Host Disconnecting and Hangs

[MauroBonder]

I have 06 Dell servers with VMware ESX4.0 Update01 in a cluster, but these hosts had a problem. When trying to do Storage VMotion of a virtual machine that was allocated in this host was disconnected from the Host vCenter, and crashed.
He was responding to Ping, but nothing on the local console could do. Anyone know of any problem with Celerra Hardware and Perc/6i with this version of ESX?

Local Disk Space - OK
Storage space in - OK
VMware HCL - OK
Firmwares - OK

I even look at other logs kernel, console, etc but have not found any relevant information.

Here is a part of the log which shows disk errors.

Thank you.

Apr 19 15:55:46 HOSTNAME.DOMAIN kernel: [   56.984181] sdbj: Write Protect is off

Apr 19 15:55:46 HOSTNAME.DOMAIN kernel: [   56.984593] SCSI device sdbj: drive cache: write through

Apr 19 15:55:46 HOSTNAME.DOMAIN kernel: [   56.984596]  sdbj: sdbj1

Apr 19 15:55:46 HOSTNAME.DOMAIN kernel: [   57.056183] sd 6:0:62:0: Attached scsi disk sdbj

Apr 19 15:55:46 HOSTNAME.DOMAIN kernel: [   57.056325]   Vendor: DGC       Model: RAID 10           Rev: 0226

Apr 19 15:55:46 HOSTNAME.DOMAIN kernel: [   57.056329]   Type:   Direct-Access                      ANSI SCSI revision: 04

Apr 19 15:55:46 HOSTNAME.DOMAIN kernel: [   57.057226] SCSI device sdbk: 62914560 512-byte hdwr sectors (32212 MB)

Apr 19 15:55:46 HOSTNAME.DOMAIN kernel: [   57.057523] sdbk: Write Protect is off

Apr 19 15:55:46 HOSTNAME.DOMAIN kernel: [   57.058330] SCSI device sdbk: drive cache: write through

Apr 19 15:55:46 HOSTNAME.DOMAIN kernel: [   57.059114] SCSI device sdbk: 62914560 512-byte hdwr sectors (32212 MB)

Apr 19 15:55:46 HOSTNAME.DOMAIN kernel: [   57.059416] sdbk: Write Protect is off

Apr 19 15:55:46 HOSTNAME.DOMAIN kernel: [   57.060219] SCSI device sdbk: drive cache: write through

Apr 19 15:55:46 HOSTNAME.DOMAIN kernel: [   57.060222]  sdbk: sdbk1

Apr 19 15:55:46 HOSTNAME.DOMAIN kernel: [   57.061316] sd 6:0:63:0: Attached scsi disk sdbk

Apr 19 15:55:46 HOSTNAME.DOMAIN kernel: [   57.061456]   Vendor: DGC       Model: RAID 10           Rev: 0226

Apr 19 15:55:46 HOSTNAME.DOMAIN kernel: [   57.061460]   Type:   Direct-Access                      ANSI SCSI revision: 04

Apr 19 15:55:46 HOSTNAME.DOMAIN kernel: [   57.062353] SCSI device sdbl: 62914560 512-byte hdwr sectors (32212 MB)

Apr 19 15:55:46 HOSTNAME.DOMAIN kernel: [   57.062855] sdbl: Write Protect is off

Apr 19 15:55:46 HOSTNAME.DOMAIN kernel: [   57.063558] SCSI device sdbl: drive cache: write through

Apr 19 15:55:46 HOSTNAME.DOMAIN kernel: [   57.064244] SCSI device sdbl: 62914560 512-byte hdwr sectors (32212 MB)

Apr 19 15:55:46 HOSTNAME.DOMAIN kernel: [   57.064553] sdbl: Write Protect is off

Apr 19 15:55:46 HOSTNAME.DOMAIN kernel: [   57.065551] SCSI device sdbl: drive cache: write through

Apr 19 15:55:46 HOSTNAME.DOMAIN kernel: [   57.065554]  sdbl: sdbl1

Apr 19 15:55:46 HOSTNAME.DOMAIN kernel: [   57.067448] sd 6:0:64:0: Attached scsi disk sdbl

Apr 19 15:55:46 HOSTNAME.DOMAIN kernel: [   57.067586]   Vendor: DGC       Model: RAID 5            Rev: 0226

Apr 19 15:55:46 HOSTNAME.DOMAIN kernel: [   57.067590]   Type:   Direct-Access                      ANSI SCSI revision: 04

[AndreTheGiant]

Have you tried to make only a copy inside the local storage or inside the Celerra Datastore to see if the problem is only on one type of storage?

Celerra seems connected with HBA and FC, correct? If yes HBA firmware is fine?

PS: have you open a support request to Dell?

Andre

[MauroBonder]

I have not had access to firmware level applied to the HBA, so I could not validate that this issue is known or not.

Yes, there is an open call in parallel at Dell....

Thank you André

[AndreTheGiant]

You can find firmware level during boot (in the POST phase).

Which kind of multipath policy do you have? I suggest to start from MRU.

Andre

[MauroBonder]

storage not compatible with vsphere. easy!