Newsletter #131 | January 2026

Read our latest newsletter and learn more about:

  • CRS804 DDQ: a compact wire-speed 4x 400G switch;

  • CAT6 to CAT7 LTE product upgrades, including LtAP LTE7, SXT LTE7, and the Wi-Fi 6 Chateau LTE7 ax;

  • User success stories;

  • Easy one-click container apps in WinBox;

  • RouterOS v7.20.7 [long-term] release;

We are back from the holidays, more exciting hardware news to follow soon!

https://mt.lv/news131

7 Likes

Somebody needs to do some decent proofreading before releasing :rofl:

Could be me but I think the descriptions are placed wrong.

CAT7 LTE kit: don't get me wrong, SXT LTE6 was/is a very good kit but still using MIPSBE for new devices ?

3 Likes

Not only.
Even if one cannot distinguish a SFP cage from an ethernet port, the left image has only 2 items in it, it should be more like:

1 Like

I expected some mentions to hap be in this month's newsletter :open_mouth:.

PDF link because it’s faster

1 Like

This:

R11e-LTE7
...
Drop-in replacement for FG621-EA – now offering 300 Mbps down / 100 Mbps up, with added Band 32 & Band 28 support.

might be more relevant than what it seems at first sight, basically it should mean that these devices:

Friendly Name Model Modem inside
Chateau LTE6 D53G-5HacD2HnD-TC&FG621-EA FG621-EA
Chateau LTE6 ax S53UG+5HaxD2HaxD-TC&FG621-EA FG621-EA
hAP ax lite LTE6 L41G-2axD&FG621-EA FG621-EA
LtAP LTE6 kit LtAP-2HnD&FG621-EA FG621-EA
SXT LTE6 kit SXTR&FG621-EA FG621-EA

See:

can be easily upgraded (and hopefully this new modem card will support e-sim)

Isn’t it a bit quick after release of 7.20.7 to write “Still on v6? RouterOS 7.20.7 long-term
release marks the right time to move to v7!” ?

At least read the forum topic before doing so…

1 Like

I like the "half-rack" form factors, so +1 on that.

But how about a "drop-in upgrade" with the North American LTE bands? There is not a single LTE/5G with Band 2/4/12/etc., beyond the now slower and older devices. It's been years now.

it’s not like it dropped from the sky. that branch has been there for a long time.

Yes true, but as many times there is a stable version (7.20.6) and then it is released as a long-term version but some quick fixes are added (see 7.20.7 release notes). That version has not been widely tested in the field, and people find new issues that were maybe caused by the quick fixes.

It seems better to first release a 7.20.7 as “stable”, wait a couple of weeks, and then release the same version (no changes except the version name) as “long-term”. People expect a long-term version to be a riskless upgrade, although of course it has been discussed many times that these names refer to maintenance/release cycles, not to stability or bugs. You will not get that changed.

proof?

Well, people report a problem in the 7.20.7 release topic and claim it works on 7.20.6.

It may be caused by specific config of course, but I do not think that would have happened when 7.20.6 had been released as long-term.

Sometimes even minor changes can cause issues, even if it is a bug fix that changes behavior that people relied on. For example (I don’t think this change is in 7.20.7) I had to do quite some work to solve an issue caused by the seemingly innocent “ippool6 - take into account "subnet-id" when specified on address;” change in 7.21/7.22. I opened SUP-208174 for it, but to keep my network usable wihout downgrading I had to redesign my address layout, change firewall settings, update DNS entries at my provider, and re-configure tunnels at some peers. While that change would normally have been only an improvement in stability.

it’s not proof though. no software is bug-proof, but the best thing there is, is the long-term release. maybe there is a specific combination of configuration that causes a problem for one person out of a million. that person will come to the forum. it is not an indication of widespread problems.

I don’t claim that. But what will likely happen is a lot of users will upgrade either within v7 or from v6 because of that announcement in the newsletter, and there will be more postings about problems they experience. It would maybe have been better to save part of that by using a version that is a bit more tested.

I think the main point behind long-term is the longer support period. Which should mean that point releases that fix only particular bugs without introducing new features are issued.

This decreases (not eliminates) the chance of breakage. But what is much more important: if something does need patching, a new lts point version will be released instead of pointing users to upgrade to the next version that contains the fix, but along with other modifications.

Altogether, I think this is a nice sort of commitment.

As a side-side note, the article linked to on varia.org:
https://www.varia.org/en/success-story-chicken-with-mikrotik/

is worth a read.

It can be used as a reference exercise in both English and Network classes, it could be used as an example of the "most gratuitous use of coniferous and deciduous" :face_with_raised_eyebrow: and to practically teach the difficulties in precisely aligning two 60° beam wide wireless devices at 50 m distance. :astonished_face:

Apart from the above it seems to me a breakthrough in ethology, clearly foxes in the North are much more educated than the common ones we have here in Southern Europe, as:

The visible camera surveillance has already had a deterrent effect on wild animals ...

(our ones can be sometimes scared with firing a gun, nothing less) :rofl:

What is left unanswered is the procedures to be implemented WHEN the motion sensor cameras actually detect a fox, what do you suggest?

  1. dress yourself, put your boots on, grab the gun, load it, run with the loaded gun through 50 meters of woods, get to the barn by the time the fox is already at your neighbour's chicken coop or anyway not less than 1 Km away [1]
  2. watch the video of the fox stealing your poultry
  3. trigger the firing of the decent amount of (AI driven) machine guns you placed covering the outside of the fence
  4. something else ...

[1] a quick brown :wink: fox attack usually lasts 5-10 seconds, maybe a particularly slow one, will take 15-20 seconds to get the hen and run away

1 Like

Where is the LtAP ax ? I remember a while ago at an exhibition, Mikrotik showcased a LtAP ax , which has 2 nano sim slot and 3 ethernet port, and support eSIM. Waiting for too long for that product !

yes, that’s the main idea. only critical fixes, no new features. this way, it can only get more stable.

Firstly, the newlsetter has not been ported to the Home Web page of Mikrotik. It still shows DEC!!
Secondly, I had to look to find it, whereas it should be prominently displayed on the first page ( meaning the latest newsletter). Please fix.

Whilst an upgrade is always welcomed, why only CAT7 and not CAT12 as offered by the competition. Cost?