I'm scared that also RB3011UiAS-2HnD version will be also bugged in first series.. This is total fiasco from MT. Customers wanted powerful units, but get unfinished and bugged devices.

My router is already in production and working good, but not 2hnd.

This is total fiasco from MT. Customers wanted powerful units, but get unfinished and bugged devices.

Maybe you want to tell something like :

I'm scared that also RB3011UiAS-2HnD version could be bugged in first series.. This could be a total fiasco from MT. Customers wants powerful units, not unfinished and bugged devices.

Most of the time, software, not hardware, is buggy. Also, all competitors have the same problems.

Dear Mikrotik,

I did monitor each CPU using System->Resources->CPU and it [flash=]seems [/flash]that all the trafic is handled by one of the CPUs, during my tests it was always the second,....1 , 2, 3 or 4 tcp streams..it does not care: always one CPU appears (really) busy while the other always< 5% of usage.

Testing was done over a 300/30 Mbits fiber connection via PPOE and "Nated", and TCP streams between a server in Internet and a PC in the internal LAN.

It does not care what port is configured as a gateway: I tried the first on each switch chip , so the 1 and the 6.
A single 300Mbits TCP input stream (via PPOE and NAT) needs about 82% of one of the CPUs...curiously always the same used.
It does not care if fasttrack is enabled or not...

Besides:
- Partition not working in RouterOS 6.33.3.
- Some strange behaviour in the LCD in RouterOS 6.33.3
- That after upgrading to 6.34rc20...I lost access to the device and needed to "netinstall": not enable to login via Winbox, web or SSH....sympthom was User/passsword correct but device thinking a lot to finally deliver an error message.
- The CPU heatsink is really hot while RouterOS reports temperature only as 34º C.
- Intermitent FC errors between ISP GPON and RB3011: I need to check this because it can be not related to the RB3011 (strange...never on a RB2011 nor an ERLITE 3).

The device works...SOHO configuration with 300/30 Mbits over fiber (external GPON from ISP), with standard firewall and NAT, using 3 VLAN in the ISP side (Internet, IPTV and IPPhone) and 1 port in the internal side (because Mikrotik ignores IGMP snooping on their devices) conected to a managed switch.

I'll wait for further refinement of this device, specially on anything related to CPU tasks/threads assignement before going forward with it, and in the meantime...crossing my fingers about hardware reliability (because the experience on RB2011 revisions, RB850x2 revisions...)

cpu-used: 12%
cpu-used-per-cpu: 13%,12%
free-memory: 996832KiB

Of course there is no loop, but 3011 don't believe me:

Oppsss.....it's so bad so see Microtik silence on RB/3011....Sorry for that, but its does not make sense.

And yes I known...someone will say that when Microtik will have something to say, he willl....but perhaps Mikrotik will not find me more...to hear them because too late.


KInd Regards

Will be keeping tabs on this thread.

RB2011 got off to a rough start and got much better as the software kept maturing.This looks like more or less the same kinda thing.

its enteriely New device for company that Never ever done ARM-based things before so some issues - was expected and reasonable to be.
some of aleady-noticed "imperfections" likely be fixed later i guess.

cpu-used: 12%
cpu-used-per-cpu: 13%,12%
free-memory: 996832KiB

Of course there is no loop, but 3011 don't believe me:

did you send this to support? or only the forum?

I have one RB3011UiAS-RM and RB850Gx2, detailed review coming soon.

I have one RB3011UiAS-RM

Could you please do me a favour and check if your RB3011 clock is working fine? Mine looses about 80 seconds per hour and NTP syncing is not working (with ntp package).

Could you please do me a favour and check if your RB3011 clock is working fine? Mine looses about 80 seconds per hour and NTP syncing is not working (with ntp package).

I have the same problem on rb3011, clock not work correctly.

Could you please do me a favour and check if your RB3011 clock is working fine? Mine looses about 80 seconds per hour and NTP syncing is not working (with ntp package).

I have the same problem on rb3011, clock not work correctly.

near NY i had similar issues on some of my MIPSBE devices aswell(both on stable or RC branches of ROS)
RB's w/o NTP packages - seems affected less.
now its ~ okay on 100% of them. weird.

Same problem with clock.

First part of review RB3011
http://lanmarket.ua/stats/obzor-marshru ... 011UiAS-RM

Part 2, RB3011 vs RB850Gx2
http://lanmarket.ua/stats/testirovanie- ... 011UiAS-RM

Part 2, RB3011 vs RB850Gx2
http://lanmarket.ua/stats/testirovanie- ... 011UiAS-RM

Thanks for this. Google translate to English is really bad but I get a good idea of the review. Thanks again!

Tried to deploy one today, managed to hit 30 mbit throughput -- after replacing with 2011, I've got close to 190 mbit/s with exactly the same configuration. Log was full of some loop/latency problems on 3011. Will try again in few months, and now I'll replace the 2011 with some CCR, since I need gigabit here.

Ooppss *) pppoe-client - implemented fastpath support.... ????
it works over RB2011 but it does NOT works over RB3011.

Same fiber connection (300/30 Mbits) and RouterOS configuration testing RB2011 and RB3011, with these results:
- RB2011...6.34.2 - CPU usage 100% (298 Mits)
- RB2011...6.35rc12 - CPU usage between 32-38% (305 Mbits)
- RB3011...6.35rc11 and sooner - CPU usage TOTAL: 45%, ONE CPU 80-90%, the OTHER CPU 2-3% (305 Mbits)
- RB3011...6.35rc12 - CPU usage TOTAL: 45%, ONE CPU 80-90%, the OTHER CPU 2-3% (305 Mbits)

Symthom: FP RX counter for PPOE client conection is always zero in RB3011 (not for FP TX...)

FYI...

rb3011 don´t support fasttrack

rb3011 don´t support fasttrack

Sorry, RB3011 don't support fastpath too.

rb3011 don´t support fasttrack

Sorry, RB3011 don't support fastpath too.

Why not ?
Why MIkrotik announces (still in the MUM) what is only partially supported?
Why ARM platform to run non optimized nor finished software?

cpu-used: 12%
cpu-used-per-cpu: 13%,12%
free-memory: 996832KiB

Of course there is no loop, but 3011 don't believe me:

You are not alone! Check out my thread http://forum.mikrotik.com/viewtopic.php ... ead#unread

I was told the recent RC would fix it but it has made no difference at all. I am not happy as I need software bridge for SFP and I currently can't use any software bridge without the dreaded error logs.


MT support have been silent once provided with supout files confirming problem.

First of all, RB3011 does support fastpath. Counters are going quite nicely.
I found CPU load distribution to be tricky. You need many connections to distribute it properly.

As far as i understood from communication with support each switch-chip is connected with 1Gbps line to each CPU,
So if switch-chip decides to take path1, CPU will have to process it with core1, it switch-chip decides to take path2, then core2.
and it doesn't matter how high the load is on CPU cores atm.

Hello,
The RB3011 block diagram has been updated to better represent internal connections among CPU cores.

Hello,
The RB3011 block diagram has been updated to better represent internal connections among CPU cores.

Any comment on the looping problem? I am getting nothing from my support ticket.

First part of review RB3011
http://lanmarket.ua/stats/obzor-marshru ... 011UiAS-RM

Do you have an English translation? Google won't translate the page.

I understand that is the switch chip, not the CPU.

First part of review RB3011
http://lanmarket.ua/stats/obzor-marshru ... 011UiAS-RM

Do you have an English translation? Google won't translate the page.

Try microsofttranslator
http://www.microsofttranslator.com/bv.a ... 011UiAS-RM

Try microsofttranslator
http://www.microsofttranslator.com/bv.a ... 011UiAS-RM

Thank you! Did not know about that one.

Why ARM platform to run non optimized nor finished software?

Tilera/EZchip/Mellanox: TILE->ARM
Freescale/NXP: PPC->ARM
Applied Micro Circuits: PPC->ARM
Atheros/Qualcomm: MIPS->ARM

Allow me to resuscitate this topic

Am I right to assume that if I have a fiber uplink to the LAN and WAN on the first switch chip the router will just use the CPU1 ??
I am not sure if SFP will connect directly to CPU1 or to the second Switch chip?

By the way, how does the router decide which CPU to use? Is it based on some kind of round-robin ?

Regards.

usually its choosen magic-driven dice roll, touched by libastral.so device/stub.
and sheduling(let alone balancing or parking) generally its tricky topic even in multi-core, let alone many-core platforms. several patents hold on-topic, several research stuff still ongoing.
and since ARM percentage/penetration even in networking keep skyrocketing - appropriate toolchain soon become mission-critical. thats why so many efforts(from countless companies) in upstream of linux do happens howdays. even legacy (and not produced anymore)Arm versions keep updated relavtively slowly among LTS stuff.