v6.41.2 [current]

Mon Feb 12, 2018 4:08 pm

I found a couple of strange entries in the routing table after the upgrade on my hAP ac.

What exactly is considered strange there?

mszru · Mon Feb 12, 2018 4:41 pm

I found a couple of strange entries in the routing table after the upgrade on my hAP ac.
What exactly is considered strange there?

The new routes to 0.0.0.0/32 in the list which were not there before.
According to the RFC 3330 it is kind of special address, and I cannot understand if it was added on purpose or by mistake during the upgrade.

Mon Feb 12, 2018 5:02 pm

Can you generate supout file from 6.41.2 downgrade and generate another file from older version and send both files to support?

fabifri · Mon Feb 12, 2018 5:10 pm

Boot loop on RB750Gr3 too.

Datacenter offline. Night shift incomming. Thanks!

Rule47 · Mon Feb 12, 2018 10:24 pm

I've got trouble connecting older devices such as Android 4.0, iPhone 4 or Blackberry probably since Rel. 6.41 to access points using CAPsMAN. I downgraded to Rel. 6.35.4, which works well with these devices. Newer devices work (tested: iPhone 6, iPhone

.

Currently installed on all devices: Rel. 6.41.2. The error I get is: "received deauth: class 3 frame received (7)".

Does someone have a clue, what to do. Users are complaining heavily.

Jo

Jotne · Tue Feb 13, 2018 1:11 am

Boot loop on RB750Gr3 too.
Datacenter offline. Night shift incomming. Thanks!

RB750Gr3 is a low cost product ment for home and small offices.
Running a whole datasenter behind it seems for me some strange.

i4ko · Tue Feb 13, 2018 6:18 am

After upgrading from version 6.39.3 [BugFix] to 6.41.1 and then to 6.41.2, my device [RB751U-2HnD] presented wireless connectivity problems.

I also upgraded the firmware, leaving it equivalent to the ROS.

The causes of the problem could not be identified in the logs.

Even after resetting and redoing all the settings one by one, it was not possible to keep the wireless connection stable.
Please write more detailed what kind of wireless connectivity problems you see.

I have seen a wireless issue on two of the RB941-2nD I have. I don't know if it is the same or not.
Initially it started with wireless devices intermittently dropping. Later on it progressed to more severe intermittent connectivity - about 50 percent of the time there will be no devices connected. Logs would show key exchange timeout and drop the wireless devices about every 30 minutes, then it changed to about every 2 minutes. It progressed further onto the wireless devices not being able to hold stable connection for more than 2-3 seconds, it would not matter if the frequency was set to auto or set frequency. Initially devices tried to connect frequently, then connection attempt became more rare, until almost no connection attempts would be made. Every time the router kicked out the device for unicast key exchange timeout. But it became even worse - the devices were not trying to connect because they could not see the SSID. Those two devices now (no longer at customer) do not advertise SSID no matter what, I can't see it even with kismet. It seems like the radio RF portion just gradually and eventually died after the 6.41.1 upgrade. Multiple resets, and power off/power on did not make any difference.
Needless to say I had to replace these two. Initially they did not have the 6.41.1 firmware installed when they started being weird. When I got the issues, I installed the 6.41.1 firmware which made things worse and ultimately the WIFI portion just does not seem to work at all. The whole degradation from a fully stable working router to fully non-discoverable wifi (even on scans) started about a month after the 6.41.0 upgrade, and tool like 3-4 days with a 6.41.1 upgrade in the middle.
Regular AP bridge, no CAPsMAN, 2GHz-N-only band, 20Mhz channel, 802.11 protocol, WPS disabled, Hide SSID not set. Frequency - regulatory domain, Country - United States, Data rates - Default, Distance - indoors, Preamble - long/both

Hmm, tried setting the country to something else - Latvia - it would make the SSID temporarily discoverable (for about 5-10 seconds) and then it will die again. At the same time the status will claim it is running, at -112dBm noise floor and 2427/20/gn(20dBm) channel (sometimes 2422).
Nothing informative in the logs, even if wireless is on debug

Clauu · Tue Feb 13, 2018 8:18 am

About firewall new feature 'tls-host' i cannot make use of it no matter what sites i try, i'm i missing something or this feature is broken?

/ip firewall filter add chain=forward dst-port=443 protocol=tcp tls-host=*google.com action=drop

Tue Feb 13, 2018 8:35 am

Clauu - Such rule should be working. Most likely some other firewall rule matches such traffic before this one. Are you using FastTrack?

Clauu · Tue Feb 13, 2018 8:53 am

Hi strods, i'm not using fastrack and tls rule was in top of all other forward chain rules.
Hm as i can see there is a fix for this feature in latest rc

What's new in 6.42rc24 (2018-Feb-08 09:42):
*) firewall - fixed "tls-host" firewall feature (introduced v6.41);

Is this related to my issue?
L.E: tested again and now it's working.. odd.

tangram · Tue Feb 13, 2018 4:43 pm

Hi,

I don't know if it's a winbox or routeros bug, but it's easy to reproduce. Connected with winbox, if you have a script open for editing and you delete that script from script list the session is dropped.
Oh and it generates autosupout

Tue Feb 13, 2018 4:46 pm

Do you use latest winbox version?

ksteink · Tue Feb 13, 2018 9:42 pm

I upgraded my RB2011UiAS-2HnD-IN with no major issues. All my basic connection features worked fine (OVPN, IPv6 and VLANs). I did noticed that after the upgrade the LED on the router was constantly ON even in my setting I have it by default on OFF. I did turned ON and OFF again and it finally turn it off.

I have been working all day with it and no issues. I still want to revisit the configuration changes on the HW off-loading but from the end user experience no services have been impacted.

aaronw · Wed Feb 14, 2018 2:47 am

Never mind. I switched to a different laptop and disabled the firewall and I was able to recover the system. I still do not know why the initial update failed. Using netinstall I was able to boot the new system and install it.

This update appears to have broken an RB850GX2 router. After the update the router never recovered and failed to boot. I have since tried to get it to boot using netinstall and that fails.

Here is the output when booting from NAND:
RouterBOOT booter 3.24

RouterBoard 850Gx2

CPU frequency: 533 MHz
Memory size: 512 MiB
NAND size: 512 MiB

Press any key within 2 seconds to enter setup..

loading kernel from nand... OK
setting up elf image... OK
jumping to kernel code
Starting...
Starting services...
connect failed (errno=2)
died with signal 6 on Thu Jan 1 00:00:12 1970

Aborted
connect failed (errno=2)
Aborted

Here is the output when I attempt to use netinstall and boot over Ethernet:
RouterBOOT booter 3.24

RouterBoard 850Gx2

CPU frequency: 533 MHz
Memory size: 512 MiB
NAND size: 512 MiB

Press any key within 2 seconds to enter setup..
trying bootp protocol.... OK
Got IP address: 192.168.0.100
resolved mac address 3C:97:0E:24:41:BC
**************************************************************************************************************
**************************************************************************************************************
**************************************************************************************************************
**************************************************************************************************************
**************************************************************************************************************
**************************************************************************************************************
**************************************************************************************************************
**************************************************************************************************************
**************************************************************************************************************
**************************************************************************************************************
**************************************************************************************************************
**************************************************************************************************************
**************************************************************************************************************
**************************************************************************************************************
**************************************************************************************************************
**************************************************************************************************************
**************************************************************************************************************
**************************************************************************************************************
********************falling back to small packet size
**************************************************************************************************************
**************************************************************************************************************
**************************************************************************************************************
**************************************************************************************************************
**************************************************************************************************************
**************************************************************************************************************
**************************************************************************************************************
**************************************************************************************************************
**************************************************************************************************************
**************************************************************************************************************
**************************************************************************************************************
**************************************************************************************************************
**************************************************************************************************************
**************************************************************************************************************
**************************************************************************************************************
**************************************************************************************************************
**************************************************************************************************************
**************************************************************************************************************
********************kernel loading failed
Basically it bricked the Routerboard. I have also cleared the configuration from the bootloader and that did not help. It seems to pass the memory test.

Everything was working fine until the update. The update process completed then the box showing the progress just hung (obviously because it could not reboot).

Any suggestions?

-Aaron

Wed Feb 14, 2018 6:35 am

Seems that either there is some issue with link between computer and router or there is another device in the path (switch, for example) causing this problem. Router starts to communicate with Netinstall, however, they both are not being able to make stable connection in order to start Netinstall process.

i4ko · Wed Feb 14, 2018 6:46 am

i4ko - Most likely you will need to re-install devices by using Netinstall. However, I recommend that you send supout file from one of your routers to support. We might see in supout file what is filling up the space.

Finally manged to update remotely. While on 6.41.1 disabled all packages except for system, dhcp and wireless, and uploaded the same from the 6.41.2 all packages file to the router (those were all i had room for), then rebooted, all in one go. Now I no longer have any packages besides these (routeros-smips is also gone) but everything is working fine. This also dropped 1.5mb in disk usage, I'll add some of the other packages where I need them.

osmoticzest · Wed Feb 14, 2018 9:35 am

Spotted the following on Superuser today, and it accurately describes what I’m seeing on the last couple of releases (potentially since 6.41.0) – the WiFi interface will just ‘stop’ at random, nothing in registration table, 0/0 tx/rx. Restarting the interface or doing something like using the ‘Scanner’ will bring things back to life. Same behaviour on both APs here.

https://superuser.com/questions/1293579 ... -is-restar

951G-2HnD + RB941-2nD

frequency-mode=regulatory-domain country=united kingdom

Nuubo · Wed Feb 14, 2018 9:37 am

Unfortunately my fears were true: bridging is still working bad because ARP is not fixed.
I am wondering why did Mikrotik eliminate master/slave options using switching?, since I am using Mikrotik I just use bridges to link wifi / ethernet interfaces but when the other day I upgraded to 6.41.2 it was impossible to keep the network running, the old master/slave interfaces of same switch were wrong, the old master interface loosing packets continuosly and the slave interface just not populate MAC addreses connected.

Please, consider my option: back to the old status, master slave is a much better option using switiching, if someone wants bridging, use bridging, but I think it is much worse.

Thanks.

tangram · Wed Feb 14, 2018 10:20 am

Do you use latest winbox version?

Thank you, I was using 3.11. In 3.12 problem is fixed.

R1st0 · Wed Feb 14, 2018 10:23 am

I have problem with neighbor discovery on CRS106. My intention is to allow neighbor discovery only on uplink port (sfp5).
When using configuration:
/interface list
add name=clients
/interface list member
add interface=sfp1 list=clients
add interface=sfp2 list=clients
add interface=sfp3 list=clients
add interface=sfp4 list=clients
/ip neighbor discovery-settings
set discover-interface-list=!clients
It works, but when exporting the config "!" mark disappears: /ip neighbor discovery-settings set discover-interface-list=clients. When I use "/ip neighbor discovery-settings pr" the "!" is there.
To reduce lines I tried similar configuration, but this one doesn´t work (no discovery on sfp5):
/interface list
add name=uplink
/interface list member
add interface=sfp5 list=uplink
/ip neighbor discovery-settings
set discover-interface-list=uplink

eddieb · Wed Feb 14, 2018 2:05 pm

updated my RB2011UAS from 6.40.5 to 6.41.2 ...
everything went smooth.
2 IPSEC site-to-site tunnels
DHCP and DNS server
PPPOE on wan

look fine here

LIV2 · Thu Feb 15, 2018 2:29 am

After upgrading my mAP2n to 6.41.2 I've found that doing /system reset-configuration no-defaults=no results in an empty configuration, and I can only connect to the router via mac-telnet
Same result when using the reset button to do a config reset, even on 6.42rc24 it also doesn't work

Thu Feb 15, 2018 6:52 am

osmoticzest - Provide supout file from your router to support@mikrotik.com
Nuubo - Please provide precise problem description and supout files to support@mikrotik.com. From which version with what exact configuration did you upgrade this device and did you upgrade directly to 6.41.2 release;
R1st0 - We will resolve this export command problem in upcoming RouterOS releases;
LIV2 - Have you tried to Netinstall your device and reset it only after that? Seems that default configuration has been changed to something else by using configuration script on Netinstall.

LIV2 · Thu Feb 15, 2018 9:56 am

LIV2 - Have you tried to Netinstall your device and reset it only after that? Seems that default configuration has been changed to something else by using configuration script on Netinstall.

I've tried Netinstall ensuring "Apply Default Config" is ticked with various versions with the following results
6.42rc20 - No default config
6.41.1 - No default config
6.41 - No default config
6.40.5 - Default config works
6.39.3 - Default config works

Nuubo · Fri Feb 16, 2018 10:50 am

Nuubo - Please provide precise problem description and supout files to support@mikrotik.com. From which version with what exact configuration did you upgrade this device and did you upgrade directly to 6.41.2 release;

I am sorry but I did the change to my configuration with a new subnet just for one computer. I thought it was some upgrade problem (I did it from 6.41 directly) I deleted the bridge and I recreated it with same results, there was no MAC address in ARP table and continuous interruptions of network traffic. I just want ask you in future releases the recover of master/slave option, I can not understand why did you eliminate the use of switching, it is much better since it is flexible and instant.

buzzdee · Fri Feb 16, 2018 7:24 pm

Hi,

seems the issue described here

viewtopic.php?f=21&t=128915&p=637801&hi ... em#p637801

still persists. When enabling/disabling ppp-devices from terminal or script, I get "no such item (4)" the first time I run the command, but it works when I run it again. Have a couple of 411U's and 912's with 3G modems, would be nice if I could use my scripts to reset those modems would work again.

macsrwe · Fri Feb 16, 2018 10:57 pm

Again - we strongly recommend to NOT use this feature if you are not aware about how it works. If you will not know min/max settings, then you will not be able to recover device in any way.

I suspect there are many routers out there where this value is set to arbitrary garbage due to recent (still current?) bug in Winbox where calling up /system routerboard settings always populates this value as 00:00:00 -- in red since zero is an invalid value -- and then demands the user set it to something nonzero before they can make the unrelated change they originally wished to make.

Sat Feb 17, 2018 10:58 am

buzzdee - First of all, please do not post version u relafed issues within new release topic. Your problem has nothing to do with release and even ia not a bug. Do not use ID in scripts or run print first.

For example:
1) will not work
/interface ethernet set 1 name=test
2) will work
/interface ethernet set ether2 name=test
3) will work
/interface ethernet print
/interface ethernet set 1 name

On #1 script does not know which interface is 1 and that is why you see such error.

buzzdee · Sat Feb 17, 2018 11:48 am

Hi strods, thanks for the quick reply.

First of all, please do not post version u relafed issues within new release topic

Sorry, I should have been more precise - what I experienced was on 6.41.2.

Heres is a snippet from the script. I'm not using ID's but the interfaces name:

# Set UMTS Interface name
:global WanName "ppp-out1"
# get its ID number #
:local WanID [/interface ppp-client find name=$WanName];
# Change the data channel setting
/interface ppp-client set data-channel="3" $WanID

If I get you right, interface should come before set-command in the last line:

/interface ppp-client $WanID set data-channel="3"

Does this also apply when using interface names instead of ID's?

Thanks, BuzzDee.

sindy · Sat Feb 17, 2018 12:26 pm

If I get you right, interface should come before set-command in the last line:
Code: Select all
/interface ppp-client $WanID set data-channel="3"
Does this also apply when using interface names instead of ID's?

Thanks, BuzzDee.

No, the order is less obvious, 1. set 2. in which item 3. which attribute:

/interface ppp-client set $WanID data-channel="3"

You can abbreviate your example into

/interface ppp-client set [find name=$WanName] data-channel="3"

And yes, you can replace the number(s) in the set, remove, unset etc. by "find" referring to any attribute.

Institor · Sat Feb 17, 2018 1:38 pm

Bridge on RB750Gr3 (6.41.2) seems to be broken.
1) No HW offload on ethernet interfaces after upgrade from 6.41.1:

Flags: X - disabled, I - inactive, D - dynamic, H - hw-offload
 #     INTERFACE                      BRIDGE                      HW  PVID PRIORITY  PATH-COST INTERNAL-PATH-COST    HORIZON
 0     eth2                           bridge_local                yes    1     0x80         10                 10       none
 1     eth3                           bridge_local                yes    1     0x80         10                 10       none
 2 I   eth4                           bridge_local                yes    1     0x80         10                 10       none
 3 I   eth5                           bridge_local                yes    1     0x80         10                 10       none
 
interface bridge port print detail
Flags: X - disabled, I - inactive, D - dynamic, H - hw-offload
 0     interface=eth2 bridge=bridge_local priority=0x80 path-cost=10 internal-path-cost=10 edge=auto point-to-point=auto
       external-fdb=auto horizon=none hw=yes auto-isolate=no restricted-role=no restricted-tcn=no pvid=1
       frame-types=admit-all ingress-filtering=no

 1     interface=eth3 bridge=bridge_local priority=0x80 path-cost=10 internal-path-cost=10 edge=auto point-to-point=auto
       external-fdb=auto horizon=none hw=yes auto-isolate=no restricted-role=no restricted-tcn=no pvid=1
       frame-types=admit-all ingress-filtering=no

 2 I   interface=eth4 bridge=bridge_local priority=0x80 path-cost=10 internal-path-cost=10 edge=auto point-to-point=auto
       external-fdb=auto horizon=none hw=yes auto-isolate=no restricted-role=no restricted-tcn=no pvid=1
       frame-types=admit-all ingress-filtering=no

 3 I   interface=eth5 bridge=bridge_local priority=0x80 path-cost=10 internal-path-cost=10 edge=auto point-to-point=auto
       external-fdb=auto horizon=none hw=yes auto-isolate=no restricted-role=no restricted-tcn=no pvid=1
       frame-types=admit-all ingress-filtering=no

2) Bridge settings freezes constantly (for example, after trying to check/uncheck "igmp snooping", bridge completely shuts down and remain disabled for approx. 1 minute).

It's slightly annoying....

GARCIADOEGOGERMAN · Sat Feb 17, 2018 5:37 pm

Bridge on RB750Gr3 (6.41.2) seems to be broken.
1) No HW offload on ethernet interfaces after upgrade from 6.41.1:

Flags: X - disabled, I - inactive, D - dynamic, H - hw-offload
 #     INTERFACE                      BRIDGE                      HW  PVID PRIORITY  PATH-COST INTERNAL-PATH-COST    HORIZON
 0     eth2                           bridge_local                yes    1     0x80         10                 10       none
 1     eth3                           bridge_local                yes    1     0x80         10                 10       none
 2 I   eth4                           bridge_local                yes    1     0x80         10                 10       none
 3 I   eth5                           bridge_local                yes    1     0x80         10                 10       none
 
interface bridge port print detail
Flags: X - disabled, I - inactive, D - dynamic, H - hw-offload
 0     interface=eth2 bridge=bridge_local priority=0x80 path-cost=10 internal-path-cost=10 edge=auto point-to-point=auto
       external-fdb=auto horizon=none hw=yes auto-isolate=no restricted-role=no restricted-tcn=no pvid=1
       frame-types=admit-all ingress-filtering=no

 1     interface=eth3 bridge=bridge_local priority=0x80 path-cost=10 internal-path-cost=10 edge=auto point-to-point=auto
       external-fdb=auto horizon=none hw=yes auto-isolate=no restricted-role=no restricted-tcn=no pvid=1
       frame-types=admit-all ingress-filtering=no

 2 I   interface=eth4 bridge=bridge_local priority=0x80 path-cost=10 internal-path-cost=10 edge=auto point-to-point=auto
       external-fdb=auto horizon=none hw=yes auto-isolate=no restricted-role=no restricted-tcn=no pvid=1
       frame-types=admit-all ingress-filtering=no

 3 I   interface=eth5 bridge=bridge_local priority=0x80 path-cost=10 internal-path-cost=10 edge=auto point-to-point=auto
       external-fdb=auto horizon=none hw=yes auto-isolate=no restricted-role=no restricted-tcn=no pvid=1
       frame-types=admit-all ingress-filtering=no

2) Bridge settings freezes constantly (for example, after trying to check/uncheck "igmp snooping", bridge completely shuts down and remain disabled for approx. 1 minute).

It's slightly annoying....

In my RB750GR3 the same thing happens ... looking for more info I found this that seems to be the reason why what happens is what happens.

Release 6.41.1 2018-02-02
What's new in 6.41.1 (2018-Jan-30 10:26):

*) bridge - fixed hw-offload disabling for Mediatek and Realtek switches when STP / RSTP configured;
*) bridge - fixed hw-offload disabling when adding a port with "horizon" set;

This will have to do with what you mention ???

Regards!!!

mpadmin · Sat Feb 17, 2018 10:32 pm

Bridge on RB750Gr3 (6.41.2) seems to be broken.
1) No HW offload on ethernet interfaces after upgrade from 6.41.1:

Flags: X - disabled, I - inactive, D - dynamic, H - hw-offload
 #     INTERFACE                      BRIDGE                      HW  PVID PRIORITY  PATH-COST INTERNAL-PATH-COST    HORIZON
 0     eth2                           bridge_local                yes    1     0x80         10                 10       none
 1     eth3                           bridge_local                yes    1     0x80         10                 10       none
 2 I   eth4                           bridge_local                yes    1     0x80         10                 10       none
 3 I   eth5                           bridge_local                yes    1     0x80         10                 10       none
 
interface bridge port print detail
Flags: X - disabled, I - inactive, D - dynamic, H - hw-offload
 0     interface=eth2 bridge=bridge_local priority=0x80 path-cost=10 internal-path-cost=10 edge=auto point-to-point=auto
       external-fdb=auto horizon=none hw=yes auto-isolate=no restricted-role=no restricted-tcn=no pvid=1
       frame-types=admit-all ingress-filtering=no

 1     interface=eth3 bridge=bridge_local priority=0x80 path-cost=10 internal-path-cost=10 edge=auto point-to-point=auto
       external-fdb=auto horizon=none hw=yes auto-isolate=no restricted-role=no restricted-tcn=no pvid=1
       frame-types=admit-all ingress-filtering=no

 2 I   interface=eth4 bridge=bridge_local priority=0x80 path-cost=10 internal-path-cost=10 edge=auto point-to-point=auto
       external-fdb=auto horizon=none hw=yes auto-isolate=no restricted-role=no restricted-tcn=no pvid=1
       frame-types=admit-all ingress-filtering=no

 3 I   interface=eth5 bridge=bridge_local priority=0x80 path-cost=10 internal-path-cost=10 edge=auto point-to-point=auto
       external-fdb=auto horizon=none hw=yes auto-isolate=no restricted-role=no restricted-tcn=no pvid=1
       frame-types=admit-all ingress-filtering=no

2) Bridge settings freezes constantly (for example, after trying to check/uncheck "igmp snooping", bridge completely shuts down and remain disabled for approx. 1 minute).

It's slightly annoying....

I have 750Gr3 and HW offload works, but you need to set Protocol Mode to "none".

/interface bridge
add name=bridge-internal protocol-mode=none

GARCIADOEGOGERMAN · Sun Feb 18, 2018 2:53 am

Bridge on RB750Gr3 (6.41.2) seems to be broken.
1) No HW offload on ethernet interfaces after upgrade from 6.41.1:

Flags: X - disabled, I - inactive, D - dynamic, H - hw-offload
 #     INTERFACE                      BRIDGE                      HW  PVID PRIORITY  PATH-COST INTERNAL-PATH-COST    HORIZON
 0     eth2                           bridge_local                yes    1     0x80         10                 10       none
 1     eth3                           bridge_local                yes    1     0x80         10                 10       none
 2 I   eth4                           bridge_local                yes    1     0x80         10                 10       none
 3 I   eth5                           bridge_local                yes    1     0x80         10                 10       none
 
interface bridge port print detail
Flags: X - disabled, I - inactive, D - dynamic, H - hw-offload
 0     interface=eth2 bridge=bridge_local priority=0x80 path-cost=10 internal-path-cost=10 edge=auto point-to-point=auto
       external-fdb=auto horizon=none hw=yes auto-isolate=no restricted-role=no restricted-tcn=no pvid=1
       frame-types=admit-all ingress-filtering=no

 1     interface=eth3 bridge=bridge_local priority=0x80 path-cost=10 internal-path-cost=10 edge=auto point-to-point=auto
       external-fdb=auto horizon=none hw=yes auto-isolate=no restricted-role=no restricted-tcn=no pvid=1
       frame-types=admit-all ingress-filtering=no

 2 I   interface=eth4 bridge=bridge_local priority=0x80 path-cost=10 internal-path-cost=10 edge=auto point-to-point=auto
       external-fdb=auto horizon=none hw=yes auto-isolate=no restricted-role=no restricted-tcn=no pvid=1
       frame-types=admit-all ingress-filtering=no

 3 I   interface=eth5 bridge=bridge_local priority=0x80 path-cost=10 internal-path-cost=10 edge=auto point-to-point=auto
       external-fdb=auto horizon=none hw=yes auto-isolate=no restricted-role=no restricted-tcn=no pvid=1
       frame-types=admit-all ingress-filtering=no

2) Bridge settings freezes constantly (for example, after trying to check/uncheck "igmp snooping", bridge completely shuts down and remain disabled for approx. 1 minute).

It's slightly annoying....

I have 750Gr3 and HW offload works, but you need to set Protocol Mode to "none".

/interface bridge
add name=bridge-internal protocol-mode=none

Then because the script of the default configuration of the RB750GR3 has set "protocol-mode = rstp" ??? It would be a problem to solve by Mikrotik support, since being in RSTP mode will never use HW OFF ...

Institor · Sun Feb 18, 2018 5:31 am

Yes, rstp was enabled on bridge...
However, disabling rstp took about 3-4 minutes, i think that's a little too slow... Changing other settings (like "fast forward") also freezes the bridge for several minutes.

kuzma2000 · Sun Feb 18, 2018 8:57 pm

Hi!
I successfully updated through Dude all my rb941 (2 pieces) ccr1009 (2 pieces) and RB WAP G-5hact2hnd from 6.41.1 to 6.41.2.
But I couldn't update any rb962.
Also, I tried to update rb962 via Winbox.
The reason is: there is not enough space.
rb962 have only 16 MB, of which only 20-30% is available.
Last time I altering them through NetInstall/
What to do now?

ATROX · Mon Feb 19, 2018 11:01 am

ap - cAP ac (RouterBOARD cAP Gi-5acD2nD)
ARP table - empty
Current Tx Power 2GHz - empty

OS - 6.41.2

Real IPQ4019 ???
In specification - IPQ4018.

tigro11 · Tue Feb 20, 2018 12:40 am

hi guys, on my rb750Gr3 with firmware 6.42rc28 I wanted to mount a hotspot for a job. I can not install the user-manager package.
Are there any solutions?
thank you all

Valerio

ATROX · Tue Feb 20, 2018 7:28 am

ap - cAP ac (RouterBOARD cAP Gi-5acD2nD)
OS - 6.41.2

Some clients are disconnected from the point here with such an error:
disconnected, extensive data loss
and
disconnected, received disassoc: sending station leaving (8)
I settings many MicroTick AP.
The settings are the same everywhere, it's on this, new such problems.

For example, this client is sitting on the spot, does not run around the office.

Well, here's a look at what the rest are doing ...

docmarius · Tue Feb 20, 2018 8:50 am

...
Current Tx Power 2GHz - empty
...

FYI: 0 dBm does not mean "empty"/zero power. It is a valid value and means 1 mW. Still odd though, if it changed by itself.

buzzdee · Tue Feb 20, 2018 6:06 pm

@sindy

Thanks for pointing out which order to use in scripting. I must have been lucky that my scripts worked in earlier versions without paying that much attention to the order.

ATROX · Wed Feb 21, 2018 10:50 am

ap - cAP ac (RouterBOARD cAP Gi-5acD2nD)
ARP table - empty
Current Tx Power 2GHz - empty

OS - 6.41.2

The new chips have a new method of obtaining power data. not yet supported Route OS.
MikroTik support answered

HiroZA · Wed Feb 21, 2018 6:55 pm

My current router was ISP supplied and has all the default config on it. If i download this update and install it, will this setup be overwritten? Will i have issues due to PPPoE settings being factory reset ?

sindy · Wed Feb 21, 2018 8:12 pm

My current router was ISP supplied and has all the default config on it. If i download this update and install it, will this setup be overwritten? Will i have issues due to PPPoE settings being factory reset ?

Even if you upgrade between stable versions, you should always backup the configuration and download the backup file from the router. Exporting it in the text form into another file (using "/export file=backup_version_today") is also a good practice because the "binary" backups from older versions are not always readable by newer versions. And noting down any credentials (in your case, especially those related to the PPPoE account) is also a good idea.

I assume that your ISP doesn't mind you to do the upgrade as they've given you administrator rights to the router.

coliflower · Wed Feb 21, 2018 9:03 pm

I am sorry but I am still confused with the new bridge implementation

I updaten one of my wAP-ac to 6.41.2 and created a bridge - this bridge has an IP 10.0.100.4/24 and PVID1 because the AP itself is a part of the administrative-VLAN1-un-tagged (Internet-Cable-Modem to Firewall to Switch to wAP-ac via TRUNKs).

Connected to the bridge via PORT is the only one available ether1 - this port has a PVID1 as well because if I understood the brief explanation of MikroTik, it is need for the un-tagged ingress (from ether-cable to port-ether1 / port-ether1 to the bridge) traffic ... On this PORT there is a cable with Trunk (vlan1-un-tagged, vlan10-tagged, vlan20-tagged, etc.).
Frame-Types = admit all ... correct / why ?
Ingress-Filtering = unchecked ... correct / why ?

The next PORT to the bridge is the physical WLAN-chip (not vAP) - this port also has PVID1 because the SSID of the WLAN-chip is the administrative-VLAN1-un-tagged.
Frame-Types = admit all ... correct / why ?
Ingress-Filtering = unchecked ... correct / why ?

Now the vAP's ... e.g. vAP10, vAP20, vAP30, etc.
For example: vAP10 has PVID10
Frame-Types = admit only VLAN tagged ... correct / why ?
Ingress-Filtering = unchecked ... correct / why ?

Next is interface / bridge / VLAN:

I created VLAN ID = 1
Tagged = empty ... correct / why ?
Untagged = bridge, vlan1, wAP-2GHz, wAP5GHz ... correct / why ?

I created VLAN ID = 10
Tagged = bridge, vlan10 ... correct / why ?
Untagged = vAP10 ... correct / why ?

At the end I activated VLAN-Filtering under /interface bridge

One additional question ... is it needed to create the VLANs under /interface VLAN, too ?

As you see there are some puzzle-brickets I know but I am not able to make a picture of them

Any help is highly appreciated - especially with explanation

sindy · Wed Feb 21, 2018 9:18 pm

I am sorry but I am still confused with the new bridge implementation

I updaten one of my wAP-ac to 6.41.2 and created a bridge - this bridge has an IP 10.0.100.4/24 and PVID1 because the AP itself is a part of the administrative-VLAN1-un-tagged (Internet-Cable-Modem to Firewall to Switch to wAP-ac via TRUNKs).

Connected to the bridge via PORT is the only one available ether1 - this port has a PVID1 as well because if I understood the brief explanation of MikroTik, it is need for the un-tagged ingress (from ether-cable to port-ether1 / port-ether1 to the bridge) traffic ... On this PORT there is a cable with Trunk (vlan1-un-tagged, vlan10-tagged, vlan20-tagged, etc.).
Frame-Types = admit all ... correct / why ?
Ingress-Filtering = unchecked ... correct / why ?

The next PORT to the bridge is the physical WLAN-chip (not vAP) - this port also has PVID1 because the SSID of the WLAN-chip is the administrative-VLAN1-un-tagged.
Frame-Types = admit all ... correct / why ?
Ingress-Filtering = unchecked ... correct / why ?

Now the vAP's ... e.g. vAP10, vAP20, vAP30, etc.
For example: vAP10 has PVID10
Frame-Types = admit only VLAN tagged ... correct / why ?
Ingress-Filtering = unchecked ... correct / why ?

Next is interface / bridge / VLAN:

I created VLAN ID = 1
Tagged = empty ... correct / why ?
Untagged = bridge, vlan1, wAP-2GHz, wAP5GHz ... correct / why ?

I created VLAN ID = 10
Tagged = bridge, vlan10 ... correct / why ?
Untagged = vAP10 ... correct / why ?

At the end I activated VLAN-Filtering under /interface bridge

One additional question ... is it needed to create the VLANs under /interface VLAN, too ?

As you see there are some puzzle-brickets I know but I am not able to make a picture of them

Any help is highly appreciated - especially with explanation

I'd suggest you to either join this topic or create a new one.

coliflower · Wed Feb 21, 2018 9:49 pm

I did, but I am still confused

Maybe because of the WLAN-aspect ?
And I opened one here - but I found other issues in between ... viewtopic.php?f=7&t=130734

davidv · Thu Feb 22, 2018 12:01 am

LHG´s AC 6.41.2
After the restarts it shows:
memory; system, error, critical; kernel failure in previous boot.
in the shutdown of the system only reboots

any solution?
Thank
.......................................................................................................
reset when you change the configuration and the connection wlan falls
problem coment: viewtopic.php?f=7&t=128916

aforster · Thu Feb 22, 2018 3:42 am

Folks,

I tried twice to move from 6.40.3 to other streams, but a few weeks after the upgrade, the router started going slow, and after a reboot, the same problem happened, exactly the same the two times: it lost part of the new bridge configurations. Even with reboots or reloading the configs, it would not fix it.. very, very strange.

It was a real pain to download the old 6.40.3 code using my cell phone as internet access to upload it to the router and restore old config.

Anyone had the same problem?

Now, Im back to 6.40.x.. just moved to 6.40.6 to see how that one is.. otherwise, will downgrade to 6.40.3 again.

jaytcsd · Thu Feb 22, 2018 8:28 am

My hAPlite locked up after upgrading from 6.3 something using quickset to download and install.

I upgraded winbox through the tool option first, it crashed.
It worked when I restarted it and I got into the router to run quickset.
It rebooted OK, I got in and was looking at the log when it died.
I've done 3 power cycles and it still won't respond to pings.

Will netinstall preserve any supout data? I think that is my only recourse to help troubleshoot what happened.
My 10+ year old RB532 is working fine, otherwise this would be a letter.

minhcoi · Sat Feb 24, 2018 6:16 am

I was upgrade my 951ui-2hnd from 6.40 to 6.41 and then i deleted all files in router... Now my 951ui become continually reset... I tried Netinstall but it cant see my router...
How can i fix this?

xtcinmymind · Sun Feb 25, 2018 1:44 pm

Hello

great job with the latest firmware work so fast and more reliable .I just wonder to know about bonding .
before i used to uncheck master port on ether1 or 2 to be able to create bonding .
Now i cant do that cos it say ether1 or 2 already on brigde1 .
Can someone tell me how to create bonding easy way on the firmware 6.41.2

thnx

sindy · Sun Feb 25, 2018 2:42 pm

Now i cant do that cos it say ether1 or 2 already on brigde1 .
Can someone tell me how to create bonding easy way on the firmware 6.41.2

Go to Bridge -> Ports, remove ether1 and ether2 from the list, go to Interfaces -> Add new -> Bonding, set ether1 and ether2 as slaves, then maybe make the new bonding interface a member of that bridge from which you've removed ether1 and ether2. Done.

xtcinmymind · Sun Feb 25, 2018 3:43 pm

Now i cant do that cos it say ether1 or 2 already on brigde1 .
Can someone tell me how to create bonding easy way on the firmware 6.41.2
Go to Bridge -> Ports, remove ether1 and ether2 from the list, go to Interfaces -> Add new -> Bonding, set ether1 and ether2 as slaves, then maybe make the new bonding interface a member of that bridge from which you've removed ether1 and ether2. Done.

Yes thnx was easy way .also in the end have to add ports with the bond created and brigde ..thnxxx again

weldox · Tue Feb 27, 2018 12:46 pm

I have my 7 milkrotik devices all stick to 6.38.5 and I am really scared in try to upgrade. I see a lot of problems in this thread....
Is it really so dangerous?

Tue Feb 27, 2018 2:57 pm

Don't update anything if it scares you. Read the release notes that describe version differences. If you think there is a change you would like to have, make your own tests with backup devices before you decide to roll out the new version to your production ones.

pe1chl · Tue Feb 27, 2018 5:36 pm

Also, when you want to be conservative set the update channel to "bugfix only" and you can update to 6.40.6 which does not
have this new style bridge instead of the existing bridge master-port setup. That should avoid a lot of problems when you
have a complicated configuration with VLAN and switching.

weldox · Tue Feb 27, 2018 5:39 pm

Thank You very much!

bbs2web · Wed Feb 28, 2018 12:25 am

We've had several instances of CCR routers losing bridge ports after upgrading from 6.41 to 6.41.2.

We run rancid as a configuration revision management system and reviewed logs after scheduling upgrades at 3am, ports simply missing.

Example:
/interface bridge port
add bridge=bridge-hosting interface=xxx-vlan12
missing after upgrade:

add bridge=bridge-nat interface=sfp11
add bridge=bridge-nat interface=xxx-vlan13
add bridge=bridge-management interface=xxx-vlan11
add bridge=bridge-management interface=sfp10
add bridge=bridge-hosting interface=sfp7

Wed Feb 28, 2018 11:29 am

Did you have any of bridge ports disabled?

lukas2511 · Thu Mar 01, 2018 12:16 am

I upgraded a RouterBoard hEX from 6.41 to 6.41.2 and had bridge ports disappearing too.

ether1 and ether2 were no longer part of any bridges (which was kinda unfortunate as those were my uplink and local-network ports)

bbs2web · Thu Mar 01, 2018 12:26 am

No, they were active and working perfectly. This occurred on another CCR where I subsequently connected via mac telnet and simply dumped all bridge ports back again, any that already existed we skipped.

I'll book a maintenance window for next weekend (10th March) and take a supout before and after upgrading.

I initially thought that it may relate to large number of bridge ports but the example above occurred on a router with 3 bridges, each with pairs...

Did you have any of bridge ports disabled?

docmarius · Thu Mar 01, 2018 9:00 am

On my CCR1009 this version is unstable.
After an unpredictable time, the DHCP server stops responding, and routing seems to go amok (routes are there, but it is like connection tracking is not working). A reboot restores functionality for a time (from minutes up to a day).
Downgrading to 6.40.6 "cures" this behavior. I do not remember seeing this on 6.41.1. This could be the beginning of a beautiful friendship with 6.40.6

b2fshen · Sun Mar 04, 2018 6:21 am

Hi!
I am having trouble sending AT commands to LTE interface. After entering of ANY AT command like "interface lte at-chat lte1 input="ATI"" I receive: "action timed out - try again, if error continues contact MikroTik support and send a supout file (13)" reply.
My board is hAP (RB951ui-2nD) running 6.41.2. I tried it with USB modems ZTE MF730M and Huawei e3372 with the same result.
From Windows PC I can
Does anybody had same issues?

kasana825 · Sun Mar 04, 2018 7:41 pm

Hello,
I have routerboard and I configure 1 as WAN and 2,3 Bridge with WAN and I have set bandwidth limit on ethernet port 3, see attached file please

but today I bought rb1100ahx4
there is no option to set bandwidth limit on interface
but there is no option to set bandwidth limit

nimbo78 · Mon Mar 05, 2018 2:33 pm

FR:
plz add ICCID to lte info. needed to identify what SIM is now installed into the LTE kit. physically difficult to take out SIM from LTE kit and insert it back to the routerboard every time when we need to see ICCID.

scrarfussi · Mon Mar 05, 2018 7:52 pm

i upgraded my rb750gr2 to 6.41.2 and i lost internet access
i can ping on the router
but devices connected to it have no internet access

bbs2web · Tue Mar 06, 2018 6:01 pm

I've sent supout files for a CHR instance before and after upgrading (6.41 to 6.41.2) where bridge ports disappear after upgrading. This problem has occurred on CCR, CHS and 750Gr3 routers, one of which only had 3 bridges with 6 bridge ports...

I'll book a maintenance window for next weekend (10th March) and take a supout before and after upgrading.

I initially thought that it may relate to large number of bridge ports but the example above occurred on a router with 3 bridges, each with pairs...

ok1mgj · Wed Mar 07, 2018 12:06 am

Hi all.

i`v upgraded 2 routerboards and on both of them i get into troubles with wireless.
RB751U-2HnD and RB751G-2HnD - both of them were on some 6.3x.x versions.

after upgrade to 6.41.2 (and corresponding firmware version) was RB unstable, after reboot and when wireless was enabled routerboard goes into booting loop.
configuration reset is unusable, because in default config is wifi enabled => booting loop.
next step was netinstall - yes, it helped. but after enabling wifi it was in booting loop again
i`v used netinstall again, did some tests and things were fine until i`v enabled wifi. Welcome to booting loop.
on 751U i`v tried 6.42rc37 - and after enabling wifi -> whooo, booting loop.
OK, something went wrong.
netinstall with older version (some 6.39.x) - and after enabling wifi..... voila, booting loop.

I have these RBs on very simple configuration - only one bridge across all ports including wifi - RBs acts only as AP for home usage.
So, now these routerboards is in stable config, but with disabled wireless.

Any idea how to solve this issue?

pe1chl · Wed Mar 07, 2018 11:28 am

Save your configuration and do a completely new netinstall.
In the past there have been changes to the wireless and at that time it already was tricky to update past that release,
it would not surprise me when it again causes problems when you still had the wireless package from before that change.

ok1mgj · Wed Mar 07, 2018 1:18 pm

Hi,
after netinstall is configuration clear, so it is not about configuration.

edit: on RB951G-2HnD is all OK and working properly.

ok1mgj · Wed Mar 07, 2018 9:22 pm

guys, please, don`t panic (..and thanks for all the fish)

after a few hours of tests with other pices of routerboards (i have many RBs of many types) i have found, where trouble is located.

and this was: power adapter.

so, i`m very sorry to all the people.

JanezFord · Thu Mar 08, 2018 1:28 pm

We've had several instances of CCR routers losing bridge ports after upgrading from 6.41 to 6.41.2.

I can also confirm that. CCR lost about half of bridge ports for no apparent reason after upgrading from 6.41 to 6.41.2. I also had to netinstall one hAP-AC with very simple config (simple CAP with all ports bridget + dhcp on bridge) which went into reboot loop after upgrade.

JF.

pe1chl · Thu Mar 08, 2018 2:07 pm

A not so widely know feature of the "older" routers (this was lost in the recent low-end models that have only 16MB of flash) is the
possibility to divide the flash in two partitions that can be copied over and one can be activated. On a CCR you can copy the existing
firmware+config to the inactive partition, upgrade, reboot and when is not troublefree you can activate the second partition and reboot again to
fall back to the previous situation (without the issue that an upgrade could have modified the configuration).
You probably already know this but I bet not every admin does.

Mon Mar 12, 2018 10:01 am

Version 6.41.3 has been released in current channel:
viewtopic.php?f=21&t=131810

Who is online