Future Support for Prime Infrastructure 1.3 vs. Prime LMS 4.3

[CARL VON HASSEL]

Anyone have thoughts as to how Cisco is going to require support/licensing going forward pertaining to the migration to Prime Infrastructure? Since we own Prime LMS 4.3 and have an existing SAS/SAU support contract I was able to get Prime Infrastructure 1.3 gratis. Although it was a PITA I now have PI on my SAS/SAU contract.

For numerous reasons (mostly PI 1.3 is feature limited and buggy) I maintain both Prime LMS 4.3.2 and PI 1.3. By the way Cisco has finally got Prime LMS to be a viable Network Management platform. It does most everything you need it to do and is mostly bug free (must be time to sunset it).

So going forward when it comes to contract renewal time do you think we will need to pay for support & licensing for both Prime LMS and Prime Infrastructure? If so I say massive fail. We will drop PI and continue forward with Prime LMS until EOS/EOL.

Thoughts?                     

[Marvin Rhoads]

Carl,

You may wany to sit in on one of the weekly product demo / Q&A sessions (details below) and provide your input to the network management product team.

Personally I agree that LMS in its current incarnation is pretty solid - definitely more feature rich than PI. (Though I would argue that there were things back in the CiscoWorks days that were done even better.) Cisco certainly knows this and is working hard to make the PI 2.0 release equal in features (reference). I believe it will come a long way beyond 1.x but would be very surprised if they succeed on all fronts.

Right now SASU for Prime covers PI and LMS products. I doubt this will be discontinued until PI has FULL feature parity with LMS.

(Click image below for larger version)

[CARL VON HASSEL]

Marvin,

Thanks, I will sit in on the product demo. Sorry if I sound so negative but I feel Cisco has the ability to do so much better in the Network Management realm. I was excited when I heard they were coming out with Prime Infrastructure even though I was really beginning to like Prime LMS. I figured they would improve on the Prime LMS product. Then I started working on PI. What a disappointment. I'm the kind of guy that hates errors and failures. When I see a device (LMS manages over 500 devices) failed a configuration archive or an inventory update on Prime LMS I want to know why. I have Prime LMS tuned now so this rarely happens. I understand that PI is on version 1 but when I change snmp settings on 500+ devices (yes you have to change them one at a a time) from the default of 2 retries with a timeout of 1 second to 3 retries with a timeout of 10 seconds. Took me about three hours because I was tired of seeing the Managed with Warnings message. The very next day all the settings were back to 2 and 1. This was fixed in PI 1.3 but you still can't change the settings for more than one device at a time (really?!?). Or when I check the box on Administration > System Settings > Monitoring Settings > Auto Monitoring and the box doesn't stay checked. Or when Clients > Current Associated Clients does not contain IP address information. These are just a few examples. On every one of these items I opened a case with TAC and the answer is always the same "It's a bug". In my opinion these are not minor issues.

A few other issues I had when I started using this product are:

It took my sales team months working with entitlement so I can download updates and patches. Yes I can get the product through PUT (product upgrade tool) and if I asked the sales team they can provide me with a link to download an update or patch but this is annoying. I don't want to waste their time or mine.

Also try opening a case with TAC on PI, it doesn't exist in the drop downs. So you have to open a case with LMS and when the engineer makes contact they have to transfer you to the PI group.

All this adds to my frustration. Don't get me wrong I believe eventually they'll get it right, version 2 will probably be usable, they'll be getting there in version 3, and they'll nail it on version 4 just in time to sunset it.

I see this in other products as well, when was the last time you've seen GD code for routers & switches? I can't tell you how many times we've had outages because of a bug. So we upgrade the code for the bug fix and find another bug, and so on and so on...

I've had management tell me to start looking for other manufacturers so we can get some stability in our network. This coming from managers who three years ago never would have entertained the thought.

You would find it hard to believe but I'm a Cisco fanboy.

[Marvin Rhoads]

You're preaching to the choir with me, Carl.

I've been using Cisco's network management products since 1995 - I'd venture to assert there are few if any people even in the BU who've used CiscoWorks classic on Solaris (or even AIX!) through CWSI and VLAN Director to CiscoWorks2000 and through LMS 2.x, 3.x and 4.x.

Cisco's big and certainly not without their bugs and product development missteps. But the competition has its issues too. Hang out in Solarwinds Thwack community a bit and you'll see some persistent issues there. And don't even get me started on Juniper's ScreenOS, JunOS and JTAC....

Perhaps we should all jsut run RANCID, Nagios and Cacti (open source configuration, fault and performance management tools) and fix the source code ourselves when things don't work.

[CARL VON HASSEL]

Agreed, I too started with the originial CiscoWorks. I use Prime LMS to manage an all Cisco enviornment but I also use

1) Solarwinds Orion - great front end and overall view of the network. Noobs love it.

2) Scrutinizer - beats Cisco hands down for Netflow

3) ManageEngine OpUtils - we used this before we were "all Cisco". Still beats CiscoWorks with managing switchports and IP addresses. Has tools that I wished CiscoWorks had.

I look at Prime Infrastructure and I see the potential.