Mode State Q-IOs Errors ============================================================================== 0 qla2xxx sdb SP A1 unlic alive 0 0 0 qla2xxx sde SP B1 unlic alive 0 0 1 qla2xxx sdh SP A0 active alive 0 0 1 qla2xxx sdk SP B0 active alive 0 0 From my first node: emcpadm getused PowerPath pseudo device names in use: Pseudo Device Name Major# Minor# emcpowera 120 0 emcpowerb 120 16 emcpowerc 120 32 <--- no more used!!! emcpowerd 120 48 From my second node: emcpadm getused PowerPath pseudo device names in use: Pseudo Device Name Major# Minor# emcpowera 120 0 emcpowerb 120 16 emcpowerc 120 32 I would like both nodes have emcpowera as [sdc, sdf, sdi, sdl], emcpowerb as [sda, sdd, sdg, sdj], emcpowerd as [sdb, sde, sdh, sdk] How can I do that ? Thanks for your support, Marco Message was edited by: Stefano Del Corno removed possibly sensitive data

rawstorage · Accepted Answer

knowledgebase article emc113184 deals with this issue,

dynamox · Answer

if the letter is already taken ..can you rename to something other emcpower device ..that will free up the slot for you to maneuver around. Also i see that your policy is set to BasicFailover ..are you using the licensed version?

aco1 · Answer

Thanks for your reply.Currently these two machines are not licensed. I'm sure my manager has requested new license for these new machines.Is there a correlation between emcpower* pseudo-devices and an unlicensed PowerPath version ?Now I'm reading the man page of emcpadm and trying to use the rename options.

dynamox · Answer

no correlation at all ..just wanted to make sure that policy is set to ClarOpt if you do have license installed , once you do get the license, install it and set policy to ClarOpt to get the benefits of load-balancing/failover.

aco1 · Answer

I've read your suggested doc and it was useful just like the "man emcpadm" command.

Just for other readers. I solved using the following commands:
From node1:
emcpadm getused
and then
emcpadm rename -s emcpowera -t emcpowere
emcpadm rename -s emcpowerb -t emcpowerf
emcpadm rename -s emcpowerd -t emcpowerg

From node2:
emcpadm getused
and then
emcpadm rename -s emcpowerb -t emcpowere
emcpadm rename -s emcpowera -t emcpowerf
emcpadm rename -s emcpowerc -t emcpowerg

Thanks,
Marco

wzglc6 · Answer

Hi I try to rename /dev/emcpowerb, but it seems to be used.I know, it is not used. This happens on a linux 5.0 x64 system.Any help is appreciated.Stefan

rawstorage · Answer

Stefan,the command 'emcpadm getfreepseudos' will show you if this if free or not.This may be something SAC should take a look at if the device is not showing up there and is not showing in powermt display dev=all.

ibmpdc2 · Answer

So rename pseudos' does not work well on boot form san devices.  How to you rename the pseudo of a boot from san device because for the device in use errors not due to the pseudo name taken but that its an open boot disk on Linux?

Dell-DylanJ · Answer

Hello,

You might give this article a look to see if it helps you at all:

https://dell.to/3wzTEM8

However, if the disk is busy, and with this being a boot from SAN, I'd expect it is, you may need to consider taking a new disk and renaming it to suit your needs, then migrating the kernel over to it.

Please let me know if this helps.

PowerPath

Was this post helpful?