P2000 Host Port Cha Down Error Message

[Michael Grech]

I have 3 P2000 SAN's, fiber channel, connected to 2 HP SANSwitch 8/8 (brocade 300) switches, that connect to 6 GL370 (ESXi) hosts with QLogic Fiber Channel adapters.  Every few days one or two of the SAN's log the following error message:

Subject: P2000 G3 FC Warning: HOST_CHAN_LINK_DOWN

Event:

Host link down. (port: 2)

EVENT ID:#A1071

EVENT CODE:112

EVENT SEVERITY:Warning

EVENT TIME:2011-07-31 10:47:14

 System Name:MSA P2000A

System Location:CEC - RIE12

System Contact:

IP Address:

I check the log files and they indicate the port went down, then immediately came back up.  I sent the log files to HP's support desk and they said the SAN looks fine so it must be out side of it.  I assume one of the Brocade switches or one of the six hosts.  Is that assumption correct and does anyone have any troubleshooting suggestions?  Thank you.

P.S. This thread has been moevd from Disk to Storage Area Networks (SAN) (Small and Medium Business). - Hp forum moderator

[Johan Guldmyr]

Hi, yes it probably is something on the SAN causing it.

Did HP only look at the P2000 logs?

You should be able to see the link errors on the SAN-switches as well.

Some initial questions though:

Are you seeing this HOST_CHAN_LINK_DOWN on all the P2000 host ports and both of the SAN-switches?
Or can you limit it down into maybe one SAN-switch, one fabric, one host ports etc. 

Are the SAN switches separated into two fabrics?

[Sudhakar Subramaniam]

These messages will be genrated because of the improper host port settings and port speed set to Auto negotiate. MSA2000 controllers supports only 2 types of port topologies (Point-to-point and Loop). If the servers are connected to MSA directly the port topology should be Loop and it should Loop on the switch port and Server HBA port. If the servers are connected to MSA through a SAN switch, the port topology should be point-to-point and it should be same on the switch port and server HBA port. Port speed should not be on AN [Auto negotiate] on MSA, Switch or HBA

. Procedure to change the host port properties on MSA:

 For MSA2000 G1 GUI Manage , General Config , host port configurtaion , Change FC Host port topologies

 For MSA2000 G2 GUI Configuration , system configuration , Host port configuration .

Regards

Sudhakar

[Johan Guldmyr]

>> If the servers are connected to MSA directly the port topology should be Loop and it should Loop on the switch port and Server HBA port. 

[Michael Grech]

Our configuration is Point-to-Point.  The servers and P2000's are connected to a SAN Switch (Brocade 300).  Is the problem with one of the Hosts (servers) or with the FC card in that particular SAN?  In other words, looking at the error message how do I determine what hardware is the problem?

[Johan Guldmyr]

It's not that easy.

'

What you see is the link between one of the host ports on the P2000 to the SAN-switch going down.

This is most likely caused by either of the ends or the cable/SFP.

But it could also be caused by other devices on the SAN.

Easiest is usually to try with another SFP/cable that you know is working.

You can also look in port errors on the brocade errors.

Try getting 

porterrshow

wait a while

and grab another one

and compare, maybe there is a HBA somewhere causing issues.

[toba_2]

Spoiler

Hello,

I hade the same problem over 3 months...I  tried all the above combinations, but without success...

Solution of the problem:

1. Check "time clock" on the HP P2000 G3 andSAN Switch to be same as on Domain Controller or Active Directory server!!!!

This is solved problem for me!

Regards!

[hms-superuser]

Hello Mate,

I thank you for posting the HP MSA issue.

I have a similar issue with my HP MSA P2000 g3 using SAS cables connecting directly to my 2 x Proliant DL630p Gen 8 servers.

At least once or twice a week i get error code 354 [Host SAS has changed (Host port: 3, 0 out 4 PHYs are up, link speed Autonegotiated)].

NB: Have no option to edit the SAS host interface in my MSA P2000 g3.

Please help me figure out a way to sort this issue, it is affecting productivity in our organization.

Many Thanks once again,