EVA Storage
cancel
Showing results for 
Search instead for 
Did you mean: 

MSA 2212fc with direct attach - link down

 
Aleksandrs Frolovs
Occasional Visitor

MSA 2212fc with direct attach - link down

Hello!
I have:
1. MSA2212fc Dual Enhanced Controller Array with:
a. latest firmware: J200P39-01 (29 Jul 2009);
b. Internal Host Port Interconnect: Interconnected;
c. Fibre Channel Topology: Loop;
2. Two HP DL360 G5 servers with:
a. latest firmware, drivers and software from HP;
b. two Emulex HBA in each server;
c. Windows Server 2008 x64 with latest updates (two node cluster is created);
3. Emulex HBA:
a. Name: FC2142SR
b. Vendor: Emulex
c. BUS: PCIe
d. Speed: 4Gb
e. HP Part Number: A8002A
f. Vendor part Number: LPe1150
g. Firmware Version: 2.72a2
h. Boot Image: 5.03a0
i. Driver Type: Storport
j. Driver Version: 7.2.20.006
4. Two HP DL360 G5 serverâ s HBA are directly connected to MSA 2212fc and configured according to the â HP StorageWorks 2000 Family Modular Smart Array reference guideâ (FC Direct Attach Configuration section):
a. Server 1 HBAs are attached to:
i. Controller A Port 0
ii. Controller B Port 0
b. Server 2 HBAs are attached to:
i. Controller A Port 1
ii. Controller B Port 1
5. Microsoft Windows 2008 MPIO is installed and configured using â HP StorageWorks 2000 Modular Smart Array Software Support/Documentation CD 2.15 (31 Jul 2009))â . Drives are appearing in Device Manager as â HP MSA2212fc Multi-Path Disk Deviceâ .
6. Emulex HBAnyware Fibre Channel Remote Management application. (4.1a35, 17 Aug 2009) is downloaded from HP web and installed for HBA management.
7. Windows 2008 cluster with all services is working fine (all LUNs are accessible from both nodes).
Problem:
1. On both cluster nodes one of two of Emulex HBA shows â Link downâ .
2. From MSA 2212fc Web management GUI I see:
a. Controller Module A:
i. Port 0:
1. SFP Detect: SFP Present
2. Receive Signal: Present
3. Link Status: Link Down
4. Signal Detect: No Signal
5. Topology: Loop*
6. Speed: 4 Gbits/sec
7. FC Address: Unavailable
ii. Port 1:
1. SFP Detect: SFP Present
2. Receive Signal: Present
3. Link Status: Link Up
4. Signal Detect: Signal Detected
5. Topology: Private Loop
6. Speed: 4 Gbits/sec
7. FC Address: 0x0000E4
8. Loop ID: 2
b. Controller Module B:
i. Port 0:
1. SFP Detect: SFP Present
2. Receive Signal: Present
3. Link Status: Link Up
4. Signal Detect: Signal Detected
5. Topology: Private Loop
6. Speed: 4 Gbits/sec
7. FC Address: 0x0000E2
8. Loop ID: 3
ii. Port 1:
1. SFP Detect: SFP Present
2. Receive Signal: Present
3. Link Status: Link Down
4. Signal Detect: No Signal
5. Topology: Loop*
6. Speed: 4 Gbits/sec
7. FC Address: Unavailable
3. What is interesting, when I start only one cluster node â both HBA are operating normally. But after starting the second cluster node â one of HBA goes down and shows â Link Downâ on each node. There is the following error in event log on both cluster nodes:
a. CLUSTERNODE1
A fibre channel HBA port has changed to a state of link down.

User Action
Check the FC cabling to this port. Reinsert the FC GBIC if it has been unplugged. Replace FC cabling or the FC GBIC if failure continues.

WBEM Indication Properties
AlertingElementFormat: 2 0x2 (CIMObjectPath)
AlertingManagedElement: "\\CLUSTERNODE1\ROOT\HPQ:HPFCHBA_FCPort.CreationClassName="HPFCHBA_FCPort",DeviceID="20000000C9800F3A:10000000C9800F3A",SystemCreationClassName="HPFCHBA_ComputerSystem",SystemName="HPQ:GUID:85408DBD-6A39-41CD-9E5A-357AEA130BBF:NODEWWN:20000000C9800F3A:PORTWWN:10000000C9800F3A""
AlertType: 5 0x5 (Device Alert)
Description: "A fibre channel HBA port has changed to a state of link down."
DevicePermanentName: "10000000C9800F3A"
EventCategory: 11 0xb (I/O Adapter)
EventID: "6"
EventTime: "20091016031105.414000-000"
HWManufacturer: "Emulex Corporation"
ImpactedDomain: 4 0x4 (System)
IndicationIdentifier: "{98484E3A-15C4-45CC-8601-CBB2664E727B}"
IndicationTime: "20091016031105.420000+180"
NetworkAddresses[0]: "X.X.X.X"
NetworkAddresses[1]: "X.X.X.X"
NetworkAddresses[2]: "X.X.X.X"
OSType: 77 0x4d (Microsoft Windows Server 2008 64-Bit)
OSVersion: "6.0.6002"
PerceivedSeverity: 5 0x5 (Major)
ProbableCause: 5 0x5 (Connection Establishment Error)
ProbableCauseDescription: "FC HBA Port State Changed to Link Down"
ProviderName: "HP Fibre Channel"
ProviderVersion: "2.4.0.0"
RecommendedActions[0]: "Check the FC cabling to this port. Reinsert the FC GBIC if it has been unplugged. Replace FC cabling or the FC GBIC if failure continues."
Summary: "FC HBA port in link down state"
SystemCreationClassName: "HP_WinComputerSystem"
SystemFirmwareVersion[0]: "2009.07.10"
SystemFirmwareVersion[1]: "2008.11.02"
SystemGUID: "30303734-3436-4247-3837-31324133534D"
SystemModel: "ProLiant DL360 G5"
SystemName: "CLUSTERNODE1.rs.local"
SystemProductID: "470064-223"
SystemSerialNumber: "GB8712A3SM"
TIME_CREATED: 129001254657118090 0x1ca4df5276acf8a




b. CLUSTERNODE2
A fibre channel HBA port has changed to a state of link down.

User Action
Check the FC cabling to this port. Reinsert the FC GBIC if it has been unplugged. Replace FC cabling or the FC GBIC if failure continues.

WBEM Indication Properties
AlertingElementFormat: 2 0x2 (CIMObjectPath)
AlertingManagedElement: "\\CLUSTERNODE2\ROOT\HPQ:HPFCHBA_FCPort.CreationClassName="HPFCHBA_FCPort",DeviceID="20000000C97A8B68:10000000C97A8B68",SystemCreationClassName="HPFCHBA_ComputerSystem",SystemName="HPQ:GUID:85408DBD-6A39-41CD-9E5A-357AEA130BBF:NODEWWN:20000000C97A8B68:PORTWWN:10000000C97A8B68""
AlertType: 5 0x5 (Device Alert)
Description: "A fibre channel HBA port has changed to a state of link down."
DevicePermanentName: "10000000C97A8B68"
EventCategory: 11 0xb (I/O Adapter)
EventID: "6"
EventTime: "20091016031334.240000-000"
HWManufacturer: "Emulex Corporation"
ImpactedDomain: 4 0x4 (System)
IndicationIdentifier: "{9899F553-A482-49B5-A425-221455F7040A}"
IndicationTime: "20091016031334.255000+180"
NetworkAddresses[0]: "X.X.X.X"
NetworkAddresses[1]: "fe80::cc2d:a883:9bcd:35ea%13"
NetworkAddresses[2]: "X.X.X.X"
NetworkAddresses[3]: "X.X.X.X"
OSType: 77 0x4d (Microsoft Windows Server 2008 64-Bit)
OSVersion: "6.0.6002"
PerceivedSeverity: 5 0x5 (Major)
ProbableCause: 5 0x5 (Connection Establishment Error)
ProbableCauseDescription: "FC HBA Port State Changed to Link Down"
ProviderName: "HP Fibre Channel"
ProviderVersion: "2.4.0.0"
RecommendedActions[0]: "Check the FC cabling to this port. Reinsert the FC GBIC if it has been unplugged. Replace FC cabling or the FC GBIC if failure continues."
Summary: "FC HBA port in link down state"
SystemCreationClassName: "HP_WinComputerSystem"
SystemFirmwareVersion[0]: "2009.07.10"
SystemFirmwareVersion[1]: "2008.11.02"
SystemGUID: "30303734-3436-4247-3837-333553445034"
SystemModel: "ProLiant DL360 G5"
SystemName: "CLUSTERNODE2.rs.local"
SystemProductID: "470064-462"
SystemSerialNumber: "GB8735SDP4"
TIME_CREATED: 129001256278902448 0x1ca4df5881546b0


Where is a problem?
Should something to be changes in HBA driver configuration?
Is it expected behavior?

P.S.> See attached document with formatted test.

Aleksandrs
4 REPLIES 4
Uwe Zessin
Honored Contributor

Re: MSA 2212fc with direct attach - link down

Maybe I have missed it in your description, but did you configure the server's FC adapters/their drivers to use loop topology, too?
.
Aleksandrs Frolovs
Occasional Visitor

Re: MSA 2212fc with direct attach - link down

All driver configuration is default.

Emulex driver default configuration: "Topology: 2=Loop then F_Port".

Active HBA (on in each node) shows "Port Type: Private loop".

Aleksandrs
Lawrance Lee_1
Frequent Advisor

Re: MSA 2212fc with direct attach - link down

Was there ever a resolution to this problem?

I would like to know if there was a setting or driver or firmware issue.....
MartinR01
Occasional Visitor

Re: MSA 2212fc with direct attach - link down

Was this issue fixed?

I'm experiencing an issue similar to yours...