Cisco thinks you're happy to wait ages for new kit, then pay premium prices Cisco has warned buyers and investors that supply chain pain is likely to persist for another few months. In remarks delivered during the company's Q1 FY22 earnings call for the three months ended 30 October, CEO Chuck Robbins said that in the first half of the quarter Cisco saw "some deterioration" in component availability, …
I'd expect a large continuing uptick in cloud and carrier orders given recent events.
The price hike has forced our project managers to resubmit new costings for all future activity, any cancellations will happen long before Cisco see any orders.
Future activity is also highly constrained by the fact that we're still waiting for products ordered in the first half of this year and there's a finite amount of cash that the company can have sitting waiting ready for things to appear - We buy kit to support activities that contribute to the bottom line.
I dunno. Once upon a time, carriers bought Cisco and Juniper. Now, probably less so. If your network is mostly optical, they don't really play in that space. If your space is Ethernet, they don't really play their either. They're typically the most expensive way to deliver a VLAN in pretty much every capex and opex sense. And the incessant drive for 'service' revenue hasn't exactly endeared them with carriers either.
Some disruption has occurred thanks to trade spats and the world catching a cold. That's also allowed profiteering. But the supply chain shifts, and capacity that was supplying Cisco can just as easily supply commodity switches. For a fraction of the cost of a 'premium' brand.
I know plenty of Cisco and Juniper shops still.
A little puzzled about Cisco not playing in optical or ethernet.
I see carriers still installing Cisco gear in the carrier hotels. I see Cisco GPON gear out there.
One thing different now vs. in the past, is there is a *lot* of commodity ethernet gear out there now, since the industry switched over to mostly commodity Broadcom chips than roll their own. So 2nd and 3rd string gear gets used a lot more.
But its not like Cisco's slice of the pie is greately diminished, just the pie got so much larger, letting others into the table too.
SOHO/SMB barely bought into Cisco in the past, and rarely do now. But telco, enterprise and alot of mid-range still buy a lot of Cisco.
Last time I think Cisco tried was with the 15454 switch they bought in. Added to my collection of polo shirts. But I've also got a few from Infinera, which although post-dating the 15454 are proper optical switches.
OK, so perhaps that's a bit harsh, and there's now a Cat for that*. But much the same problem, ie vendor lock-in with consumables at markups that would make an ink jet vendor blush. And then certifications, service contracts, spares etc. Other vendors are often far cheaper and easier to deal with when you're buying components in city, state or country sized quantities.
Or just take a look at the club pushing for 25gs-PON, because that's a better roadmap than legacy Ethernet players like Cisco are clinging to. As you say, basic Ethernet has long been a commodity, not a 'premium' service like routing. But such is politics. Cisco engineers helped spawn tag-switching, which became MPLS. But Cisco's marketing types decided that would be a premium router license feature, and crippled the Cats so they couldn't (MPL) Switch.
Funnily enough, carriers noticed the bait and switch, and other vendors have been happy to displace the old guard. After all, in an MPLS world, the Internet is just another VRF.
* Catalyst PON. AKA CaPON. Tastes like chicken, only a lot less bang for your cluck.
Exactly this. We were Cisco fanatics for years because they had the cool features we wanted and more importantly they delivered the closest thing you could get to a guarantee of no unscheduled downtime.
Neither of these is the case anymore. The features in question have long since been commoditized, and Cisco reliability is nowhere close to what it once was. In fact, we're seeing much better reliability from products that cost less than 1/10 as much - because they have far simpler software stacks based on mostly on generic Linux functionality that's stable, tested, and mature. I'm fine with that if it gives us the features and performance that we need. And I'm delighted with spending less money for better uptime.
Carriers prefer Huawei kit rather than cisco/juniper, BUT they have been banned buy the US from using it AND the US has banned Huawei from buying most chips.. so back to Cisco it is.
Really big fish (like Google, Amazon, Facebook) use their own stuff, much cheaper and better.
So it is probably carriers and corporate.
I'm getting more dissatisfied with Cisco daily. Just had to go through a 259 page PDF manual to get three relevant lines of code to program new switches for "Smart Care." Two of those lines errored out and I had to engage tech support. Yes, a tech support case to register switches, FFS.
They make systems and documentation so unnecessarily complex that I'm sure they are employing >500 more people than they really need.
I'm currently playing with Ubiquiti switches in my home lab and they are refreshingly easy to program and use. There are other options out there.
While I'm not quite ready to go full Ubi at work(though I have their APs at home, and they push packets quite well) I am really done with Cisco.
Between the switches and the phone system, Cisco makes the things that are most likely to get me fired at this point. The call manager ate itself and we found out the hard way that their "Backups" are not, that even with Cisco TAC, a sales rep sympathetically applying a motivating boot, and an onsite cisco consultant, restoring the backup files to anything other than the original UCS box they set it up with resulted in a non-funtioning system. We ended up re-installing everything from scratch basically. Now I spin off full vm copies because I know their backup/restore procedure is a lie.
Currently it has a lovely quirk where if I restart the core switch stack(which can take 20mins) due to something like a 1 line change to the MTU setting, the call manager core dumps because it lost it's network connection for too long. Too long being shorter than the minumum time to reboot the switch it's plugged into, also made by Cisco. The phone system then needs to be coaxed back into booting via the VMware console, takes forever to start, and has to have services restarted manually. So if there is a blip in the core, the network goes down for just shy of half an hour, and the phones will be offline for another 30 mins after that at least. Thank god for UPS technology.
We are considering alternative bids in the next renewal, needless to say.
We are considering alternative bids in the next renewal, needless to say
We are in the same boat. We were (past tense) happy to pay an arm-and-a-leg for something reliable and stable.
Nowadays there are vendors asking us "what numbers do you want to see" if we switch to them.
Go with Cisco, they say. They have the best TAC support. Not sure `bout that anymore. I have several TAC Cases where agents, consistently, play the game of "stop-the-clock" with me.
Do it!
I had a fun week in Frankfurt being indoctrinated so I could sell Cisco's telepresence gear. In which I learned more about CallMangler than I wanted. Like the brain being basically an under specced and very over priced linux box. And then some fun conversations about why we couldn't just spin that up in our cloud as a VM. Then we could manage the environment and do sensible stuff like simple backup imaging and synchronous replication for DR. But Cisco wants to flog huge margin tin, and of course 'services'.
I can't remember if they did decide to allow a virtual Mangler, but a number of other large carriers were pushing for that as well. Although Cisco pitches itself as 'carrier grade', it often isn't, which can make it a major PITA to offer their stuff as a service with a decent (and honest) SLA. Or being told we could promote their stuff as a 'cloud' service, when realistically it was just hosting core tin in an off-prem DC. Which also creates problems and costs to rack, feed, and manage their tin.
Their was also some other fun stuff. To get their top-end Telepresence solution certified, the room had to meet Cisco specifications, which meant some construction needed. So it was fun telling project managers that they'd need to manage building contractors to code and to Cisco. That also went as fair as paint. ISTR Cisco had done a deal with the holders of Bob Ross IP , which meant a choice of 3 Cisco Powered paints. Exorcist Green, Diarrhea Brown, or a slightly more tolerable blue. Wrong paint, and Cisco wouldn't take the solution into support.
Funnily enough, after talking to other CCDAs, they had the same thing, the cert trains you to push tin, but also when & why not to push Cisco. Their were also other potential challenges. If you're an honest engineer, you design a solution that's best for your client. If you're a Chartered engineer, you risk losing that status, if you knowingly offer a solution that sucks donkey balls. Even if that's the vendor certified design.
But TL;DR Bismarck was dead on about sausages. VoIP and video are mature and more open technology now, and there are a lot of good and cheaper solutions on the market. That includes core services, a slew of VoIP handsets as well as important stuff like regulatory compliance for logging. And there's a good push for open source solutions, either managed or DIY. You may not see those saving Jack Bauer, or in a Marvel show, but they work. And you can often get better support via Reddit or Discord than a big vendor's TAC. It shouldn't be a suprise that large carriers like AT&T have been moving in that direction.
Same for us.
In those days CISCO issued a decree that switches we bought last year are no longer supported ... And, lo, corporate IT came upon them, and the glory of the CFO said "ye shall spend all this left over budget on new CISCO stuff, just make sure it's done this financial year"
And they did announce that there was a 9month(*) waiting list for the new kit and we were sore displeased.
(* we're used to getting 'impregnated' by CISCO but we didn't realise that's how the switches were made)
My daughter and son-in-law are experiencing 9 month lead times and sky high prices for parts. Not for networking gear but the parts needed to overhaul a six cylinder engine. The parts cost estimate alone costs about $25K. Sounds ridiculous? Especially as brand new crate engines of similar output can be had for a fraction of that price. The difference is that its for a light aircraft. Once you're in aviation you're in a world of monopoly supply so costs go through the roof and lead times are just "take it or leave it". This is the price you pay for restricting competition.
We are told repeatedly how competition is the cornerstone of our economic system, how the race to satisfy customers with the best price/performance drives product innovation but this is really a half-truth. The reality is that as a business matures the suppliers consolidate into a handful who divvy up the market among themselves. They also defend that market against what they deem as unfair competition -- that is, anyone who produces similar product at a significantly cheaper price point. If they can government to enforce the monopoly (like they do with aviation) then their business is secure. They just need to find an angle (with aviation, its safety; communications, its security).
I do most of my work in the SMB space ($10s to $100s of $Ms in revenue) and we've had lead-time issues with Cisco for years. Granted, these were "weeks and months" not "months and quarters." But still. I get that companies don't want to have inventory on the books, but Cisco has been taking JIT to such a ridiculous extreme that any disruption was going to be painful and a huge disruption has created absurdity.
Let's just call this what it is: a company deliberately shifting its inventory management risks (and associated costs) onto its customers.
Yes, carrying inventory costs them money - it ties up capital and capital has cost. But inavailability costs their customers far more. It is perfectly clear where Cisco's priorities lie, and they are not with their customers.
I started moving my customers away from Cisco a while ago because of the costs of project delays, along with their noticeable decrease and ongoing decline in software quality over the last decade or so.
Perhaps other customers and consultants should consider the same.
My impression is that CISCO is now Oracle levels of Evil.
But it has become the "Nobody ever got fired for buying X" of networking with Cyber being the next threat.
CISCO switches have a bunch of features which we never use, but nobody wants to take the risk of buying some "lesser" system
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
