Ubiquiti updates

[mackintire]

Actively developed devices with huge potential and fall short should get re-reviewed annually until they are 3 years old.

Most of Ubiquiti's stuff would fall into this category.

I think you performed a review of the ERL with firmware 1.0.2 firmware and some updated comments for versions beyond 1.2


1.6 was just released and included UI changes and rewrite for the default setup wizard "for first time users".

You may also want to test out the AWESOME fq_codel QOS plugin. Magic voodoo that works to mitigate bufferbloat (usually on the WAN side).

It would also be useful to see if the performance issues with eh 8 port edgerouter are resolved as seen from your end.

Best Regards,

Michael

 

[Trip]

A very good suggestion. One could say the same for select Asus, Netgear, Linksys models... they all seem to be releasing updates for at least a year or two on their higher-end devices.

Anyways, the 1.6+ edgerouter firmware pretty much makes all the capability accessible from the GUI and, from my vantage point at least, drastically increases usability by the masses.

 

[Cloud200]

As much as I would love to agree with you on the improvements ubiquiti has made to the firmware . . . every so often I get one shipped with version 1.0.2-1.2 and that is a scary thought.

In addition, one can hardly call someone that currently appreciates an ERLite average. Most users that I deal with on a daily basis can't even tell the difference between a WAN and a LAN port without color coded ports and cables let alone set a static IP address.

Once they are shipped with an ID-10-T proof firmware that has the wizard's settings (DHCP, NAT masq, Firewall, etc) applied by default then I would say they are getting closer to being a drop in replacement for a WRT54G

 

[zentec]

Once they are shipped with an ID-10-T proof firmware that has the wizard's settings (DHCP, NAT masq, Firewall, etc) applied by default then I would say they are getting closer to being a drop in replacement for a WRT54G

They are not intended to be a drop in replacement for a WRT54G and I am sure UBNT would be loathe to have that consumer market. These are price sensitive support intensive people who at the drop of a hat will take to Amazon to vent their frustration over lack of support because they can't figure out the RJ-45 on the far left is a serial port and thusly, the "router sucks UBNT sucks".

I think the ERL-3 has already attracted the attention of the DD-WRT crowd however, I believe there is a package for it. It probably doesn't support the hardware acceleration, so that sort of defeats the purpose.

 

[Cloud200]

They are not intended to be a drop in replacement for a WRT54G and I am sure UBNT would be loathe to have that consumer market. These are price sensitive support intensive people who at the drop of a hat will take to Amazon to vent their frustration over lack of support because they can't figure out the RJ-45 on the far left is a serial port and thusly, the "router sucks UBNT sucks".

I think the ERL-3 has already attracted the attention of the DD-WRT crowd however, I believe there is a package for it. It probably doesn't support the hardware acceleration, so that sort of defeats the purpose.

Then what would be the purpose of the review?
As in who is the intended reader audience?

We already know the product is more complex than the average consumer gateway device. We already know it is fast as all heck. We already know that if there is an issue your only recourse is to seek help via the forums.

It seems to me that a single update paragraph on the original article might be all that is needed rather than a full revamp of the review.

 

[zentec]

Then what would be the purpose of the review?
As in who is the intended reader audience?

We already know the product is more complex than the average consumer gateway device. We already know it is fast as all heck. We already know that if there is an issue your only recourse is to seek help via the forums.

It seems to me that a single update paragraph on the original article might be all that is needed rather than a full revamp of the review.

Networking professionals, people who can roll their own that need better performance than a consumer router can produce?

I don't necessarily disagree with your premise, but I believe that not only has the hardware evolved since the original article (at least the packaging is different), but the 1.6 firmware is comprehensive enough that many people won't need to venture into the Vyatta CLI. Changing the original article may detract from the point that UBNT has actually been continuing development.

 

[Trip]

Also of note are how many larger-scale changes took place from 1.5 to 1.6 (entirely different version of linux, complete hierarchical access to all settings, etc) edgeos is obviously a bit buggy at 1.6.0 (per the UBNT forums) but once they push a few sub-releases, it should stabilize. I for one am looking forward to purchasing a unit or two once that time arrives. And I can't see that taking too much longer than a few months at most.

 

[OP06D]

I would like to add a +1 for an updated review of both the ERL and the Edgerouter Pro/8. Especially for the Edgerouter Pro. In the SNB review, there were issues with routing speed and hardware offloading. I have an Edgerouter 8 (same cpu as Pro) and it appears the issues Doug found in the Pro have been fixed with the 1.6.0 firmware. Many users with gigibit connections are reporting near gigibit routing speed with the Pro/8 and 1.6.0 firmware, as apposed to Doug's sub par 304.1 wan to lan and 553.6 lan to wan numbers. The new wan-2lan2 wizards in 1.6.0 make these routers extremely easy to setup even for the technically challenged. I purchased an ERL and the 8 at the same time. I figured I would just use the ERL as a test device before I deployed the 8. After a few hours with the ERL, it went back in the box and the 8 was is in the rack at work and I haven't looked back since. The ERL has replaced an old CISCO device in my home. Great devices IMHO.