- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
vmhba# different numbers on different hosts
I have ucs blades and they all have the same hardware config and same service profiles. but under storage adapters, I see the vmhba has different numbers on some hosts
some hosts show vmhba1, vmhba2
some hosts show vmhba3 and vmhba4
why is that?
each hosts only have 2 vhbas
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
ESX host enumerates devices based on the PCI order in which they are discovered. Particularly with UCS-Service profiles you can strictly govern the order and naming convention, this is not possible on other hardware platforms.
You can get around by modifying the esx.conf file.
Did you clone the service profile or carve out seperate ones for each server ? If its the latter , check the device order - remember that vHBA & vNIC are both carved out from the CNA cards, so order of both NIC and HBA are important.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
yes he profile was cloned from a template. but was unbinded and rebinded back to the template. I dont understand why the other hosts shows a different number since both of them are cloned from the same template
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Can you please post /etc/vmware/esx.conf file outputs from both hosts ?
I'm only interested in this section,
===
/vmkdevmgr/pci/m00008502/alias = "vmnic1"
/vmkdevmgr/pci/s00000005.00/alias = "vmhba3"
/vmkdevmgr/pci/s00000006.01/alias = "vmnic7"
/vmkdevmgr/pci/m00008501/alias = "vmnic0"
/vmkdevmgr/pci/s00000005.01/alias = "vmhba4"
/vmkdevmgr/pci/s00000006.00/alias = "vmnic6"
/vmkdevmgr/pci/m00008504/alias = "vmnic3"
/vmkdevmgr/pci/m00008503/alias = "vmnic2"
/vmkdevmgr/pci/s00000002.01/alias = "vmnic5"
/vmkdevmgr/pci/s00000002.00/alias = "vmnic4"
/vmkdevmgr/pci/s00000001.00/alias = "vmhba1"
/vmkdevmgr/pci/s00000001.01/alias = "vmhba2"
/vmkdevmgr/pci/p0000:00:1f.2/alias = "vmhba0"
===
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I see the same issue.. I received email alerts regarding loss of storage connectivity and I see that vmhba3 is the culprit, we have HP blades and only have 2 HBAs in place, most hosts show vmhba1 and 2, but a small # of hosts show vmHBA2 and 3..
I don't expect a specific resolution to my environment but wouldn't mind seeing some expertise and advisement from the community on this scenario from a high level.