2.0 Upgrade - Page 3

peter.robertson · ‎09-20-2011

Has anyone completed the upgrade to 2.0? And if so, how was your experience?

I have the firmware bundles downloaded and am preparing to run the upgrade but wanted to know if anyone has run into any issues. I have checked and do not have any FCoE VLAN overlap.

kg6itcraig · ‎09-30-2011

Maybe, see my post below.

Craig

My UCS Blog http://realworlducs.com

Robert Burns · ‎10-17-2011

Greetings All,

The UCS 2.0 patch 2.0(1q) has been released which addressed the two major bugs detailed above.

Release Notes:

http://www.cisco.com/en/US/docs/unified_computing/ucs/release/notes/OL_25363.html#wp232083

Software:

http://www.cisco.com/cisco/software/release.html?mdfid=283612660&flowid=22121&softwareid=283655658&release=2.0%281q%29&relind=AVAILABLE&rellifecycle=&reltype=latest

Highlights of major issues fixed:

•Using the UCSM GUI, you are now able to disassociate a service profile that is currently bound to a template. (CSCts95454)

•When you assign an org to a locale from the GUI, the operation sometimes fails due to an internal error. This error is now corrected. (CSCts60863)

•The PCI Device address of a VNIC will not change after an upgrade of UCS Manager from Release 1.x to Release 2.0(1q). (CSCts96949)

•When the DHCP server is using an option 67 (RFC 2132) to report the bootfile name to the gPXE client, gPXE will receive the boot parameters and the boot will function normally. (CSCts86689)

BIOS

•When the BIOS is upgraded on a B230-M1 blade from Release 1.x to Release 2.0, the PCI address is preserved. (CSCts86890)

•A B230-M1 blade discovered while running 1.4 BIOS release image and now running a UCS 2.0 release BIOS image will associate and disassociate normally. (CSCtj54470)

•A Blade with a service profile with a 22 character or longer name will boot as expected from the local disk after upgrading the BIOS from a 1.x release to the BIOS in the 2.0(1q) release. (CSCtt13313)

Regards,

Robert

kg6itcraig · ‎09-30-2011

Since the code update been trying to figure out why if we select “Enable Failover” for our MetaData vNIC template (updating), or corresponding vNIC’s in service profiles traffic will not flow between upstream switch and FI’s. I can try downing a port-channel and that does not fix issue. Have tried each. Have to select an FI and not enable failover for traffic to flow. This was not the case before update.

Upstream 3750 switch config for MetaData VLAN.

!

interface Port-channel8

description UCS trunk to FIA

switchport trunk encapsulation dot1q

switchport trunk allowed vlan 512

switchport mode trunk

!

interface Port-channel9

description UCS trunk to FIB

switchport trunk encapsulation dot1q

switchport trunk allowed vlan 512

switchport mode trunk

!

############################

Corresponding UCS CLI output below.

UCS02-A(nxos)# show cdp neighbors

Capability Codes: R - Router, T - Trans-Bridge, B - Source-Route-Bridge

S - Switch, H - Host, I - IGMP, r - Repeater,

V - VoIP-Phone, D - Remotely-Managed-Device,

s - Supports-STP-Dispute

Device-ID Local Intrfce Hldtme Capability Platform Port ID

UCS02-B(SSI135004JB) mgmt3 144 S I s N10-S6200 mgmt3

VLP127C7000NS002(JAF1429CKER)Eth3/3 164 R S I s N7K-C7010 Eth10/26

VLP127C7000NS002(JAF1429CKER)Eth3/4 164 R S I s N7K-C7010 Eth10/25

SAN127C3750NS001 Eth3/5 176 S I WS-C3750X-48 Ten3/1/1

SAN127C3750NS001 Eth3/6 176 S I WS-C3750X-48 Ten3/1/2

UCS02-B(nxos)# show cdp neighbors

Capability Codes: R - Router, T - Trans-Bridge, B - Source-Route-Bridge

S - Switch, H - Host, I - IGMP, r - Repeater,

V - VoIP-Phone, D - Remotely-Managed-Device,

s - Supports-STP-Dispute

Device-ID Local Intrfce Hldtme Capability Platform Port ID

UCS02-A(SSI15120GEM) mgmt3 166 S I s N10-S6200 mgmt3

VLP127C7000NS002(JAF1429CKER)Eth3/3 136 R S I s N7K-C7010 Eth10/28

VLP127C7000NS002(JAF1429CKER)Eth3/4 136 R S I s N7K-C7010 Eth10/27

SAN127C3750NS001 Eth3/5 146 S I WS-C3750X-48 Ten2/1/1

SAN127C3750NS001 Eth3/6 146 S I WS-C3750X-48 Ten2/1/2

#####################

UCS02-A /eth-uplink # show pin-group MetaData expand

Ethernet Pin Group:

Name: MetaData

Ethernet Pin Target:

Fabric Endpoint

------ --------

A fabric/lan/A/pc-2

B fabric/lan/B/pc-2

#####################

UCS02-A /eth-uplink # show interface

Member Port:

Fabric ID Port-channel Slot Port Oper State State Reason Lic State Grace Period

---------- ------------ ----- ----- ------------- ----------------------------------- -------------------- ------------

A 1 3 3 Up Unknown 0

A 1 3 4 Up Unknown 0

A 2 3 5 Up Unknown 0

A 2 3 6 Up Unknown 0

B 1 3 3 Up Unknown 0

B 1 3 4 Up Unknown 0

B 2 3 5 Up Unknown 0

B 2 3 6 Up Unknown 0

####################

UCS02-A# scope eth-uplink

UCS02-A /eth-uplink # scope fabric a

UCS02-A /eth-uplink/fabric # show port-channel detail

Port Channel:

Port Channel Id: 1

Name:

Admin State: Enabled

Oper State: Up

Speed: 10 Gbps

State Reason:

flow control policy: default

Port Channel Id: 2

Name:

Admin State: Enabled

Oper State: Up

Speed: 10 Gbps

State Reason:

flow control policy: default

UCS02-A /eth-uplink/fabric # show port-channel expand

Port Channel:

Port Channel Id: 1

Name:

Admin State: Enabled

Oper State: Up

State Reason:

Member Port:

Slot Id Port Id Membership Oper State State Reason Lic State Grace Period

------- ------- ---------- ---------- ------------- --------- ------------

3 3 Up Up Unknown 0

3 4 Up Up Unknown 0

Port Channel Id: 2

Name:

Admin State: Enabled

Oper State: Up

State Reason:

Member Port:

Slot Id Port Id Membership Oper State State Reason Lic State Grace Period

------- ------- ---------- ---------- ------------- --------- ------------

3 5 Up Up Unknown 0

3 6 Up Up Unknown 0

####

UCS02-A /eth-uplink/fabric # show port-channel detail

Port Channel:

Port Channel Id: 1

Name:

Admin State: Enabled

Oper State: Up

Speed: 10 Gbps

State Reason:

flow control policy: default

Port Channel Id: 2

Name:

Admin State: Enabled

Oper State: Up

Speed: 10 Gbps

State Reason:

flow control policy: default

UCS02-A /eth-uplink/fabric # show port-channel expand

Port Channel:

Port Channel Id: 1

Name:

Admin State: Enabled

Oper State: Up

State Reason:

Member Port:

Slot Id Port Id Membership Oper State State Reason Lic State Grace Period

------- ------- ---------- ---------- ------------- --------- ------------

3 3 Up Up Unknown 0

3 4 Up Up Unknown 0

Port Channel Id: 2

Name:

Admin State: Enabled

Oper State: Up

State Reason:

Member Port:

Slot Id Port Id Membership Oper State State Reason Lic State Grace Period

------- ------- ---------- ---------- ------------- --------- ------------

3 5 Up Up Unknown 0

3 6 Up Up Unknown 0

The VLAN for MetaData is 512

UCS02-A(nxos)# show vlan id 512

VLAN Name Status Ports

---- -------------------------------- --------- -------------------------------

512 VLAN0512 active Po1, Po2

Remote SPAN VLAN

----------------

Disabled

Primary Secondary Type Ports

------- --------- --------------- -------------------------------------------

UCS02-B(nxos)# show vlan id 512

VLAN Name Status Ports

---- -------------------------------- --------- -------------------------------

512 VLAN0512 active Po1, Po2

Remote SPAN VLAN

----------------

Disabled

Primary Secondary Type Ports

------- --------- --------------- -------------------------------------------

####

UCS02-A(nxos)# show vlan brief

VLAN Name Status Ports

---- -------------------------------- --------- -------------------------------

1 default active Po1, Po2, Eth1/21, Eth1/22

Eth1/23, Eth1/24, Eth1/25

Eth1/26, Eth1/27, Eth1/28

Eth1/29, Eth1/30, Eth1/31

Eth1/32, Eth1/33, Eth1/34

Eth1/35, Eth1/36, Eth1/37

Eth1/38, Eth1/39, Eth1/40

Eth3/1, Eth3/2, Eth1/1/1

Eth1/1/2, Eth1/1/3, Eth1/1/4

Eth1/1/5, Eth1/1/6, Eth1/1/7

Eth1/1/8, Eth2/1/1, Eth2/1/2

Eth2/1/3, Eth2/1/4, Eth2/1/5

Eth2/1/6, Eth2/1/7, Eth2/1/8

Eth3/1/1, Eth3/1/2, Eth3/1/3

Eth3/1/4, Eth3/1/5, Eth3/1/6

Eth3/1/7, Eth3/1/8, Eth4/1/1

Eth4/1/2, Eth4/1/3, Eth4/1/4

Eth4/1/5, Eth4/1/6, Eth4/1/7

Eth4/1/8, Eth5/1/1, Eth5/1/2

Eth5/1/3, Eth5/1/4, Eth5/1/5

Eth5/1/6, Eth5/1/7, Eth5/1/8

Eth6/1/1, Eth6/1/2, Eth6/1/3

Eth6/1/4, Eth6/1/5, Eth6/1/6

Eth6/1/7, Eth6/1/8, Eth7/1/1

Eth7/1/2, Eth7/1/3, Eth7/1/4

Eth7/1/5, Eth7/1/6, Eth7/1/7

Eth7/1/8, Eth8/1/1, Eth8/1/2

Eth8/1/3, Eth8/1/4, Eth8/1/5

Eth8/1/6, Eth8/1/7, Eth8/1/8

Eth9/1/1, Eth9/1/2, Eth9/1/3

Eth9/1/4, Eth9/1/5, Eth9/1/6

Eth9/1/7, Eth9/1/8, Eth10/1/1

Eth10/1/2, Eth10/1/3, Eth10/1/4

Eth10/1/5, Eth10/1/6, Eth10/1/7

Eth10/1/8

2 fcoe-vsan-2 active veth11560, veth11568, veth11576

veth11584, veth11592, veth11600

veth11608, veth11616, veth11624

veth11632, veth11640, veth11688

veth11690, veth11706, veth11708

veth11716, veth11718, veth11726

veth11728, veth11736, veth11738

veth11746, veth11748, veth11756

veth11758, veth11766, veth11768

veth11776, veth11784, veth11792

veth11800, veth11808, veth11816

veth11832, veth11840, veth11848

veth11856, veth11864, veth11872

veth11880, veth11888, veth11896

veth11904, veth11912, veth11920

veth11928, veth11936, veth11944

veth11952, veth11960, veth11968

veth11976, veth11984, veth11992

veth12000, veth12008, veth12016

veth12024, veth12032, veth12040

veth12048, veth12056, veth12064

veth12072, veth12080, veth12088

veth12096, veth12104, veth12112

veth12120, veth12128, veth12201

7 VLAN0007 active Po1, Po2

10 VLAN0010 active Po1, Po2

20 VLAN0020 active Po1, Po2

512 VLAN0512 active Po1, Po2

4044 SAM-vlan-management active Eth1/1/9, Eth2/1/9, Eth3/1/9

Eth4/1/9, Eth5/1/9, Eth6/1/9

Eth7/1/9, Eth8/1/9, Eth9/1/9

Eth10/1/9

4047 SAM-vlan-boot active

4048 VLAN4048 active Po1, Po2

Totally do not get why this is happening.

Craig

My UCS Blog http://realworlducs.com

v-crsimp · ‎10-17-2011

In the end it was a config error I didn't notice until the update. Had a VLAN not global, but only one one FI. Was nothing todo with the code update.

v-crsimp · ‎10-17-2011

OK let's "Man Up" try this again. I am pushing to the back of the line. Will see how this goes for ya'll first.

New Reliease Patch 2.0(1q).

http://www.cisco.com/cisco/software/navigator.html?mdfid=282558030&flowid=22121

Good Luck!

Craig

kg6itcraig · ‎02-21-2012

2/21/2012

After taking a break to let some code updates roll, ran the 2.0(1w) code update without error. Nice and smooth like usual.

My standard order for code updates.

1) CIMC

2) IOM (which will wait until the FI’s update)

3) UCSM

4) FI X (Which ever is the current passive FI)

5) FI X (Second FI)

Craig

My UCS Blog http://realworlducs.com

Robert Burns · ‎02-21-2012

Thanks for the feedback Craig (as always).

Don't forget to update the blade components (Adaptors, BIOS and Board Controllers) after the FIs are completed

Rob

Jing Ren · ‎11-12-2012

Hi Lorenz,

I think from the screenshots you've attached, they only showed the VLAN ID and VSAN ID. It should not be a problem if they are the same. Could you please click on the default VSAN and check FCoE ID is also 1? Otherwise it could match the description of CSCtr74091 at

http://www.cisco.com/web/software/datacenter/ucs/UCS_2_0_Software_Advisory.htm

lorenzobexer · ‎02-23-2012

Hi,

we have a UCS Infrastructure consisting of 2 FI (6120XP) and 2 chassis. We are currently on 1.4(3l) and want to upgrade to UCS 2.0(t).

I think we have a case of VLAN overlapping here, can someone confirm that by this screenshots:

We use B200-M2 Blades with ESXi running on them, and all VM disks + ESXi boot disks are located in the SAN.

So i wondered if it really is such a good idea to change the default VSAN and so disabling the disks of all VMs for a moment. We already had problems with Linux systems setting their filesystem to read-only after a short SAN failure.

Isn't it a better idea to just change the default Ethernet VLAN and to only risk a short network connectivity outage?

Also I'm a little worried by the experience of Craig back in September. Even after the VLAN overlapping was fixed, all of the blades crashed during the upgrade.

That's just something that can't happen to our production environment as it would take weeks to fix all the problems this would cause.

Has this issue been fixed or would it be better to go to the latest 1.4 firmware?

Thanks in advance and Regards

kg6itcraig · ‎02-24-2012

Yes you are stuck with the overlap issue.

Don't think you can remove/change VLAN 1. It is hardcoded I believe on Cisco.

You are stuck with changing the VSAN to two. UCSM will tell you if a change is going to reboot stuff and you can cancel. We changed for the code update and all servers were off. I don't remember what happened when we changed, sorry. Maybe try the simulator, or just call TAC. TAC has been beaten up quite a good deal over this and has the answers.

1.4m to 2.0(1w) went fine and didn't bother our Hyper-V servers. We don't do ESX. Was a very smooth update. None of the PCI Bus renumbering that caused issues first 2.0 code level appeared.

One bit of worry was blades show to be "Configuring" as each FI/IOM's reboot. But they are up.

Craig

My UCS Blog http://realworlducs.com

Robert Elmes · ‎02-25-2012

We recently upgrade a test/demo environment from 1.3(1c) to 2.0(1w). Normal update process followed, but when it came to the Fabrics the first one upgraded ok but the second fabric crashed and wouldn't upgrade.

Raised issue with TAC but we needed to get on with the upgrade, as we only had a short test window on this system, so in the end powered off/on the failed fabric. This resolved the issue and the fabric upgraded. TAC have now come back to us with information on the reason. New bug: CSCty19988.

kg6itcraig · ‎02-25-2012

Robert Elmes,

Is there anything unique about your environment that caused this? What do you have attached to your FI's? Are you boot from SAN, do you use SAN (and if so what switches are attached), do you have port-channels etc? If you don't mind me asking.

We have a few more UCS's to upgrade in the next few weeks. We try to load our test UCS up with as tricky a configuration as possbile, so it will run into issues the production ones might.

But still, once in a while...

Craig

My UCS Blog http://realworlducs.com