I just finished installed/configured VMware Hypervisor 5.1.0 on HP Proliant DL380 G7 server. Below is the error I got when connecting to the host via vSphere Client:
In the hypervisor Network Adapters, the static IP address that I have is linked to vmnic0 showing below.
Please let me know if this is incorrect.
And in the DNS Configuration, I set the Primary and Alternate DNS Server to match ours. Then the Hostname is changed to CompanyHost.
Again, please let me know if my configuration is incorrect. If it's all correct, I wonder why connecting to this host is not working. Any suggestion is much appreciated.
1. As long as one port is connected you are fine. Anyway, I'd suggest you verify the connections to ensure proper cabling. The network ports on the DL380G7 are counting from right to left! (see HP Product Bulletin)
2.-3. DNS resolution is not critical at all for a stand alone host, but should be in place in case you are going to add the host to a vCenter Server environment.
André
Are you sure vmnic0 is plugged into the right switch/port?
Can you ping out from the host with the test management network option in the DCUI?
Many thanks for the response. They all failed. Okay, I have a question regarding the HOST Name, is this something I just create or do I have to have my server IT team create a sub domain and point it to the address of my vSphere host? For example, our public domain is myCompany.com and my host name will be myVMHost.myCompany.com. Is this sub domain must be created in the DNS server and pointed to the vSphere host IP address?
ICMP traffic (pinging) is on a completely different OSI layer than DNS (name resolution) as long as you try to connect to your host via IP DNS related problems won't manifest.
If you can connect via IP but not via host name then you can start thinking about a DNS problem, but it is best practice to have correct forward and reverse lookup configuration in your DNS server for the host. So myVMHost.myCompany.com should resolve to the correct IP in an ideal world.
For your case start with the standard troubleshooting.
Step 1 - Already done, you cannot reach the outside or your standard gateway via ping.
Step 2 - Double check the NICs and cables (do you see lights on the NIC ports and switch ports)
Step 3 - Double check network settings configuration like subnet mask and VLANs
Step 4 - Look for incoming packets on the switch (if it is a managed switch), are those dropped, are any packets received at all
Step 5 - Try plugging a laptop into the switch and configure the ports accordingly, can you ping the host now, can you ping the laptop from the host
Tried all those steps. My only other concern right now is, perhaps the guy who suppose to assign a static IP to this VM NIC did not do it correctly; therefore, I cannot ping that IP address. The next step is to check with him. Idealy, I would need our DNS server to resolve my myVMHost.myCompany.com but that is no require, correct? The host name can be anything but, does it has to be resolved by the DNS server?
Get correct IP address, Subnet mask and gateway IP, configure with management NIC, need to confirm with remote technicians for connectivity, if server have multiple NIC's be sure on which NIC is connected which network segment, after configuring console IP test management network, after that ping IP from any system, try to access server using browser https://x.x.x.x/. Thanks..
I just got a verification from our server team and yes, the IP address linked correctly to the VM NIC. However, it's still not working. So, what's next? The gateway, subnet mask are all correct. I can't access the host via browser nor ping it.
Try to ping the default gateway from the ESXi host's console or using "Test Management Network" from the GUI to see whether this works.
André
That failed. I can ping the default gateway from my desktop but when tried it on the TMN, it failed.
Well, in this case start investigating on the physical side, i.e. switch port settings like VLANs (802.1Q), port security, ...
André
Are you referring to the physical port on the back of the server and switch? I've checked. Everything seems to be fine.
Despite our server team said they've verified the IP address is link to the vm nic, is there a way to test this by issuing a local/private IP address to this and do some test pining this way? If so, what local IP address I should use and gateway?
Okay, something new, I went into the Network Adapters, and see that vmnic0 status is "Disconnected (...)". So, why? How do I get it to be connected?
Disconnected could mean either a network cable issue (defective/wrong cable, connected to a wrong switch port), a disabled switch port, or (less likely in this case) a driver issue on the ESXi side.
André
... can you confirm the correct physical network adapter (vmnic) is selected/connected? Take a look at the DCUI to see whether any of the network adapters shows up as connected.
André
Okay, so the HP Proliant DL380 G7 with two multipurpose NIC shows with four NIC in the Network Adapters. The first two NIC shows status as "Disconnected" and the bottom two shows up as connected. I toggled and select all four for fault-tolerence and now in TMN, pining the gateway works but DNS failed. Any idea?
vmnic2 and vmnic3 shows up as connected but 1 and 2 is not.
Now that you can ping the gateway, check whether you are able to reach the host using the vSphere Client. Then check the DNS settings in the GUI (DNS servers). Does ping or nslookup return the proper IP address?
I sometimes saw that a host refused to resolve DNS names, when it wasn't able to contact the DNS server when it was powered on. If all the settings are ok try rebooting the host to see whether this solves the issue.
André
Okay, I'm able to connect to the host via vSphere Client; however, here are a couple of things I like to know.
1. Does the DNS resolution work from your client? Did you add a DNS record (Host-A entry) in your DNS server?
2. Can't tell you what's wrong. I'd try to switch cables to see whether this changes things.
3. Depends on your networking concepts (e.g. iSCSI, vMotion, ...) and how many (additional) NICs you have.
André
Okay, this is interesting. I unplugged the ethernet cable from the physical NIC 1 and the vmnic4 got disconnected. I unplugged ethernet cable from the physical NIC2 and vmnic3 got disconnected. Apparently, vmnic0 and vmnic1 does not get affected but this. HP says since their two NIC are multipurpose port, it shows up as four but apparently, 1 and 2 is not working. However, pining the DNS still failed but I'm able to connect via vSphere Client now -- one step closer.