Does anyone have a good contact at Ubiquity Networks? Finding a pattern I don’t like.
The company has mostly fallen apart. Their sales are going up, but their responsiveness and customer support have been declining over the last five years.
Agree 1000% with the sentiments expressed by Mike.
Unfortunately despite much research I’ve been unable to find a suitable replacement vendor. All the other vendors seem to want to ram cloud-management down your throat which I absolutely do not want. My network, my control, not under the auspices of someone else’s magic cloud.
That is a big problem. In terms of their UniFi product line, there are no reasonable alternatives.
Upper management is the biggest problem. They have severe ADD.
A ton of companies have these kinds of issues. They just plain don’t hire enough people in the right areas to really excel.
Except, you could argue they are exceling. Stocks are going up up up, and folks buy the product.
I really wish stock holders would ask the proper questions in the quarterly calls.
I deploy Ubiquiti equipment quite a bit, both in WLANs and WISP distribution networks. It’s excellent quality at a dirt cheap price. As with all software-based products, there will be bugs. Your or my pet bug may never get fixed, based on market demand. That’s simply capitalism, not low quality. None of us can afford to pay for perfection, because it would never ship.
I deployed 400 HP-Aruba APs at SFO, and that installation requires a full-time network engineer to manage the system. I’ve deployed many more times that in Unifi APs and they run perfectly well with only periodic software updates to accommodate new client device types. Unifi is 75% cheaper than Aruba, for essentially the same result.
-mel
Bugs exist in hardware products too. The difference is that with the
software products, you'd hope for them to be fixed, whereas the ones
in hardware generally turn into RMA.
My current pet peeves with Ubiquiti are all on the router side of
things. OSPFv3 (IPv6) doesn't work correctly past EdgeOS v1.10.9,
and their BGP blows chunks - I've got an Infinity connected to a
pair of route reflectors handling a single IX (two route servers)
and it loses its mind, with the bgpd process actually going away.
Whether that's Ubiquiti's fault or should be blamed on ZebOS is a
debatable question. If you've got a vendor supplying your routing
software, it seems like fixing advertised features that are clearly
broken would be a matter of applying pressure on the upstream vendor
whose code used to work and then was broken, not a matter of "market
demand."
What's not debatable is that this has been the status quo for around
nine months. That's nine months without proper IPv6 support. And
this is their high end 10G full BGP tables router. Buyer beware.
The wifi side of things? Yes, the Ubiquiti stuff is very inexpensive
and it provides better value-per-dollar than just about anything else
out there.
... JG
Kind of OT for NANOG, but stock value is a terribly inaccurate way to measure if a company is “excelling.” Wall Street knows nothing of how to run a company, prioritizing quarterly profit over long-term success. Not hiring additional staff makes your quarterly numbers look good, but it isn’t good for the long-term attractiveness of your product. A good business doesn’t just target new suckers, they also keep existing customers happy. Eventually you run out of suckers and all you have is a bunch of burned bridges in your wake.
I subscribe to several feature requests in their community that are YEARS old with little to no response from Ubiquiti. Some of them can’t be hard for Ubiquiti to implement because they’re running on the exact same hardware and underlying OS and some of them you can configure in JSON files, but they just aren’t available in the GUI. They just don’t care. They’d rather push out Flavor Flav cameras or lighting.
They came out with a new product in a particular family and opened a new feature request section for it. I commented something similar to, “Start with feature parity with the existing product, then start working through the years of feature requests there. Come back when you’re done.”
This doesn’t just afflict equipment manufacturers. Network operators are in the same boat. Both groups have companies profiting hundreds of millions or billions of dollars every quarter, can’t spare a few hundred grand a year for a couple dev-ops guys to just bang out automation or features. Yes, I understand you rarely get twice the work from twice the people, but there are opportunities to make this better.
For you an I, a hundred grand of reinvestment in the product and business makes perfect sense. Make a good product, you will sell more of it, the customers win, the business wins, the shareholders win.
For those who ascribe a different line of thinking, a few hundred grand of reinvestment in the product means THEY DIDN’T MAXIMIZE SHAREHOLDER VALUE FOR MY 0.02% OWNERSHIP , TIME TO FILE SUIT!
Your or my pet bug may never get fixed, based on market demand. That’s simply capitalism, not low quality.
No, that’s low quality, full stop. Bugs need to be fixed in software that you are selling. I bought a product and I expect it to work. If they are going to tout themselves as enterprise grade, which they do (Narrator: They’re not), then they need to fix bugs in their production software.
JG,
I empathize with your BGP problems. I’ve had problems with BGP on anything other than Cisco for my entire networking life. It’s just the nature of the beast, although that’s not an excuse for ubiquity not fixing it.
But what is an excuse is market demand. How many people do you think speak BGP on ubiquiti routers? I know ubiquiti, like every company, likes to claim that they do everything. But no company can do everything, so you have to find out where their strengths are and avoid their weaknesses.
Personally, I always put a pair of stacked Cisco layer3 switches at the edge of every BGP network. This gives me reliable, redundant BGP peering that operates at wire speed and can still carry full backbone tables. Use Cisco hardware let me do this for less money then I would pay for a buggy ubiquiti router.
-mel via cell
I’ve been deploying Ubiquiti products since Robert was going around to the conferences himself, peddling his mini PCI card radios. I’ve deployed most of their product lines. I’m familiar with how they operate.
For the carrier side of things Mikrotik is a fairly standard replacement for UBNT stuff.
Mostly agree, but the “your pet bug” argument has validity.
When the thing that isn’t working is basic functionality, (e.g. IPv6) , there is no excuse for that not to work, and a company that tries to spin 'basic functionality" as “feature request” tends to dig their own grave.
But many, many bugs are a result of a Large Spending Companies that told a vendor “implement this thing or I won’t buy from you anymore”. Sometimes that thing is a good feature that advances the entire industry, and sometimes that thing is a bad feature that only helps their situation and they are simply throwing money to make it someone else’s problem.
This is why we see the historical pattern. Vendor enters the space and they are great! Over some years their code bloats all to hell and turns into buggy garbage because they are spending 90% of their time supporting 10% of the features from 5% of their customers because they spend the most money. Eventually someone starts something new and enters the space and here we go again. Sound familiar?
Gonna have to disagree with you there.
I’m not sure if it was a cashflow issue or what, but they launched the Unifi Dream Machine Pro after a very short testing period.
A client bought two thinking they could replace their office firewall with it.
There were so many show-stopping issues with the product, it was basically a brick for ~4 months.
You couldn’t configure the device unless it had a non RFC-1918 address on its WAN interface. It crashed frequently both due to software bugs and memory usage. Software updates frequently b0rked things…and backup/restore was broken. Interfaces would disconnect and reconnect for ~30 seconds several times per hour. I tested it for my client and had a list of ~15 show-stopper bugs that prevented us from putting it into production.
Thankfully they were “quick” to fix it. It’s been ~3 months and all the show-stoppers seem to be resolved. Things like logging and graphing are still broken, but that doesn’t stop internet access.
Regardless, they shouldn’t have pushed a completely broken device out the door as being ready for public sale. It should still be in beta today in my opinion.
-A
JG,
I empathize with your BGP problems. I???ve had problems with BGP
on anything other than Cisco for my entire networking life. It???s
just the nature of the beast, although that???s not an excuse for
ubiquity not fixing it.But what is an excuse is market demand. How many people do you
think speak BGP on ubiquiti routers? I know ubiquiti, like every
company, likes to claim that they do everything. But no company
can do everything, so you have to find out where their strengths
are and avoid their weaknesses.
Well, my point was more about the nature of software (it's fixable!)
and the "market pressures deter edge case bux fixes" argument which
appears to be a fallacy if you, as a hardware vendor, have paid a
license for some professionally developed product, like ZebOS.
ZebOS is the commercial offspring of Zebra, which forked Quagga,
which forked FRR. I have minor complaints about all of them, but
the open source developers have generally done well over the years.
ZebOS is integrated into a variety of networking devices. A quick
Google suggests this includes F5, SonicWall, Ubiquiti, Fortinet,
and other devices.
Ubiquiti produced its EdgeRouter Lite back in late 2012, able to do
a million PPS on a $100 platform, so there's little doubt about
their ability to create devices that do "hardware assisted" software
packet routing. I was kinda hoping that the marriage of their
hardware and ZebOS would result in a usable product. I am pretty
sure that's what Ubiquiti expected to happen, so that they would not
need to worry about the finer points of arcane routing protocols.
I don't really have a need to do a bazillion PPS. There's still an
Ascend GRF 400 here, and having passed the 150K routes mark, it now
serves to lift my office laser printer to a better height. It's
also part of why I try to avoid buying the hardware routers. There's
no budget for it and hardware routers generally provide far more
router than is needed here.
Personally, I always put a pair of stacked Cisco layer3 switches
at the edge of every BGP network. This gives me reliable, redundant
BGP peering that operates at wire speed and can still carry full
backbone tables. Use Cisco hardware let me do this for less money
then I would pay for a buggy ubiquiti router.
[assuming that was supposed to be "Used Cisco hardware"]
Veering way off topic here, I wasn't aware that there were layer 3
stackable Cisco switches that could handle full BGP tables. The
Ubiquiti Infinity is $1,800. I am curious what you're using. I
have nothing against used hardware.
... JG
This thread has taken a very NANOG turn. Whether the company has or
hasn't fallen apart, I'm sure someone is still there to contact.
Some say the poster is still looking for a contact at Ubiquiti to this day...
All of the traditional POCs are gone, most of them not replaced. There may not be anywhere to send the OP, other than to wish them luck on the forums.
What are you looking for OP? If it’s a vulnerability you should have no problem with the provided link. If it’s something else, let us know.
Most of my contacts have left there but there is still someone I can get you in touch with.