My personal knee jerk reaction to MikroTik at this time… STOP the crap with RouterOS Enterprise. WHY?! What is this drive and business decision to pull development resources to these “enterprise” features baked into RouterOS and or for Routerboard hardware?
The RouterOS Enterprise should be a seperate PRODUCT or operating system. Stop incorporating the linux features into RouterOS “just because its cool”. Like a new developer learning cool new tricks and wants to show them off.
Adding the SMB, NFS, and now iSCSI for storage to RouterOS “Enterprise” is stupid. Leave these to NAS or other products. Why would we as professionals want to have our ROUTERS communicate to network storage devices? This is a security risk of its own.
Focus on ROUTING and WIRELESS improvements and enhancements. Otherwise, fork off “RouterOS Enterprise” to “MikroTik BoxOS or MikroTik EntraOS”
Leave the “RouterOS Enterprise” for what the actual name implies…
We’re slowly getting rubbed the wrong way with MikroTik new products and the quality of the RouterOS releases.
AX = broken
new AX Outdoor no longer has standard 48v 802af/at…
wifi-qcom-ac = broken
SA-timouts.
no CAPsMAN controller sync capability
UserMan is hotmess
More than just wifi. We’ve already started moving to another vendor for AX.
RouterOS7 still lacking refinement of many other requested features [Layer3, MLAG enhancements, etc]. But is “good enough” for majority of our needs. For our larger customers than require enterprise customers, SLA’s and uptime – we use other products.
You’re complaining the cheaper product isn’t keeping up with a more expensive product?
There are some v7 things that need sorting but don’t deploy it if you’re not happy with it or don’t upgrade it if you’re not happy with how it runs.
I think that what was really meant is that the (admittedly cheaper) Mikrotik environment (hardware+software) is not keeping up with some more expensive product (which is expected and OK) but also misses in practice some basic features (reliability of Ax connections is one) that seemingly work just fine in cheaper or equally cheap products, and that this latter is due to mis-directed resources, assigned to add features (and bells and whistles) that more expensive products already have, thus neglecting the resolution of known problems/issues.
I don’t think that it is easy to manage a (vast) software project such as RouterOS and - at the same time - manage the development of the hardware it runs on, but - as seen from the outside - it seems like the priorities are largely mixed up.
The amount of changes in each new release (and the number of bugs - minor or major - that are introduced in each release and hopefully fixed on the next one) are IMHO a sign of a semi-random approach, of course I don’t know what really happens inside Mikrotik, I can only tell what it seems like happening, a bunch of evidently brilliant programmers working on whatever crosses their mind, with no apparent coordination or directives, a lot of throwing things at the wall and see what is sticking.
A few of the issues that are daily reported for newish versions are clearly “niche” and are the reason why public Betas should exist, but most are so “wide” that it is not possible that they managed to get through an appropriate testing/quality control system.
Mind you, even if personally I disagree with quite a few of the directions the development is taking, I wholly respect them, the problem is when the crave for these new, supposedly better, features leads (as said very likely because of lack of resources to do both) to not working or only partially working new features AND not working or only partially working old ones.
I started out writing a long comprehensive list of concerns i had over the direction of RouterOS in the last couple of years, but decided that most wouldn’t really bother reading it
In summary I hope that Mikrotik will understand that most implementations of the CCR/CRS range of devices are not implemented in mass at home or end points and that their role within a network is far more critical than a CPE with all the bells and whistles like Kid Control and media streaming.
We need to go back to having the option to remove all unwanted services from a device, and really expand on the core networking technologies with a possible “core” OS. For example the implementation of BGP in V7 is half baked and lacks critical features such as ECMP, ease of reporting/troublshooting (session data doesn’t update automatically) and inefficient route filtering. Here is a small example of MT capabilites vs Vender X taken from a peer with a route reflector.
Vender X Capabilities: .capabilities=mp,rr,em,gr,as4,ap,err,llgr,fqdn
Mikotik Capabilities: .capabilities=mp,rr,gr,as4
IPv6 implementation is weak with some serious bugs. Rather than focusing limited resources to try get the next feature out, and rather focus on expanding and stabilising the current OS would go a long way.
Its becoming more and more difficult to “trust” devices in key network areas when with each release moves further and further away from core networking and ever closer to a consumer CPE operating system.
Remove the bloat, focus on routing packets, stabilise the operating system for key devices and deliver a routing operating system that lives up to its potential. There are too many “funnies” that don’t make sense and reduce the trust in the product. Implementing configuration changes that dont provide the expected result.
Well said, 100000000% this. MikroTik is completely missing the mark, and losing their values. Hence my comment they need to step this ROSE bullshit. Fork it or pivot to a new operating system that fills this niche or market. Leave these storage protocols OFF the ROUTER products. STUPID… Also a security risk. Their development time is better spent fixing issues and adding real features that we as professionals been asking.
I doubt a lot of resources are spent to make ROSE package. This could be just recompile of well-known linux packages and that’s it, not that much resources wasted. Agree on ax troubles tho. 2 versions no fix.
