v6.35.2 [current] is released!

Fri May 06, 2016 10:20 am

What's new in 6.35.2 (2016-May-02 10:09):

*) discovery - fixed identity discovery (introduced in 6.35.1);
*) firewall - fixed policy routing configurations (introduced in 6.35rc38);
*) log - fixed time zone adjustment (introduced in 6.35.1);
*) queue - fixed interface queue type for ovpn tunnels;
*) snmp - fixed snmp timeout (introduced in 6.35.1);
*) vrrp - fixed missing vrrp interfaces after upgrade (introduced in 6.35.1).

If you experience version related issues, then please send supout file from your router to support@mikrotik.com. File must be generated while router is not working as suspected or after crash.

nxs02 · Fri May 06, 2016 3:21 pm

keep up the good work guys

napismizpravu · Fri May 06, 2016 4:42 pm

RB433UAH 6.35.2 > IPv6 Tunnel no works local, no advertise IPv6 adresses (6.35.1 works)

edit: reboot routerboard IPv6 tunel works in version 6.35.2 ...depression RouterOS after updating

105547111 · Sat May 07, 2016 3:33 am

What about load balancing - Is this fixed?

MartijnVdS · Sat May 07, 2016 9:00 am

What about load balancing - Is this fixed?

I'd say the VRRP fix would imply that?

dgnevans · Sat May 07, 2016 4:00 pm

Simple queues not working 100%. When simple queue applied web speeds fluctuate and picks drop to external websites. Removing simple queue resolves. Reverting to v6.32.4 resolves the issue.

pe1chl · Sun May 08, 2016 10:07 am

What about load balancing - Is this fixed?
I'd say the VRRP fix would imply that?

VRRP has nothing to do with load balancing...

Cha0s · Sun May 08, 2016 2:38 pm

What about load balancing - Is this fixed?

*) firewall - fixed policy routing configurations (introduced in 6.35rc38);

I think this fix covers 'load balancing'

Deantwo · Mon May 09, 2016 10:39 am

*) queue - fixed interface queue type for ovpn tunnels;

Fixed what and how?

I am having problems with SSTP interface queues after upgrading routers from 5.x versions.
See: http://forum.mikrotik.com/viewtopic.php?f=1&t=108175

Don't know if a supout file is necessary when it is an easily reproduced issue.

docmarius · Mon May 09, 2016 12:52 pm

On my hAP SNTP Server does not synchronize with a USB GPS starting with 6.35 (GPS reception is working, GPS monitor is updating).
Worked on 6.34. Reverting to 6.32.4 solves the issue.

Mon May 09, 2016 1:41 pm

On my hAP SNTP Server does not synchronize with a USB GPS starting with 6.35 (GPS reception is working, GPS monitor is updating).
Worked on 6.34. Reverting to 6.32.4 solves the issue.

post your "/port" settings please, as ours still works. SNTP can't have a server, you mean regular NTP probably

pe1chl · Mon May 09, 2016 3:49 pm

Running 6.35.2 on a CCR1009-8G-1S-1S+ the display of the FAN RPM is erratic.
When viewing the health page on the browser about 1 out of 3-4 automatic refreshes the FAN rpm line disappears.
When displaying the FAN RPM on the frontside display it drops to 0.
(fan is running at constant speed judged by the sound)

maruhpro · Mon May 09, 2016 5:54 pm

It's a bug or a feature?

Some clients directly connected to the router (cable) or to a panel (wireless 5.8/2.4) appear in arp address with 00:00:00:00:00:00 mac.

In version 6.28 it's ok.

Mon May 09, 2016 5:59 pm

It's a bug or a feature?

Some clients directly connected to the router (cable) or to a panel (wireless 5.8/2.4) appear in arp address with 00:00:00:00:00:00 mac.

In version 6.28 it's ok.

feature, that is how incomplete ARP entries are indicated, so that you have easier time to debug ARP related problems

planetcaravan · Mon May 09, 2016 11:11 pm

Hi,

I've just updated a brand new LHG 5 from 6.35.1 to 6.35.2 and it bricked! Can't access, can't see on WinBox, ether1 is up only for 1 second.

Mon May 09, 2016 11:52 pm

_removed_

docmarius · Tue May 10, 2016 9:57 am

On my hAP SNTP Server does not synchronize with a USB GPS starting with 6.35 (GPS reception is working, GPS monitor is updating).
Worked on 6.34. Reverting to 6.32.4 solves the issue.
post your "/port" settings please, as ours still works. SNTP can't have a server, you mean regular NTP probably

Yes, i meant NTP server.
Exact router model is RB951Ui-2nD EU, Firmware 3.33

/port
set 0 name=usb1

/system gps
set enabled=yes port=usb1 set-system-time=yes

/system ntp server
set enabled=yes

Trying to use it as a NTP server gets error: "server not synchronized" on the other RB.
GPS is a ublox 6 USB mouse.

As a hint on the GPS mouse:
- It works on Windows
- It works on Omnitik (checked until 6.34)
- It fails to work on NetMetal5 RB922UAGS (checked different firmwares, it gets a GPS data set, then stops)
- It fails on Linux (kernel 3.16) unless plugged into a USB hub - it requests the wrong driver (uhci vs. ohci, ok with USB hub)
- It worked on hAP until 6.34.x, fails on 6.35 and 6.35.2

ssbaksa · Tue May 10, 2016 11:55 am

Hi,

I have upgraded my RB1200 from 6.34.4 to 6.35.2 and all worked OK on the first glance. This is our main router firewall in test lab for connecting corporate net to test lab. I have checked connectivity to (almost) all mayor segments - no problems there.
In the morning my colleagues reported that 2 VLAN (VLAN310 and VLAN311) segments of 42 which we are using are not to be reached. I have tested ping from corporate and it failed. Ping from router was successful as long I am using the same interface - routing was broken for those two segments only. Reboot didn't help either.
I dint have any other choice but to revert to previous firmware. Only one thing I have forgot, to build support.rif while new firmware was applied. I have created one after downgrade. I am not sure that this will help to pinpoint error.

What can be done now?

Br

Sasa

Tue May 10, 2016 12:09 pm

/port
set 0 name=usb1

/system gps
set enabled=yes port=usb1 set-system-time=yes

/system ntp server
set enabled=yes

Try this:

set 0 baud-rate=4800

maruhpro · Tue May 10, 2016 5:26 pm

It's a bug or a feature?

Some clients directly connected to the router (cable) or to a panel (wireless 5.8/2.4) appear in arp address with 00:00:00:00:00:00 mac.

In version 6.28 it's ok.
feature, that is how incomplete ARP entries are indicated, so that you have easier time to debug ARP related problems

How can I debug ARP related problems?
After update to 6.35.2 many clients have 00:00:00:00:00:00 mac address.

docmarius · Tue May 10, 2016 6:30 pm

Try this:
Code: Select all
set 0 baud-rate=4800 

I set the baud rate to 4800 (was 9600).

Just to be clear on this: GPS Monitor works in both cases (position and time are updated on 6.32.4 and 6.35.2 and both baud rates):

/system gps> mon
        date-and-time: may/10/2016 15:11:02
             latitude: N 45 43' 32.796''
            longitude: E 21 14' 43.796''
             altitude: 96.800003m
                speed: 0.055560 km/h
  destination-bearing: none
         true-bearing: 0.000000 deg. True
     magnetic-bearing: 0.000000 deg. Mag
                valid: yes
           satellites: 9

- The NTP server on 6.35.2 still remains in INIT state (linux client)/ Last Bad Packet Reason: server-not-synchronized (and no update) for ROS client.
- On 6.36rc11 the USB port shows to be invalid and can not be used. After deleting the invalid entry, no usb port is shown.
- Reverting to bug fix 6.32.4 fixes everything and NTP server is working, even with default 9600bps.

Senux · Tue May 10, 2016 6:45 pm

It's a bug or a feature?

Some clients directly connected to the router (cable) or to a panel (wireless 5.8/2.4) appear in arp address with 00:00:00:00:00:00 mac.

In version 6.28 it's ok.
feature, that is how incomplete ARP entries are indicated, so that you have easier time to debug ARP related problems
How can I debug ARP related problems?
After update to 6.35.2 many clients have 00:00:00:00:00:00 mac address.

Also there is ghost clients with 00 mac. Mikrotik claims solved this issue, but maybe only in dreams

maruhpro · Tue May 10, 2016 7:49 pm

feature, that is how incomplete ARP entries are indicated, so that you have easier time to debug ARP related problems
How can I debug ARP related problems?
After update to 6.35.2 many clients have 00:00:00:00:00:00 mac address.
Also there is ghost clients with 00 mac. Mikrotik claims solved this issue, but maybe only in dreams

and I can't do downgrade, because appear a message in log "omitting package routeros-mipsbe-6.32.4; newer package routeros-mipsbe-6.35.2 is already installed"

Tue May 10, 2016 7:58 pm

You need to press downgrade button. Simple reboot doesn't the downgrade on its own.
For 6.32.4 you can use automatic download and upgrade function with bugfix chain set.

drbunsen · Wed May 11, 2016 2:35 pm

- The NTP server on 6.35.2 still remains in INIT state

I can confirm the same on RB3011 for v6.35 and v6.35.2

temandroid · Thu May 12, 2016 9:18 am

Have a problem on x86 system - can't access to ROS by WinBox or WebFig, but telnet is ok.
In WinBox no changes after phrase "Downloading plugins" , but in log "user logged in by winbox"...

Thu May 12, 2016 9:31 am

Have a problem on x86 system - can't access to ROS by WinBox or WebFig, but telnet is ok.
In WinBox no changes after phrase "Downloading plugins" , but in log "user logged in by winbox"...

firewall? what winbox version? mac-winbox?

acidsas · Thu May 12, 2016 12:46 pm

Something strange is going on with ether5 on RB750GL. Port 5 is disconnected, but there is a lot of traffic according to counters. Up to 40Gbit per second

I'll mail support to support.

Towerboy1983 · Thu May 12, 2016 1:20 pm

Hi, all
Not sure if this is the correct place.

I have a CCR1009-8G-1S-PC cloud core with OS 6.35.2 that we want to use as a Dude server.

We got Dude loaded onto the router, works almost perfectly.
We are experiencing 2 problems with this router now

1 For some reason or other, if we open an router in Dude, it does not go full screen, if we open it full screen, the original size is the only working area.
2 After we got Dude working, we aren't able to safe any settings. Not even if I try and reset the router does the settings disappear.

All help will be appreciated.

Vitokhv · Sat May 14, 2016 11:12 am

6.32.4 (Bugfix only)
6.35.2 (Current)
6.36rc12 (Release candidate)

that this stable, no error?
for RB751U-2HnD

tesomahok · Sat May 14, 2016 1:23 pm

I have upgraded a couple of my routers (RB750 and RB450) from from 6.33.3 to 6.35.2 and IPSec vpn on all of them have the same problem - the tunnel is up, but the the traffic can't get thru. Reconfigure the tunnels does not fix the problem, so i downgraded back to version 6.33.3 and that fixed the problem immediately.

wencs · Sun May 15, 2016 4:26 pm

Report BUG system clock time zone error Etc/GMT+4 error assing GMT offset -04:00
and correct America/Caracas/ -04:00

Sun May 15, 2016 8:08 pm

Report BUG system clock time zone error Etc/GMT+4 error assing GMT offset -04:00
and correct America/Caracas/ -04:00

This is NOT a bug. In fact, this works as expected. A quote from Wikipedia:

The special area of "Etc" is used for some administrative zones ... In order to conform with the POSIX style, those zone names beginning with "Etc/GMT" have their sign reversed from the standard ISO 8601 convention. In the "Etc" area, zones west of GMT have a positive sign and those east have a negative sign in their name (e.g "Etc/GMT-14" is 14 hours ahead/east of GMT.)

YrPa · Mon May 16, 2016 8:13 pm

Hi all!
Trying to upgrade from 6.0 via WinBox connection but getting "ERROR : missing wireless-6.35.2-mipsbe.npk"
I am new in RouterOS, so what is wrong?

Mon May 16, 2016 10:34 pm

Use wireless cm2 or rep packages instead original wireless package.

Thu May 19, 2016 7:00 pm

feature, that is how incomplete ARP entries are indicated, so that you have easier time to debug ARP related problems
How can I debug ARP related problems?
After update to 6.35.2 many clients have 00:00:00:00:00:00 mac address.
Also there is ghost clients with 00 mac. Mikrotik claims solved this issue, but maybe only in dreams

Mikrotik "claims" to have solved an issue where the ARPs were being sent out via the wrong interface (misplaced ARP). All-zero MAC addresses in the ARP table are NOT misplaced ARPs.

An IP address with all-zero MAC in the ARP table means that the Mikrotik has sent an ARP request for that IP address, but it has not received a reply to the ARP request. The zeros are a place-holder until something replies to the ARP.

The router has sent an ARP for that IP address because it has received a packet with the unused IP address as a destination. When the Mikrotik tries to forward this packet, it must first know the MAC address, right? So it tries to send an ARP to discover the MAC of the IP address. In this case, there is no host using the IP address, so no reply is made for the ARP, so the ARP entry stays set to all-zeros.

This all-zero MAC behavior is there so that if someone sends lots and lots of packets to the same unused IP address, the Mikrotik will not send a flood of ARP requests. If the Mikrotik didn't store the "pending" requests, then every single packet for the unused address would generate an ARP broadcast. This would lead to a possible way to flood the network with broadcasts.

This 00:00:00:00:00:00 behavior is helpful for troubleshooting because if you see it, then you know you're not receiving ARP replies from that IP address. If you have that address configured on a device, then seeing 00:00:00:00:00:00 tells you that the host cannot answer ARPs for some reason - usually a network connectivity issue.....

andersonlich · Fri May 20, 2016 7:12 am

using v6.35.2 on my CCR1072.
problem using SFP S-RJ01, this is not issued on v6.34.2.
already test back to back with notebook using IP. the result is can't ping to router.
the CCR detected the SFP-T, and config was disabled autonego and fixed to 1g + full duplex.

kuxu · Sat May 21, 2016 5:38 pm

Using v6.35.2 on my 2011UiAS-2HnD
Simple queues not working

shuey · Mon May 23, 2016 3:56 pm

I'm trying to install this version, but it doesn't have the server and client all in one package (like v3.6 did - one single installation file and you're up and running). I've downloaded the client install, and the server npk file, but the wiki doesn't explain how to "upload" the file in order to get it to work. And when I launch the client, it can't connect to the server (which obviously makes sense because I don't know how to incorporate the npk file).

Does this current version no longer function on the same level as v3.6? Do I need RouterOS in order to use the latest version of The Dude? If so, it looks like the "free" version isn't so "free" anymore, correct? (because RouterOS actually costs money to use?...)

Mon May 23, 2016 4:29 pm

I'm trying to install this version, but it doesn't have the server and client all in one package (like v3.6 did - one single installation file and you're up and running). I've downloaded the client install, and the server npk file, but the wiki doesn't explain how to "upload" the file in order to get it to work. And when I launch the client, it can't connect to the server (which obviously makes sense because I don't know how to incorporate the npk file).

Does this current version no longer function on the same level as v3.6? Do I need RouterOS in order to use the latest version of The Dude? If so, it looks like the "free" version isn't so "free" anymore, correct? (because RouterOS actually costs money to use?...)

Are you talking about the Dude? In this case, this is the wrong topic, and the question was already answered in the manual and elsewhere:

http://wiki.mikrotik.com/wiki/Manual:Th ... stallation
http://wiki.mikrotik.com/wiki/Manual:Th ... nToNewDude
http://wiki.mikrotik.com/wiki/Manual:Th ... _on_VM_CHR

shuey · Mon May 23, 2016 4:42 pm

Are you talking about the Dude? In this case, this is the wrong topic, and the question was already answered in the manual and elsewhere:

http://wiki.mikrotik.com/wiki/Manual:Th ... stallation
http://wiki.mikrotik.com/wiki/Manual:Th ... nToNewDude
http://wiki.mikrotik.com/wiki/Manual:Th ... _on_VM_CHR

Thanks for the info Normis! (and for your patience - I'm still trying to wrap my head around all this new information, but your links helped a lot!)

mohnori · Tue May 24, 2016 3:26 am

Hello
in PPPoE profile there is IdleTimeout its work like Session Timeout . that's mean if I put 5 minutes in Idle Timeout even if the user is active its will disconnected after 5 minutes , my router is CCR 1036 12G 4S. Please fix it.

JanezFord · Wed May 25, 2016 2:58 pm

Using v6.35.2 on my 2011UiAS-2HnD
Simple queues not working

Make sure you are not using fastrack .... my 2011 simple queues work fine on v6.35.2 but without fastrack.

JF

notToNew · Wed May 25, 2016 5:11 pm

Using v6.35.2 on my 2011UiAS-2HnD
Simple queues not working
Make sure you are not using fastrack .... my 2011 simple queues work fine on v6.35.2 but without fastrack.

JF

Hm, I have the same issue with simple queues, which worked fine in 6.32.x.
I'm not using fasttrack (all fasttrack-rules in firewall are disabled!).
If it matters, I'm using 2 bridges (bridge-wan, bridge-lan).

dgcapel · Fri May 27, 2016 12:13 pm

Problem with 6.35.2
I set the public IP address in a interface : example ether6. And this IP address are disabled. If the DHCP client in this interface try to get Address, this return a error:

12:04:17 dhcp,error dhcp-client on ether6 failed to add IP address XXX.XXX.XXX.XXX: already have such address (6)

But .... the IP address are disabled!! If edit the IP address (disabled) or delete : work.
Thanks

Fri May 27, 2016 12:20 pm

It's common behaviour.
Disabled = inactive != not assigned.
Same problem if you try to add to the firewall/address list the second item with same IP. There is now way even if previous is disabled.

Fri May 27, 2016 12:32 pm

It is not a problem, you cannot add same address multiple times in the list and it will not be allowed in the future too.

Fri May 27, 2016 12:37 pm

I set the public IP address in a interface : example ether6. And this IP address are disabled. If the DHCP client in this interface try to get Address, this return a error

This seems to have recently been addressed in 6.36rc16. A quote from here:

Version 6.36rc16 has been released.

Changes since previous version:
...
*) address - allow multiple euqla ip addresses to be added if neither or only one is enabled;
...

rushlife · Fri May 27, 2016 1:05 pm

on ccr1072 is there problem, LCD does not working, whole time is screen empty / white

dgcapel · Fri May 27, 2016 3:00 pm

It is not a problem, you cannot add same address multiple times in the list and it will not be allowed in the future too.

OK! Thank you

wispwest · Fri May 27, 2016 9:17 pm

We're still seeing our logs FLOODED with SNTP date/time updates or changes, either something isn't right, or MIkrotik decided to just recently start LOGGING every time the SNTP updates, which I cannot understand why anyone needs that information, and it just floods the log cache so that you cannot see anything IMPORTANT in the last few days...

Also I hope there isn't simple queue problems like people say, I just upgraded a 1100AHx2 which has 490 Simple Queues. What are the symptoms for the simple queue issues?

blispx · Fri May 27, 2016 10:45 pm

Dude client crash under Wine

wine version:

wine --version
wine-1.9.9

dude crash dump:

6.35.2 BREAKPOINT at: 7bc812c8

eip=7bc812c8 eflags=246
edi=a5e7ec esi=a5cfbc ebp=a5dfc8 esp=a5cf8c
eax=7b4287c5 ebx=7eeda000 ecx=a5cf80 edx=0

log:
cw:1411 backtrace:[416733, 4168a3, 500cbc, 74008c, 5007c4, 502dc6, 6f8b7a, 4010b9, 40129c, 7b4602dc, 7b46130a, 7bc812bc,
 7bc8426d, 7bc8129a, 7bc53a07, f7572d7d]

unreliable backtrace:

modules:

stack: 0 4292518052

wine debug:

WINEARCH=win32 wine /home/blispx/.wine/drive_c/Program\ Files/Dude/dude.exe 
fixme:heap:RtlSetHeapInformation (nil) 1 (nil) 0 stub
fixme:process:SetProcessDEPPolicy (1): stub
fixme:heap:RtlSetHeapInformation (nil) 1 (nil) 0 stub
err:winediag:WSASocketW Failed to create a socket of type SOCK_RAW, this requires special permissions.
Pinger: socket failed: 10013:
prog dir: C:\Program Files\Dude\
modules:
400000-85e000	dude.exe
63a40000-63a72000	libgobject-2.0-0
64f80000-64fa8000	libfontconfig-1
65300000-65329000	libpixman-1-0
65340000-65364000	libgdk_pixbuf-2.0-0
65580000-655b8000	libpango-1.0-0
66440000-66470000	libcroco-0.6-3
685c0000-6867b000	libglib-2.0-0
68dc0000-68e20000	libcairo-2
68e40000-68e73000	librsvg-2-2
68f40000-68f60000	libexpat-1
69c80000-69cb0000	libpng12-0
6b280000-6b291000	libpangowin32-1.0-0
6c240000-6c263000	libjpeg-62
6d4c0000-6d4cf000	libpangocairo-1.0-0
6d580000-6d5c5000	libgio-2.0-0
6d700000-6d725000	libpangoft2-1.0-0
6dd00000-6dd0c000	libgmodule-2.0-0
70c00000-70c64000	libfreetype-6
70f40000-70f8b000	libxml2-2
7ac10000-7ac79000	riched20
7b420000-7b7e2000	KERNEL32
7bc10000-7bcf6000	ntdll
7d890000-7d9ac000	oleaut32
7d9b0000-7d9f1000	usp10
7dab0000-7dadc000	uxtheme
7dce0000-7dd60000	winex11.drv
7dd70000-7dd84000	imm32
7e340000-7e367000	msacm32
7e370000-7e420000	winmm
7e430000-7e449000	iphlpapi
7e450000-7e487000	winspool.drv
7e490000-7e572000	comdlg32
7e580000-7e66d000	comctl32
7e670000-7e6a6000	ws2_32
7e6b0000-7e71e000	shlwapi
7e730000-7e95a000	shell32
7e970000-7e9da000	rpcrt4
7e9f0000-7eb13000	ole32
7eb20000-7eb2d000	version
7eb40000-7ec7f000	user32
7ec80000-7ec93000	msimg32
7eca0000-7ed09000	advapi32
7ed20000-7ee25000	gdi32
7ee40000-7eee4000	msvcrt
ERROR: createWindow failed error=1411:Class does not exist.

DEBUG: cw:1411 backtrace:[416733, 4168a3, 500cbc, 74008c, 5007c4, 502dc6, 6f8b7a, 4010b9, 40129c, 7b4602dc, 7b46130a, 7bc812bc, 7bc8426d, 7bc8129a, 7bc53a07, f75fdd7d]

notToNew · Sat May 28, 2016 7:28 pm

We're still seeing our logs FLOODED with SNTP date/time updates or changes, either something isn't right, or MIkrotik decided to just recently start LOGGING every time the SNTP updates, which I cannot understand why anyone needs that information, and it just floods the log cache so that you cannot see anything IMPORTANT in the last few days...

Also I hope there isn't simple queue problems like people say, I just upgraded a 1100AHx2 which has 490 Simple Queues. What are the symptoms for the simple queue issues?

Do you update the time cia the "Cloud" - option AND via the ntp?
maybe they both don'tlike each other, and one should be enough

bedior · Sat May 28, 2016 7:41 pm

RB2011UiAS-2HnD-IN on 6.35.2 give slower L2TP/IPsec (AES256 + SHA1):
http://www.speedtest.net/result/5349439075.png (CPU 60% on download)
On version 6.32.2:
http://www.speedtest.net/result/5349455563.png (CPU 100% on download)

Why?

Cha0s · Sat May 28, 2016 7:42 pm

We're still seeing our logs FLOODED with SNTP date/time updates or changes, either something isn't right, or MIkrotik decided to just recently start LOGGING every time the SNTP updates, which I cannot understand why anyone needs that information, and it just floods the log cache so that you cannot see anything IMPORTANT in the last few days...

Also I hope there isn't simple queue problems like people say, I just upgraded a 1100AHx2 which has 490 Simple Queues. What are the symptoms for the simple queue issues?

Do you update the time cia the "Cloud" - option AND via the ntp?
maybe they both don'tlike each other, and one should be enough

mAP Lite has a time keeping bug as mentioned here: http://forum.mikrotik.com/viewtopic.php ... te#p527731
Every 15 minutes the clock runs faster by 1.8seconds.
So sntp every 15minutes tries to correct the time, hence the log flooding.

I am not sure which model wispwest uses. I'm just mentioning it in case anyone else has mAP Lite and wonders why it keeps flooding the logs.
I've also opened a ticket a few days ago (Ticket #2016052566000621) but I haven't received any response yet.

MIkrotik decided to just recently start LOGGING every time the SNTP updates, which I cannot understand why anyone needs that information

It only logs the changes in time by NTP when the time difference on the local clock and the NTP clock is off by 1+ second.
Otherwise on boards that time keeping works ok it will only log the time change after a reboot and the time offset will be at most a few ms.

Regarding the Simple Queues, I am not sure if there is a problem. I only use Queue Tree and they work fine with 6.35.x up to 6.36rc16 on RB850Gx2.

anima3 · Mon May 30, 2016 4:34 pm

Problem with v6.35.2 on CRS226, trunk interface on SFP+ after reboot -it is frozen! -but i can connect to management VLAN!? Of course the end users not receiving DHCP addresses!
And I can't create sup-output file:
[mxx@CPS_Server_CRS226_SW_1] > system sup-output file=support_30_50_2016.rif
system is busy (12)
[mxx@CPS_Server_CRS226_SW_1] > system sup-output file=supout_30_50.rif
system is busy (12)
[mxx@CPS_Server_CRS226_SW_1] >

After reboot nothing good. I must wait CRS226 is up and must change SFP+ patch 1m to (and back) SFP+ patch 3m -now,everiting is working (to next power lost)!

In every building i have ~5 CRS226. After power lost, I can't every time change SFP+ patchcords!

Now, to go back to v6.30.4 I need for each installation through the console cable.

I think, the MikroTIK team is forgotten about this mipsbe hardware

alienID · Tue May 31, 2016 3:12 pm

Hello,
It seem I am not alone, I am using RB750GL my simple queue rules didn't work after upgrade it to 6.35.2 so I need to revert it back to 6.32.4 to make it works.
I am also notice that in simple queue mikrotik didn't state the correct bandwidth that target use for example the client download 1 Mbps in simple queue it's only state about 1/4 of that 250 - 300 kbps after I revert it back to 6.32.4 it seem the number is much closer and also simple queue is works.

wrobli · Sat Jun 04, 2016 1:37 pm

Hi.
When i update CPE to 6.35.2 pppoe session reconnection many times

back to 6.35 resolve problem

Mon Jun 06, 2016 2:34 pm

About queues issue. In older version PPPoE did not have FastTrack support. Now after upgrade to this version support exists. It means that tunnel traffic can bypass queues:
http://wiki.mikrotik.com/wiki/Manual:Wiki/Fasttrack
"Note that not all packets in a connection can be fasttracked, so it is likely to see some packets going through slow path even though connection is marked for fasttrack. This is the reason why fasttrack-connection is usually followed be identical action=accept rule. Fasttracked packets bypass firewall, connection tracking, simple queues, queue tree with parent=global, ip traffic-flow(restriction removed in 6.33), ip accounting, ipsec, hotspot universal client, vrf assignment, so it is up to administrator to make sure fasttrack does not interfere with other configuration"

kristaps · Wed Jun 08, 2016 2:04 am

on ccr1072 is there problem, LCD does not working, whole time is screen empty / white

This issu will be fixed in next version

bedior · Wed Jun 08, 2016 8:46 am

What about LT2P speed? This is connected with Fasttrack?

Michel · Thu Jun 09, 2016 10:42 am

Hi.
When i update CPE to 6.35.2 pppoe session reconnection many times

back to 6.35 resolve problem

Same here, my 2 PPOE Clients reporting error many times, I am getting also an error while I try to disable a PPOE client
"non rate limit queues are usless on this interface"

Thu Jun 09, 2016 11:41 am

CCR1072 LCD issue will be fixed with bootloader. You will need to upgrade bootloader to 3.33 version.

malbadrawy · Sun Jun 12, 2016 9:53 am

Hi All
first sorry for my English language
I am using (The Dude client - ve 4bet 3) it was working, but when I tryied to install the new version it is giving me error "Error: Connect failed:10061: No connection could be made becaus..."
Notes: I am using it for alert me up/down , tracroute , http access devices
so could any one advice
All the best

Mon Jun 13, 2016 11:50 am

http://forum.mikrotik.com/viewtopic.php?f=21&t=109309

Who is online