Thanks for checking out the forums for the BIG-IP LTM VE. This is where you can come to discuss your experiences with LTM VE, ask questions of others that are using it, share suggestions, and look for help. This is not official technical support so there are no SLAs but the DevCentral community usually does an outstanding job of helping each other out, and you'll likely find a wealth of knowledge and experience here. So let us know how you like it, or if you haven't seen it yet check out the BIG-IP LTM VE page on DevCentral to get started. Click here Colin

I finally had a chance to check this out this weekend and thought it was awesome. Thank you everyone at F5 Networks for getting this out to the community!

I'm glad to hear you're digging it naladar. I have to admit I'm a pretty big fan myself. I've found a bunch of uses for it already and this is just the trial version. Very cool stuff. Hopefully others are having similar experiences. Colin

Hi you all, Of course I had to try to load LTM VE on my MacBook Pro (which BTW is awesome, thanks for rating my rule in the 2009 contest!) with VMware Fusion. I had some trouble setting up two host-only networks to simulate a client and server side within my Mac. But looking at the vmx files on other systems I was able to figure out how to set it up (after which you don't want to touch the GUI network settings as it does not understand the custom configuration and defaults to one host-only network again). I now have a fully operational development environment running on my Mac with one Ubuntu client, one LTM VE and two Ubuntu servers... great! Keep up the good work guys! Oh BTW, is it possible to downgrade to version 9.4.x too? Still loads of devices running v9 for which there needs to be iRules developed too ;-) Cheers, Sake PS OK... one can always wish for more, when will the iRule editor be available natively for MacOS/X? ;-)

This is certainly very interesting. I am slightly concerned with the 90 day trial, I'm pondering if its worth me spending the time building a lab only to have the rug pulled out from under it in a few months. Its not entirely clear if its just a case of applying for license renewals every 90 days (So that you guys can keep control of installations, force upgrades, or eventually make it a commercial products) or if it really is a case of 90 days and then pay. (And then we're into wrangling with my company taking forever to raise a PO for *any* amount of money and having to arrange a server to host it on, rather than using my home desktop!) EULA isn't clear at all (Its just the generic one on the download link and it looks like i will have to install and activate to find out on the rest) PS, do we know if this will tie in with enterprise manager if i do get it running on proper servers?

Firstly thanks F5 for having the sense to release this product. There's nothing worse than not being able to spend time learning a product properly, and for the vendor to actually encourage this is an obvious yet rare thing to do! I agree with Steve on the the license issue. What happens after 90 days. My plan for this is for it to have a permanent place in my lab, but this requires a permanent licensing scheme. I don't mind paying a small amount, and I don't mind it being crippled.. I'm also curious how F5 are going to use this product to increase/maintain share. F5 units are really good, and really expensive. That's fine for production, but when you also have to put them into your pre-prod, staging, and development environments, as well as one for the network geeks to play with, it gets silly. One thing this new product could do really well is provide different levels of cripple - so for example a free (or really cheap) version for the lab, which maybe reboots every 6 hours and only allows 256kbs of throughput - so there's no way it can sneakily be used for small production sites. Then maybe a 'still cheap but not given away' version for development environments, which maybe doesn't reboot itself but is still low performance only. Then maybe something which is performance limited, but you're allowed to burst for load testing.. Then the real boxes of course for live deployments. By doing this, F5 becomes a much more viable proposition commercially, and it will at the end of the day help sell more LTMs. I say this as I'm in a position where we're refreshing our dev, staging, and preprod envionments, and the F5 part of the project will cost more than all the rest of the network infrastrucuture combined. Some inexpensive VM options would mean we keep F5.. Also I really want one of these in my Lab ;-)

Welcome to the BIG-IP LTM VE forums

45 Replies

net_enthused_25
Nimbostratus
Mar 10, 2010
Thanks Treepose,

The netstat command revealed that the 172.16.X.X

management network was in the routing table, but only specified as a broadcast address 172.16.232.255 or a subnet; 172.16.232 /24

I'm obviously a networking newbie and don't know the correct interpretations for a lot of this information. Plus the VMware adds another layer of abstraction to straighten out. For example, the 172.16.232.128 (my management IP) should only exist with the virtual emulation, but my router has to know it's there in order for my browser to communicate with it. Is there a way to do that? For now, all I can do is access the GUI through the internal self IP (which I know is not recommended) .

What's setting me back even worse is that "b reset" - to wipe out all my settings and start from scratch - is registering as an invalid command. But again, maybe this is an artifact of working with a simulation and not the real thing. Nonetheless - it's nice to have an BIG IP to poke around on !

I'm suspecting the problem might have something to do with the way I set up my host (Using a real qualified domain from dynDNS the first time. Now I'm using a dummy domain, but maybe it doesn't matter). My original BIG IP VE setup had a management IP that was actually accessible..... BUT - none of the pool members were reachable !

Living and learning.
treepose_102627
Historic F5 Account
Mar 10, 2010
b reset isn't a supported command for BIG-IP. There are manuals around to peruse... here's the method: "b import default" - and this KB entry applies (taken from the release notes: )

"bigpipe import command (CR136004)

Use of the b import default command can generate Security-Enhanced Linux (SELinux) errors. You can ignore these errors."

If your workstation (Web Browser) is on the same subnet as your LTM VE's mgmt port, your router shouldn't come into play. Host only and NAT'd networks accomplish this. Short of that, reviewing how layer2 and layer3 networking plays with each other is probably a good idea.
net_enthused_25
Nimbostratus
Mar 11, 2010
Thanks, Treepose

I'll try and read up on the literature (there's no shortage of it) and figure out what's going on.

I initially thought it a good idea for the management IP to be on a separate subnet from my actual subnet, as I understood that this was closer to a real world usage scenario.

Thanks again,

dwc
treepose_102627
Historic F5 Account
Mar 11, 2010
Net_enthused: One style of deployment involves three subnets. One for mgmt, one for "internal" traffic and one for "external" traffic. This is covered in the release notes with configuration examples.
net_enthused_25
Nimbostratus
Mar 13, 2010
I think I understand that deployment configuration fairly well, as that's what I'm trying to establish.

The problem is that if my management port is located on a 172.16.232.x , and my actual computer subnet is 192.168.1.x , my browser has no idea how to reach the mgmt IP.

I'm guessing that my router searches for 172.16.232 and ends up hopping out into the internet instead of back into my box. I tried to establish a static route in my router, but that didn't work. This issue might be related to the layer of abstraction that the VMware adds, but I'm not sure.
net_enthused_25
Nimbostratus
Mar 13, 2010
addendum -

I'm pretty sure at this point it's an issue with my VMware

network settings. And nothing to do with my router. I have the VMware Fusion network settings set for "Host Only" and they seem to be pointing at the right 172.16 network... but how that bridges to my own actual 192.168 network is still hazy to me. If it's all a layer 2 issue, they're all on the same box and should be honking at one another just fine. Hmm....
qe_102628
Nimbostratus
Mar 13, 2010
net_enthused: Your Fusion host should have a virtual interface in the 172.16.232 network. It's will be one of the vmnet interfaces that shows up when "ifconfig -a" is ran inside of a terminal window. I hope that will clear up your haze.
net_enthused_25
Nimbostratus
Mar 17, 2010
Yep - the vmnet networks were the key. I had actually

designated an management IP with the exact IP address of the Fusion virtual machine; which created a conflict. When I stayed with that network but changed to a unique host address, my problems were basically solved. I'm sure it was staring at me in the documentation somewhere.

Good call qe; thank you !
hoolio
Cirrostratus
Jan 21, 2011
Did this post get moved at some point? It looks like it's in the Advanced Design/Config section, not the VE forum.

Aaron
BA_7682
Nimbostratus
Aug 29, 2012
Is this forum still active?

I cant create a topic in F5 VE group