The Riverbed Blog (testing)

A blog in search of a tagline

Should Cisco WAAS users upgrade to version 4.3?

Posted by riverbedtest on December 6, 2010

Cisco recently released the newest version of WAAS–version 4.3.  As with the earlier 4.2 release, there are no major new functional capabilities introduced by Cisco into this release, other than the availability of a virtual WAAS edition (vWAAS) and WAAS Express (both of which incorporate older pre-existing WAAS software capabilities).  If you are an existing WAAS user, is it worth upgrading to the newest WAAS version 4.3 that is now available?  To answer this question, I took a look at the WAAS 4.3 release notes posted on Cisco's website.  Here are some observations that might cause WAAS users to think a bit first before upgrading:

1)  According to the release notes, WAAS version 4.3 has 35 "resolved caveats" while introducing 39 "open caveats".  Most of us can accept that introduction of new features and functionality into any software product inevitably will be accompanied by bugs and other issues.  Certainly, that should be expected in any software product, including WAAS…until you consider that WAAS 4.3 has no major new software features or functionality!  It is startling that software that is largely unchanged from previous releases would still have 39 "open caveats."

2) The "Requirements and Guidelines" section lists a total of 28 distinct stipulations to consider before performing the upgrade to WAAS version 4.3.  And these aren't things that you can blissfully ignore as we typically do when accepting "terms and conditions" while installing new software.  Some of these requirements–for example, changed default values for the WCCP source IP mask–can potentially mess up entire WAAS deployments if they aren't addressed properly.  However, there is one particular requirement among the 28 listed that you have to think very carefully about:

  • After upgrading application accelerator WAEs, verify that the proper application accelerators, policies, and classifiers are configured. For more information on configuring accelerators, policies, and classifiers, see the "Configuring Application Acceleration" chapter in the Cisco Wide Area Application Services Configuration Guide.

What does this mean?  Well, for those with involved or complex WAAS deployments, the above requirement is rather disturbing when you consider what it potentially entails.  The requirement calls for careful inspection of rather detailed application optimization (AO's) configuration settings–of which there may be as many as a dozen AO's–on each and every WAAS device in the network.

Now, why would Cisco insist that customers doublecheck and validate their detailed pre=existing configuration settings?  For many WAAS users these configurations are the product of months or years of troubleshooting and headaches.  Could it be that the upgrade process might corrupt the previously stable configuration?  Could the upgrade to 4.3 destroy the months or years of work and toil needed to get the WAAS deployment to a reasonably-stable condition?  And what if the upgrade is unsuccessful?  Another thing to keep in mind here is that Cisco's documentation indicates downgrading from WAAS version 4.3 to a previous version is also a very intricate, potentially dangerous process. 

Based on these observations, it seems to me that upgrading to WAAS 4.3 involves far more risk than reward.  After all, what would you get?  There are no new functional capabilities in WAAS 4.3. Cisco's documentation indicates that WAAS 4.3 introduces more known bugs than bug fixes, and that there are an astounding 28 different requirements that you have to carefully pay attention to.  And if the upgrade doesn't proceed as hoped, reverting to the previous software version appears to be almost as dangerous as the process of performing the upgrade in the first place.  To me, it seems WAAS users have far more to lose than to gain by upgrading to WAAS 4.3.

14 Responses to “Should Cisco WAAS users upgrade to version 4.3?”

  1. The Cisco bashing reflects poorly on your brand.
    I am by no means drink the Cisco 600 pound gorilla kool aid and consider Riverbed, F5, Airohive and Arista to be better-than-Cisco at the things they do best. Let YOUR product’s technical superiority and peripheral resources (whitepapers etc) define our desire to buy WAN optimization from you.
    I subscribe to this blog so that I can accurately voice to my customers the reasons why Riverbed warrants investment over WAAS. Every post you construct attacking WAAS makes me wonder why you feel the need. Act like the incumbent. Own this space. Don’t ‘fight’ for it.

  2. Josh Tseng said

    I appreciate the thoughtful feedback and your support for Riverbed.
    As to your question about the purpose of this blog and others like it, there are a lot of people out there who are interested in knowing the differences between Cisco WAAS and Riverbed, and this post was intended for them. I wish everybody out there was as enlightened as you, but unfortunately this is not the case. While we do have internal competitive materials, there is no publicly-available place for them to find this information.
    The purpose is not to “bash” Cisco or any other competitor, and it’s unfortunate if it came off wrong for you. I have tried my best to stick to the facts, for the benefit of those who want to know the details about WAAS from the other side, not just from the Cisco sales rep. I deal with these customers every day, including some who had previously purchased WAAS because they assumed it was just like Riverbed, and at the time of purchase didn’t have access to objective data points telling them otherwise. This blog was intended for people like them who I’m unable to talk to personally.
    So please do accept my regrets, but also understand that there are those out there who are interested in this material.
    Best regards,

  3. Matt said

    First of technically this is good information to share however, I have had this discussion with you in the past, I will repeat what I said to you before, you need to stick to the facts on the the issues involved with your competitors and do less attacking / opinionating of their product.
    Overall its a very informational post. I just think the tone needs changed.

  4. Larry Chaffin said

    The only reason Cisco wins deals over Riverbed is price. We all know their quote list pricing is a joke, everyone gets 45% off or more. Which leads to the question why are their list prices so high? Just so they can give huge discounts? I have been seeing in the 50% now with Cisco giving WAAS modules away with router purchases. But then I see customers calling after a year saying we need to look at Riverbed as the WAAS is not doing what we want and Riverbed does so much more.
    Customer who buy just on price, I can understand that. Everyone has a budget and needs to stick to it. But looking at the reporting, the RSP and how much more Riverbed can do presents a value that customers need to really look at as an investment.
    Plus Cisco has done nothing to the WAAS in a along time but fix bugs. They are selling wan acceleration because they have to sell everything. But you cannot be good or great at everything. They are losing router and switch market share now due to this, you really think they are going to put money to the WAAS?
    They will be fixing bugs for years to come.

  5. A customer and a Partner. said

    I’m constantly amazed by the new Marketing lows set by this company. Couple of things in the article
    1) “The only reason Cisco wins deals over Riverbed is price”
    So you are saying they’ve never beaten you in a environment that had firewalls in place whereby the customer didn’t want to modify to allow your WAN opt to work. I’ve seen it.
    Are you saying that module integration in the router isn’t a key selling point for them? I’ve seen it used to defeat you.
    Are you saying that their original idea of virtual servers running on a WAN optimization product, completely MS certified from a single support vendor has never beaten you? Uhmmm, yea, I’ve seen that one too.
    I guess that point has been beaten to death.
    2) Now to the completely ludicrous statement of
    “of which there may be as many as a dozen AO’s–on each and every WAAS device in the network.”
    The AO’s that you mention are
    CIFS and Windows Print
    I count 6 separate protocol specific AO’s
    Not the dozen that you mention.
    3) The configuration of thousands of WAAS devices (routers, appliances, modules,etc) are all done by a central manager and it’s only recommended to double check your configuration to make sure everything is as it should be, as any proper organization would do.
    4) When you speak about “losing router and market share” you are just showing your ignorance in an area that you know nothing about. Or maybe you are starting to spread rumors because you are planning to create a router or switch platform? Good luck with that.
    You said this forum was to educate customers, but instead you are slagging other vendors and flinging FUD. Is that it?
    BTW, since this is a public form and not everyone has access to market share information, this link is decent
    Cisco is up to pre-recession numbers, despite HP/3Com’s tactics and still maintains 70+% Market share in routers. Increasing SP Ethernet/Core share and Security. Not perfect, but they’re doing the right things.
    Now let this forum and company get back to what their focus is and not spreading FUD and slagging others.
    Good luck.

  6. Larry Chaffin said

    Let’s talk price, did not see anywhere you mentioned discount, now that’s funny. Cisco is all about inflated list price to give 48% to what 60% to customers to win deals. Then they still make a ton of money on it. That either means the list pricing is really bad or the equipment is just cheap. You tell me. It is a wonder that Cisco can give such huge discounts on gear and still make such huge profits.
    Cisco needs some real list prices that are half of what they are now. Be real for the customers.
    By the way, tell Cisco to Build America, not China and India.
    Let talk about the Cisco firewalls, is it true that Cisco put a standard policy in to block the Riverbed Steelhead Probe in all ASA’s? That’s the word on the street. Who would use and excuse like we cannot make a simple command change to an ASA ,then going to oh we cannot buy Riverbed now, we need to buy Cisco like our Cisco AM told us. That’s bullshit. We make the changes all of the time and laugh, things Cisco Engineers and AM’s will tell customer to get them to buy the WAAS.
    Now let’s get to the virtual server on the WAAS, it sucks. Customer say it I have seen it.
    The strategy of the buy a router and get a WAAS module free, well lets just say those modules are not that good either. I have seen many customer turn them off and pull them out after the router utilization goes high due to the WAAS Module. Don’t tell customers that one do you?
    Cisco can try to have the great marketing machine fix the WAAS with customers, but it does not work. I have had two customer pull it out as it did not work, was horrible to configure and did not work in a large environment.
    Going back to Market share, Cisco cannot even beat Blue Coat at the moment, look at Gartner, where is Cisco? I would not be surprised if Juniper or Expand came back to beat Cisco. Cisco needs to buy blue coat and expand just to get the WAAS to where it needs to be.
    I forgot Cisco is a Niche Player also in Gartner for, we all wonder when the WAAS will be there since they have made no technological advances in the product in years. BS and Marketing can only get you so far, if Gartner looks at how many WAAS Cisco gives away, I wonder if the market share would drop below expand and move them to the Niche?
    Remember the saying “ You never get fired for buying Cisco, but maybe you should”

  7. A customer and a Partner. said

    Wow, Larry, you are really off aren’t you.
    We can play the “I have seen a customer remove vendor XXXX game all day” because I’ve seen Riverbed removed when it broke the environment, management reporting was off by 15% as it doesn’t take into consideration the IP overhead, system failures, etc etc. But why?
    Let’s take your last final statement. It truly shows that you don’t spend the time to investigate your claims and just throw FUD around.
    “I forgot Cisco is a Niche Player also in Gartner for, we all wonder when the WAAS will be there since they have made no technological advances in the product in years.”
    You linked to a completely wrong product. Is that how Riverbed does business? BTW, RVBD is not even on the best. Better check ya QA before you post.

  8. Marko, Australia said

    Got 30 waas in production and now testing waas express! Initially we had lots of issues (4 years back) with the firmware but now its all settled down and working like a charm except for ica compression!
    Havent played with Riverbed but hay, the management only likes Cisco!

  9. Josh Tseng said

    Hi Marko, you don’t know what you’re missing. Give us a call if you want ICA optimization and better performance for all other apps. Here’s a customer-provided screen shots of ICA improvements:
    Congrats on getting WAAS working after 4 years.

  10. Eric Johnson said

    We just purchased and implemented a ~$70,000 7-site WAAS solution with WAAS Mobile. This blog post was a major factor in our decision.

  11. Josh Tseng said

    Thanks for reading my blog. I’m glad you are doing your due diligence by looking at Riverbed’s web site before buying Cisco.
    Good luck to you,

  12. Ciscojunkie said

    Cisco WAAS only works with IPv4

  13. Shaun said

    I too agree with what Harry said at the start of this. We choose Cisco over Riverbed (and we tested both) based on Riverbed’s arrogance and attacking of Cisco constantly. We saw huge gains with the Cisco product and while price was somewhat of a factor, the Riverbed didn’t have anything more compelling to make us switch. So it came down to the relationship and I hate to say it you guys really need to tone it down. If your technology is as good as it is, then let it speak for yourself. Maybe take a page out of F5’s book. They know they have the best product in that space, do they hammer everyone? No. In the end we are very happy with our WAAS deployment and these upgrades that we did (and always do) have not impacted our deployments at all.
    I will say on a positive note Riverbed had a better reporting tool however that wasn’t enough to stop me from gagging on your guys arrogance

  14. There are 28 distinct stipulations to consider upgrading for a new version. And I am excited for this new output but I have to undergo for these guideline. It will worth it!

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: