elihuj's Posts

One of the older labs I built had external storage by way of Openfiler. I presented storage from Openfiler to my ESXi host via iSCSI. This was a single ESXi host and a separate desktop for storag... See more...
One of the older labs I built had external storage by way of Openfiler. I presented storage from Openfiler to my ESXi host via iSCSI. This was a single ESXi host and a separate desktop for storage. In order to consolidate (+ do more stuff) I took my ESXi host and created a nested lab. This allowed me to have nested ESXi hosts within my physical host, as well as install Openfiler as a VM to present storage to the nested hosts. I installed my nested hosts on SSD, and used my regular 2TB HDD to present storage to the hosts via Openfiler. This allowed me to do much more within my lab. Just some ideas.
Hi AWelcomer‌, what is your HA Isolation Response set to in your cluster settings?
Are you using a router with the routes defined connected to the host?
That was my question as well. What happens with the hosts/VM's during the APD situation?
We have a 2008 MSCS cluster setup with Physical mode RDM's within ESXi 5.0. I was looking over the whitepaper I used to setup the cluster, and as vGuy‌ says, it does state that Migration with vMo... See more...
We have a 2008 MSCS cluster setup with Physical mode RDM's within ESXi 5.0. I was looking over the whitepaper I used to setup the cluster, and as vGuy‌ says, it does state that Migration with vMotion of clustered virtual machines is a setup limitation of the MSCS configuration. Have a look at this table that shows the different clustering solutions.
Also remember that the host will disconnect/reconnect from vCenter. There is no cause for concern on this.
This was a concern that we faced in the past as well. One of our biggest issues was keeping up with VM Tools. We would have VM's on hosts several build levels behind. Rest assured though, there i... See more...
This was a concern that we faced in the past as well. One of our biggest issues was keeping up with VM Tools. We would have VM's on hosts several build levels behind. Rest assured though, there is no issue in running your VM's on hosts with different build numbers or different tools versions.
Promiscuous Mode! Thank you grasshopper‌, that was it. Enabled it at the port group level, and pings started working.
Where was it asking you to reboot? You may want to check the vmkernel.log log file for any clues.
EDIT: Pings on VM2 are working now. I thought I set the default gateway correctly, but when I checked again it was still using the 192.x address. I changed it to the correct Vyatta IP (10.10.10.2... See more...
EDIT: Pings on VM2 are working now. I thought I set the default gateway correctly, but when I checked again it was still using the 192.x address. I changed it to the correct Vyatta IP (10.10.10.252) and pings immediately started. Good stuff. I'm still wondering why a physical interface is needed. If the Vyatta VM is configured with two vNIC's; one with a physical interface and the other without one, shouldn't it still be able to route packets on the interface without a physical NIC?
Well.. some progress. VM2 is an Openfiler server with two interfaces. vNIC1 is on the 192.168.10.x subnet, and vNIC2 is on the 10.10.10.x subnet. After adding a route to the 10 network on my work... See more...
Well.. some progress. VM2 is an Openfiler server with two interfaces. vNIC1 is on the 192.168.10.x subnet, and vNIC2 is on the 10.10.10.x subnet. After adding a route to the 10 network on my workstation, I can ping the 10 address on VM1 (Vyatta).. but as I said, I still cannot ping the 10 address on VM2. So for ha ha's, I decided to use another VM to test. I configured VM3 with a 10.10.10.x address and put it in the same port group as the others... no joy. Then I thought about what abhilashhb‌ said in regards to no physical interfaces, so I added a physical NIC to the port group. Once I did that, pings from my workstation to VM3 worked! I'm still puzzled as to why VM2 still doesn't ping from my workstation. I'm thinking maybe a firewall rule or something on the VM. More troubleshooting!
Hey grasshopper.. hoping to make it a fun lab. I thought having some additional VLANs to play with might make things a little more exciting. A lot of the how-to's I've read on it though do use L3... See more...
Hey grasshopper.. hoping to make it a fun lab. I thought having some additional VLANs to play with might make things a little more exciting. A lot of the how-to's I've read on it though do use L3 switches and such. I was hoping to go with the poor man's approach! This is the community fork of VyOS. And yes, I have both VM's using the eth0 (192.168.10.200) as their default gateway. Still no joy.
This is where I was thinking that Vyatta would take over... With the Vyatta VM, eth1 has a direct connection to the 10.x network. So when I ping 10.10.10.200 (Vyatta) from my workstation it wo... See more...
This is where I was thinking that Vyatta would take over... With the Vyatta VM, eth1 has a direct connection to the 10.x network. So when I ping 10.10.10.200 (Vyatta) from my workstation it works (via my static workstation route). When I ping VM2 (10.10.10.202) from my workstation, wouldn't it go to the Vyatta and be routed accordingly? Despite there being no physical uplink, wouldn't it use the connected interface (192.x) and route it? Apparently not!
I have a physical host running ESXi 5.5 with 2 NIC's connected to an unmanaged switch. I'd like to setup some additional VLANs with Vyatta to simulate different scenarios. Here is my current setu... See more...
I have a physical host running ESXi 5.5 with 2 NIC's connected to an unmanaged switch. I'd like to setup some additional VLANs with Vyatta to simulate different scenarios. Here is my current setup: Port group 1: 1 NIC, Management traffic only (192.168.10.0/24 subnet) Port group 2: 1 NIC, VM traffic (192.168.10.0/24 subnet) Port group 3: 0 NIC's, Used for Vyatta I have two VM's in port group 3. VM1 has a vNIC in port group 2, and a vNIC in port group 3. VM2 has a vNIC in port group 3. Here is the IP configuration: VM1: 192.168.10.200, 10.10.10.200 VM2: 10.10.10.202 VM1 is my Vyatta VM. In this configuration, I am successfully able to ping between the 2 VM's on the 10.x network. I am unable to ping from my workstation to VM2, however. I added a static route on my workstation to the 10.10.10.0/24 network via 192.168.10.200. Afterwards, I am able to ping the Vyatta VM on the 10.x network. I cannot ping VM2 on the 10.x network though. I feel as though I'm missing something.. probably something simple.
Looks good LucD. I removed the "where" statement, and I can output all the path statuses. Thank you again for your help!
Okay.. as I'm looking over this again the script will output ONLY dead paths correct? If there are no dead paths, it will be blank right?
Thank you for the reply LucD. How can I add in the syntax to pull the host information from my text file? Get-VMHost | % {      $server = $_       ($_ | Get-View).config.storagedevice.mu... See more...
Thank you for the reply LucD. How can I add in the syntax to pull the host information from my text file? Get-VMHost | % {      $server = $_       ($_ | Get-View).config.storagedevice.multipathinfo.Lun | %{$_.Path} | `       select Name, PathState | `       Add-Member -pass NoteProperty Server $Server.Name | `       Select Server, Name, PathState } | ConvertTo-Html -head "Test" | Out-File "C:\Test.htm" Can I change the first line to "Get-VMHost -Name (Get-Content "C:\Scripts\esxhosts.txt")"?
I am trying to get a script working to show path status for multiple hosts, with the hope to show dead paths, and export the results to CSV. I found this post, and have been attempting to build o... See more...
I am trying to get a script working to show path status for multiple hosts, with the hope to show dead paths, and export the results to CSV. I found this post, and have been attempting to build on it. When I run the script against multiple hosts, the CSV does not show the hostnames in the output. Also, will this script output dead paths? & { $esx = Get-VMHost -Name (Get-Content "C:\Scripts\esxhosts.txt")   foreach($hba in (Get-VMHostHba -VMHost $esx -Type "FibreChannel")){     $target = $hba.VMhost.ExtensionData.Config.StorageDevice.ScsiTopology.Adapter |       where {$_.Adapter -eq $hba.Key} | %{$_.Target}     $luns = Get-ScsiLun -Hba $hba -LunType "disk" -ErrorAction SilentlyCOntinue | Measure-Object | Select -ExpandProperty Count     $nrPaths = $target | %{$_.Lun.Count} | Measure-Object -Sum | select -ExpandProperty Sum     $hba | Select @{N="VMHost";E={$_.Entity.Name}},@{N="HBA";E={$hba.Name}},     @{N="Target#";E={if($target -eq $null){0}else{@($target).Count}}},@{N="Device#";E={$luns}},@{N="Path#";E={$nrPaths}}   } } | Export-Csv "C:\Scripts\pathstatus.csv" -NoTypeInformation -UseCulture
That's a good point you bring up Frank. I was thinking about turning large pages off to help TPS kick in sooner. What are the recommendations for this?
Very helpful Bayu, this was exactly what I was looking for. Only thing I thought was interesting.. in misconception #2 it says "if a non-head extent member goes down, you won’t be able to acce... See more...
Very helpful Bayu, this was exactly what I was looking for. Only thing I thought was interesting.. in misconception #2 it says "if a non-head extent member goes down, you won’t be able to access the VMs whose virtual disks have at least 1 block on that extent." We had two 1TB LUNs with one of the extents as showing offline. There was 50GB free on the datastore, meaning data must have been on the offline extent. We were able to create a new 2TB datastore and sVmotion everything off. We were all curious as to how everything stayed up (not that we are complaining).