K2000 & RSA - Not all images sent from RSA
Hi All
We are running a test with 10 machines internally, where we are using a scripted installation to "remotely" re-image all of them simultaneously.
We have a RSA, which is sitting on top of our switch right here, which is configured to sync with K2000.
When we go to do the re-image, we are noticing strange behaviour... Out of the 10 machines, 9 of them are pulling the image from the RSA, but 1 (sometimes 2) is pulling the image from K2000 (which is located in another country).
This dramatically slows down our deployment.
Our RSA is a physical machine running ESXi, and has the virtual appliance installed there.
The switch we are using is a Cisco Catalyst 2960-X 24 port. So bandwidth isn't an issue (as far as we can tell).
We've tried setting the network on the RSA from Auto-Negotiate to 1000mbps Full-Duplex, but when we made that change, the RSA did not respond over the network. We had to change it back.
We've run the test a few times now, and every time it's a different machine that connects to K2000 and tries to pull the image from there.
Our RSA has the DHCP server running... although we also have another DHCP server on the subnet, but has an exclusion range for the range that the RSA is broadcasting on. What we can tell, is that all the machines are picking up an IP from the RSA.
Can anyone help? perhaps, is there a way to turn off a PXE response from K2000 (since we will only be using RSA's during operations)?
Many Thanks
2 Comments
[ + ] Show comments
Answers (0)
Please log in to answer
Be the first to answer this question
-Corey - cserrins 9 years ago
We downloaded the manipulator, but if I'm honest, it's not that straight forward.
Could you possibly let us know how to configure the manipulator to point directly to the RSA?
We will be having numerous RSA's in different regions across Europe... so in essence we could create a KBE for each location that the RSA is pointing to... Not the best way of doing things, but if it works, then I'm all for it.
Right now, we've figured that we can get them all to boot to the RSA, if we boot 5 machines at a time to PXE, with a 5 minute or so wait before we boot the next 5.
But if we can create a KBE that directly points towards the RSA, then this would be much better than our current workaround.
*edit* The other DHCP is just a standard Windows Server that is somewhere in Germany... the process to get anything changed on that is a real nightmare... (politics you know...), but what we have seen is that the machines are picking up IP's from the RSA...
My feeling is towards network limitations of some kind, but I cannot find any information in regards to incoming network limitations on the ESXi OS. - Flappers 9 years ago
-Corey - cserrins 9 years ago