v6.13 released!

Fri May 16, 2014 10:18 am

What is new at 6.13

What's new in 6.13 (2014-May-15 16:03):

*) console - comments are now accepted where new command can start, that is,
where '/' or ':' characters can be used to start new command, e.g.
/interface { # comment until the end of the line
print
}
*) backup - backups by default are encrypted now (with user password).
To use backup on older versions, you should disable encryption with dont-encrypt
flag when creating it;
*) files with '.sensitive.' in the filename require 'sensitive'
permission to manipulate;
*) lcd - reduce CPU usage when displaying static screens;
*) l2tp - fixed occasional server lockup;
*) pptp - fixed memory leak;
*) sstp - fixed crashes;

Simply click “Check for updates” in QuickSet, Webfig or Winbox packages menu. If you run v5 or older, download the newest NPK package from our webpage, upload to your router, and reboot: http://www.mikrotik.com/download

NOTE:
We plan to have full release of CAPsMAN for general public in RouterOS v6.14!
Please, make all additional tests and report all persisting problems in wireless-fp package.

rextended · Fri May 16, 2014 10:31 am

STILL EXIST ON 6.13

http://forum.mikrotik.com/viewtopic.php ... 88#p416454
BUG SIGNALED FROM 6.10 AND STILL NOT FIXED???

Opened another ticket for that: [Ticket#2014041566000226] 6.12 UNFIXED BUG: user-manager profile limitation

I wait again the fix on 6.14....

ropeba · Fri May 16, 2014 11:19 am

When we can expect solving of bug "IGMP proxy stop working on upstream interface down/up"? Described in: http://forum.mikrotik.com/viewtopic.php ... mp#p412317

janisk · Fri May 16, 2014 11:22 am

When we can expect solving of bug "IGMP proxy stop working on upstream interface down/up"? Described in: http://forum.mikrotik.com/viewtopic.php ... mp#p412317

we are working on this issue. no fix yet.

Fri May 16, 2014 12:23 pm

rextended, I will check for the issue, and try to find out why it is not fixed yet.
And we will try to fix it at 6.14 version.

Fri May 16, 2014 12:36 pm

Separate topic is made of OpenVPN bridning.
Please post all your notes there,
http://forum.mikrotik.com/viewtopic.php ... 90#p426390
Thank you very much for the cooperation.

joegoldman · Fri May 16, 2014 12:39 pm

Hi Guys,

Just updated 951-2n to 6.13. I use RIP over a PPTP client interface, and as with every update since probably 6.9 or even earlier, after the first upgrade reboot, RIP routes do not become active. Another reboot of the whole board is required to get the routes back active (even tried disabling/enabling RIP before reboot).

I did report this a few versions ago but only on the forums.

Thanks,
Joe

panevdd · Fri May 16, 2014 1:42 pm

Just upgraded from 6.12 to 6.13 (x86).

From the Winbox -> Queues -> Simple Queues - the list is empty.
In the terminal/ssh it's ok.

Anybody with the same problem?

congo · Fri May 16, 2014 2:17 pm

Hello,

Does v6.13 contain the fix for [Ticket#2014050866000407] ?

Chupaka · Fri May 16, 2014 2:34 pm

[admin@TestPlace] > system clock pr
            time: 14:30:01
            date: may/16/2014
  time-zone-name: Etc/GMT-3
      gmt-offset: +03:00
[admin@TestPlace] >

"-3 is not equal to +3" is still not fixed

odge · Fri May 16, 2014 2:52 pm

Hey, I see in v6.12 you allow DNS names in a PPTP adapter instead of using a script.

But one thing you should definitely add to the interface, is the ability to select the outgoing interface or route... because in multiple connection scenarios, you still need to update the route/mangle using a script... real bummer, but an easy fix to do it on the interface itself!

slvnet · Fri May 16, 2014 3:45 pm

I'm new to MT, so maybe it was somethere on the forum...

I got and installed ROS 6.12RC20.
Now after pressed Download and Upgrade it Download and Reboot. After reboot still 6.13RC20.
Download manualy to Files -> Reboot. Still 6.13RC20.

How to upgrade to ROS 6.13 final ?

Best

Fri May 16, 2014 3:50 pm

Just upgraded from 6.12 to 6.13 (x86).

From the Winbox -> Queues -> Simple Queues - the list is empty.
In the terminal/ssh it's ok.

Anybody with the same problem?

Is it happening only on x86 boxes?
Please contact support@mikrotik.com
If you access the Queue Simple from the Webfig do you see all the simple queue rules?

rextended · Fri May 16, 2014 4:13 pm

rextended, I will check for the issue, and try to find out why it is not fixed yet.
And we will try to fix it at 6.14 version.

Very thanks!

Chupaka · Fri May 16, 2014 4:30 pm

I'm new to MT, so maybe it was somethere on the forum...

I got and installed ROS 6.12RC20.
Now after pressed Download and Upgrade it Download and Reboot. After reboot still 6.13RC20.
Download manualy to Files -> Reboot. Still 6.13RC20.

How to upgrade to ROS 6.13 final ?

it was described at the same place where RCs are available:

NOTE 2:
In some rare cases it might be impossible to do any package managment
(upgrade/downgrade/enable/disable/uninstall/install) in 6.13rc13+ rc versions,
board will simply reboot itself staying on previous version
This is fixed for the release version of 6.13 (2014-May-15 14:33) and newer:
Please, use Netinstall for upgrade/downgrade

rextended · Fri May 16, 2014 5:37 pm

I'm new to MT, so maybe it was somethere on the forum...

I got and installed ROS 6.12RC20.
Now after pressed Download and Upgrade it Download and Reboot. After reboot still 6.13RC20.
Download manualy to Files -> Reboot. Still 6.13RC20.

How to upgrade to ROS 6.13 final ?

Best

Use netinstall.

rextended · Fri May 16, 2014 5:49 pm

Just upgraded from 6.12 to 6.13 (x86).

From the Winbox -> Queues -> Simple Queues - the list is empty.
In the terminal/ssh it's ok.

Anybody with the same problem?

THIS PROBLEM DO NOT EXIST IN MY TWO x86 MACHINE.
>>> NETINSTALLED <<< (because I can not upgrade from 6.13rc23 to 6.13 final)

McTedson · Fri May 16, 2014 6:03 pm

Another winbox issue :

rextended · Fri May 16, 2014 6:10 pm

And where is the issue?
I not see any issue!

McTedson · Fri May 16, 2014 6:13 pm

Here's for comparison what it should look like.

Chupaka · Fri May 16, 2014 6:16 pm

plus, 'Unknown' should not be here - it doesn't show up in Terminal

McTedson · Fri May 16, 2014 6:17 pm

Must be some new link speed on some not yet released hardware

rextended · Fri May 16, 2014 6:21 pm

SFP & SFP+ speed...

McTedson · Fri May 16, 2014 6:28 pm

SFP & SFP+ speed...

No, SFP has a separate tab with all infos.

rextended · Fri May 16, 2014 6:30 pm

The tab are separate, but the lisnk speed, is speed...

panevdd · Fri May 16, 2014 6:49 pm

Just upgraded from 6.12 to 6.13 (x86).

From the Winbox -> Queues -> Simple Queues - the list is empty.
In the terminal/ssh it's ok.

Anybody with the same problem?
Is it happening only on x86 boxes?
Please contact support@mikrotik.com
If you access the Queue Simple from the Webfig do you see all the simple queue rules?

I'll upgrade some other devices in Monday and will see is the problem still here.
I've contacted support and got Ticket#2014051666000581.
From the Webfig the Queue list is empty, too.
I have some customers with a Cyrillic characters in their names; maybe this is a problem, but in the IP -> Firewall there is no such problem.

McTedson · Fri May 16, 2014 6:59 pm

The tab are separate, but the lisnk speed, is speed...

SFP does not have Status tab on my router.

tedkuban · Fri May 16, 2014 7:07 pm

The problem described for 6.12 is not solved:

Upgrade from 6.11 cause stop IPSec communications.

dg8ngn · Fri May 16, 2014 7:10 pm

Hi,

it seems BCP bridging (PPP tunnel bridging) did break from 6.12 to 6.13. I used http://wiki.mikrotik.com/wiki/Manual:BC ... _bridging) for implementation. Instead of PPTP I use L2TP.

Using "export" the configuration of 6.12 and 6.13 are identically. Downgrading from 6.13 to 6.12 worked. Only the "Server"-Side is affected... I can provide the configuration if necessary.

Bye,
Jann

lorsungcu · Fri May 16, 2014 7:41 pm

dg8ngn - can you go into more detail about what is broken with BCP in 6.13? I need to upgrade (apparently L2TP is broken in 6.12, which is half the reason I upgraded...), but rely pretty heavily on BCP.

slvnet · Fri May 16, 2014 7:53 pm

I got and installed ROS 6.12RC20.

How to upgrade to ROS 6.13 final ?
it was described at the same place where RCs are available:

NOTE 2:
In some rare cases it might be impossible to do any package managment
(upgrade/downgrade/enable/disable/uninstall/install) in 6.13rc13+ rc versions,
board will simply reboot itself staying on previous version
This is fixed for the release version of 6.13 (2014-May-15 14:33) and newer:
Please, use Netinstall for upgrade/downgrade

I got only link to 6.13RC20 before,
http://forum.mikrotik.com/viewtopic.php?t=84801#p425567

Thanks a lot. I'll do netinstall

steen · Fri May 16, 2014 9:52 pm

Hello Folks!

Sorry to sat that CRS and simple vlans, trunk and access ports still does not work.

I did reset configuration with no defaults and with defaults, result is the same.
I did not to a netboot install, that test remains.

However the CRS does not hang anymore when trying, otherwise it fails exactly like before.

Tested CRS examples (http://wiki.mikrotik.com/wiki/Manual:CRS_examples)
Port based VLAN [FAIL]
Inter VLAN routing [FAIL]

lorsungcu · Fri May 16, 2014 10:04 pm

Found more issues with DHCP in BCP bridging in 6.12. Going to open a ticket but dont expect a fast enough response to matter. I dont see anything in the changelogs between 5.26 and 6.12 that would cause DHCP to break as it has, can someone from mikrotik confirm if anything is different?

Rudios · Fri May 16, 2014 10:31 pm

Found more issues with DHCP in BCP bridging in 6.12. Going to open a ticket but dont expect a fast enough response to matter. I dont see anything in the changelogs between 5.26 and 6.12 that would cause DHCP to break as it has, can someone from mikrotik confirm if anything is different?

I have experienced DCHP problems with BCP myself but as far as I remember the problems were solved in 6.11 or 6.12.
I don't know what your exact problems are?
I experienced problems where devices never got an IP address from the DCHP server over a PPTP link with BCP. My issue is also mentioned here

lorsungcu · Fri May 16, 2014 11:20 pm

Yeah that sounds like what I'm seeing, although I'm using L2TP. I need 6.x for IPSec but it breaks all kinds of other stuff. This is maddening.

toni17 · Sat May 17, 2014 2:45 am

Queue Tree Global Parent still not working till 6... , This is very important for my Business .
I cant see my Huawei K4305 in system ports , so i cant send and receive sms .

Chupaka · Sat May 17, 2014 4:05 am

Queue Tree Global Parent still not working till 6...

details, please

c0d3rSh3ll · Sat May 17, 2014 4:55 am

Hello Folks!

Sorry to sat that CRS and simple vlans, trunk and access ports still does not work.

I did reset configuration with no defaults and with defaults, result is the same.
I did not to a netboot install, that test remains.

However the CRS does not hang anymore when trying, otherwise it fails exactly like before.

Tested CRS examples (http://wiki.mikrotik.com/wiki/Manual:CRS_examples)
Port based VLAN [FAIL]
Inter VLAN routing [FAIL]

Any other user can confirm this bug?
The next week I need to do port based vlan in crs trunking to a rb but this news is so bad for me...

sent from my mobile phone using tapatalk

meetriks2 · Sat May 17, 2014 8:21 am

Well 6.13 seems be the first to have no problem with much higher cpu load compare to v5.6
(or is follow the path of v6.7 and it will show up over a week or so)

http://forum.mikrotik.com/viewtopic.php?f=2&t=84729

stefan803 · Sat May 17, 2014 9:15 am

Hi,

since 6.13 I have a really bad sstp-throughput. I tried 6.9 - 6.12 yesterday and it's always been about 1 MB/s (that's about what the line can do). With 6.13 (inkl. the rc-versions) I get a maximum of about 200 kB/s.

The setup:

Lan1 Lan2
MikroTik <-> AVM Router <-> Internet <-> AVM Router <-> Mikrotik

(The AVMs need to be used as phone is voip)

Interesting: When connecting directly in Lan1 to the SSTP server the speed is fine, it's just happeining over the internet ... And did not happen before 6.13.

Do somebody know what might have changed in 6.13 that makes it beeing so slow over internet.. or that makes other components on the path beeing slower as with 6.12?

Thanks!

wichets · Sat May 17, 2014 12:25 pm

Hi report bug

i use the 6.3 with aircard 312u it is working properly but after i have upgrade to 6.13 the error massage is occurred
please see error on the attached file

rgds/wichets

Maggiore81 · Sat May 17, 2014 1:32 pm

Just upgraded from 6.12 to 6.13 (x86).

From the Winbox -> Queues -> Simple Queues - the list is empty.
In the terminal/ssh it's ok.

Anybody with the same problem?

me too.

rebooted two times and they are back.

nicklowe · Sat May 17, 2014 6:45 pm

I have noticed that the performance of SSTP is not as great as it could be. I had been hoping that things would improve with 6.13 - we were told that the OpenSSL library was being updated to the latest version.

This latest OpenSSL version implements cipher suites that use the ChaCha20-Poly1305 and AES in GCM.

Mikrotik should offer these in RouterOS with all the TLS dependent features like SSTP and use them in preference where available to drastically improve throughput and security for site-to-site tunnels.

The performance benefits of ChaCha20-Poly1305 are awesome:

http://googleonlinesecurity.blogspot.sg ... https.html
https://www.imperialviolet.org/2013/10/07/chacha20.html

It would be an easy performance win, especially on the lower end, resource constrained hardware.

For security reasons, RC4 should also be completely removed from RouterOS and definitely not used in preference anywhere - it presently is for SSTP. The stream cipher it is no longer considered secure due to biases. It is quickly being deprecated from other platforms such as Windows:

https://technet.microsoft.com/library/security/2868725 (2013)
https://technet.microsoft.com/library/security/2960358 (2014)

skibi82 · Sat May 17, 2014 7:00 pm

To fix problem use netinstall

I'm new to MT, so maybe it was somethere on the forum...

I got and installed ROS 6.12RC20.
Now after pressed Download and Upgrade it Download and Reboot. After reboot still 6.13RC20.
Download manualy to Files -> Reboot. Still 6.13RC20.

How to upgrade to ROS 6.13 final ?

Best

skibi82 · Sat May 17, 2014 7:02 pm

When there are plans to implement Xstp in CRS switch?

BinaryCrash · Sat May 17, 2014 9:51 pm

Just upgraded from 6.12 to 6.13 (x86).

From the Winbox -> Queues -> Simple Queues - the list is empty.
In the terminal/ssh it's ok.

Anybody with the same problem?
Is it happening only on x86 boxes?
Please contact support@mikrotik.com
If you access the Queue Simple from the Webfig do you see all the simple queue rules?

It happened to me also on a CCR-1016-12G.
queue simple just stop updating on winbox, then i reopen winbox and it is empty.
At terminal/ssh i can see the rules there.
In winbox and webfig the list is empty.
If i reboot the router, i can see it again, but seconds after the list get empty.

The queue is working, just the list does not show up.

I have over 1100 simple queue rules.

BinaryCrash · Sat May 17, 2014 10:00 pm

Also i would like to report two of my CCR-1016-12G got error: "file operation error" when i try to save any backup or save export file.
Same error show when i try to delete any file.

this happened to 2 routers that have more than 1000 simple queues and store graphs.
The backup on both got 19 MB on the last run, automated.

I did reset the configuration and imported using terminal and it is working again.
Only happened on the routers with over 1000 simple queues.
The other routers with just a few queues didn't have this issue.

All same model, version 6.12. Firmware ok.

Looks like the router stop writing to disk after some amount of data. Like a limit.

I still have the backups if you would like to test it.
I didn't test this problem on version 6.13 as i don't have a spare core router to test it yet.

I also got some problems using export compact and trying to restore some arp entries didn't have interface on it.
Maybe to do this we need to use verbose. Version was 6.12 when i did this.

Chupaka · Sun May 18, 2014 1:02 am

queue simple just stop updating on winbox, then i reopen winbox and it is empty.
At terminal/ssh i can see the rules there.
In winbox and webfig the list is empty.

please check not only "/queue simple print", but also "/queue simple print stats" - I sometimes have such problem on v6.7 with queue tree, winbox list is empty, and printing stats gets timeout. it sometimes happens when I change queue type on highspeed queue (>1Gbps of traffic)

BinaryCrash · Sun May 18, 2014 1:52 am

queue simple just stop updating on winbox, then i reopen winbox and it is empty.
At terminal/ssh i can see the rules there.
In winbox and webfig the list is empty.
please check not only "/queue simple print", but also "/queue simple print stats" - I sometimes have such problem on v6.7 with queue tree, winbox list is empty, and printing stats gets timeout. it sometimes happens when I change queue type on highspeed queue (>1Gbps of traffic)

/queue simple print stats
keep waiting for response and nothing happens, until i press control+C to cancel.

I have two routers.
One with 6.12 and one with 6.13
The one with 6.12 have 1300+ queues.
The other with 6.13 have 1228 queues.

Commands stats, rate, packets, bytes does not work in 6.13 terminal.

6.12 router (A):

ros code

/interface monitor-traffic aggregate
    rx-packets-per-second:     18 962
      rx-drops-per-second:          0
     rx-errors-per-second:          0
       rx-bits-per-second:  109.3Mbps
    tx-packets-per-second:     18 356
      tx-drops-per-second:          0
     tx-errors-per-second:          0
       tx-bits-per-second:  104.5Mbps

6.13 router (B):

ros code

rx-packets-per-second:    17 000
      rx-drops-per-second:         0
     rx-errors-per-second:         0
       rx-bits-per-second:  97.5Mbps
    tx-packets-per-second:    16 231
      tx-drops-per-second:         0
     tx-errors-per-second:         0
       tx-bits-per-second:  89.9Mbps

steen · Sun May 18, 2014 7:07 pm

Hello Folks!

Sorry to sat that CRS and simple vlans, trunk and access ports still does not work.

I did reset configuration with no defaults and with defaults, result is the same.
I did not to a netboot install, that test remains.

However the CRS does not hang anymore when trying, otherwise it fails exactly like before.

Tested CRS examples (http://wiki.mikrotik.com/wiki/Manual:CRS_examples)
Port based VLAN [FAIL]
Inter VLAN routing [FAIL]

Any other user can confirm this bug?
The next week I need to do port based vlan in crs trunking to a rb but this news is so bad for me...

sent from my mobile phone using tapatalk

We now also tested netinstall, result is exactly the same as before, the device does not work with this very basic vlan configuration.
Is there anybody out there who have got this switch work with vlans and trunk, if so please share how you did it ?

StubArea51 · Sun May 18, 2014 11:55 pm

Thanks for 6.13 MikroTik....looking forward to testing in our Data Centers

Mon May 19, 2014 8:46 am

[admin@TestPlace] > system clock pr
            time: 14:30:01
            date: may/16/2014
  time-zone-name: Etc/GMT-3
      gmt-offset: +03:00
[admin@TestPlace] >

"-3 is not equal to +3" is still not fixed

That this is not a bug, you should be using the "manual" menu, not changing the timezone name. Apparently there is a international timezone called "GMT-3" which sets your offset to +3. Don't touch the name, only set offset if you want custom offset. We use the timezone names from the official list: http://www.iana.org/time-zones

JanJoh · Mon May 19, 2014 10:11 am

I may be doing something silly, but it would appear that the "Hey, look at me guys, I am now a hub!"-behaviour of my CRS125 has returned after upgrading to 6.13

But the previous fix command (/interface ethernet switch port set [find] learn-restricted-unknown-sa=yes) does not seem to be accepted anymore?

So, what may i be missing?

rextended · Mon May 19, 2014 10:20 am

Code: Select all
[admin@TestPlace] > system clock pr
            time: 14:30:01
            date: may/16/2014
  time-zone-name: Etc/GMT-3
      gmt-offset: +03:00
[admin@TestPlace] > 
"-3 is not equal to +3" is still not fixed
That this is not a bug, you should be using the "manual" menu, not changing the timezone name. Apparently there is a international timezone called "GMT-3" which sets your offset to +3. Don't touch the name, only set offset if you want custom offset. We use the timezone names from the official list: http://www.iana.org/time-zones

[...]
# These entries are mostly present for historical reasons, so that
# people in areas not otherwise covered by the tz files could "zic -l"
# to a time zone that was right for their area. These days, the
# tz files cover almost all the inhabited world, and the only practical
# need now for the entries that are not on UTC are for ships at sea
# that cannot use POSIX TZ settings.
[...]
# We use POSIX-style signs in the Zone names and the output abbreviations,
# even though this is the opposite of what many people expect.
# POSIX has positive signs west of Greenwich, but many people expect
# positive signs east of Greenwich. For example, TZ='Etc/GMT+4' uses
# the abbreviation "GMT+4" and corresponds to 4 hours behind (-4) UT
# (i.e. west of Greenwich) even though many people would expect it to
# mean 4 hours ahead (+4) of UT (i.e. east of Greenwich).
#
[...]

Zone	Etc/GMT-14	+14	-	GMT-14	# 14 hours ahead of GMT
Zone	Etc/GMT-13	+13	-	GMT-13
Zone	Etc/GMT-12	+12	-	GMT-12
Zone	Etc/GMT-11	+11	-	GMT-11
Zone	Etc/GMT-10	+10	-	GMT-10
Zone	Etc/GMT-9 	+9 	-	GMT-9
Zone	Etc/GMT-8 	+8 	-	GMT-8
Zone	Etc/GMT-7 	+7 	-	GMT-7
Zone	Etc/GMT-6 	+6 	-	GMT-6
Zone	Etc/GMT-5 	+5 	-	GMT-5
Zone	Etc/GMT-4 	+4 	-	GMT-4
Zone	Etc/GMT-3 	+3 	-	GMT-3
Zone	Etc/GMT-2 	+2 	-	GMT-2
Zone	Etc/GMT-1 	+1 	-	GMT-1
Zone	Etc/GMT+1 	-1 	-	GMT+1
Zone	Etc/GMT+2 	-2 	-	GMT+2
Zone	Etc/GMT+3 	-3 	-	GMT+3
Zone	Etc/GMT+4 	-4 	-	GMT+4
Zone	Etc/GMT+5 	-5 	-	GMT+5
Zone	Etc/GMT+6 	-6 	-	GMT+6
Zone	Etc/GMT+7 	-7 	-	GMT+7
Zone	Etc/GMT+8 	-8 	-	GMT+8
Zone	Etc/GMT+9 	-9 	-	GMT+9
Zone	Etc/GMT+10	-10	-	GMT+10
Zone	Etc/GMT+11	-11	-	GMT+11
Zone	Etc/GMT+12	-12	-	GMT+12

I hope this clarifies everything permanently.

skillful · Mon May 19, 2014 10:47 am

Just upgraded from 6.12 to 6.13 (x86).

From the Winbox -> Queues -> Simple Queues - the list is empty.
In the terminal/ssh it's ok.

Anybody with the same problem?

I confirm this issue on RB1200.

Simple queues are gone in both winbox and webfig. Sometimes the queue are there but frozen in time. After a reboot, simple queues display properly in winbox for sometime before freezing or disappearing. This appears to be a display issue because the simple queues are still functional and limiting clients though.

wolfeyes · Mon May 19, 2014 11:52 am

Has anyone noticed any port flapping issue with this release on the CRS125?

samotoka · Mon May 19, 2014 11:57 am

winbox still crash when conected via pptp

bysard · Mon May 19, 2014 12:27 pm

winbox still crash when conected via pptp

Same here.

_saik0 · Mon May 19, 2014 3:37 pm

Hi,

Don't know if there should be a separate topic, but here goes.

I'm running a L2TP/IPSec/OSPF VPN between multiple MikroTik 2011UAS routers.
I'm having issues on 6.13 where I have to frequently flush SAs to reconnect dropping L2TP connections.
On 6.10 this happened from time to time, but on 6.13 it's now happening during the first 30minutes of established L2TP connection.
I can only see the message

ipsec,debug <peer-ip> give up to get IPsec-SA due to time up to wait.

The DPDs seem to be exchanged (successfully!) even after this happens, but it doesn't try to flush the old SA and establish a new one.
Only manual flush helps, and this is needed only on one side of the VPN.
This is from the side with "send initial contact" option enabled.

On the other side, there's ROS 6.6 for one tunnel and 5.14 for the other.

I've seen so many issues with l2tp/ipsec in the 6.x release (including the buffer and complete loss of connectivity) that i don't even know which version to use as a safe starting point

tedkuban · Mon May 19, 2014 4:19 pm

The problem described for 6.12 is not solved:

Upgrade from 6.11 cause stop IPSec communications.

6.11 can connect with 5.26 through IPSec, however some days later there is an error of "No buffer space available". 6.12 and 6.13 don't connect neither with 6.11, nor with 5.26. In this regard a question - Can I downgrade the CRS125 to version 5.26?

visalink · Mon May 19, 2014 5:14 pm

Code: Select all
[admin@TestPlace] > system clock pr
            time: 14:30:01
            date: may/16/2014
  time-zone-name: Etc/GMT-3
      gmt-offset: +03:00
[admin@TestPlace] > 
"-3 is not equal to +3" is still not fixed
That this is not a bug, you should be using the "manual" menu, not changing the timezone name. Apparently there is a international timezone called "GMT-3" which sets your offset to +3. Don't touch the name, only set offset if you want custom offset. We use the timezone names from the official list: http://www.iana.org/time-zones

Normis, and the following is a bug in the log or not?

Chupaka · Mon May 19, 2014 5:17 pm

Apparently there is a international timezone called "GMT-3" which sets your offset to +3.

I hope this clarifies everything permanently.

ha-ha, thanks, guys! at last, after years, I got an answer

"In order to conform with the POSIX style, those zone names beginning with "Etc/GMT" have their sign reversed from what most people expect."

visalink · Mon May 19, 2014 5:34 pm

I'm always testing the RouterOS on a large distance, mikrotik made a mess in version 6.13rc20.
There is an alternative solution to upgrade the version 6.13rc20 except with netinstal?

rextended · Mon May 19, 2014 7:13 pm

I'm always testing the RouterOS on a large distance, mikrotik made a mess in version 6.13rc20.
There is an alternative solution to upgrade the version 6.13rc20 except with netinstal?

NO.

It's not a good idea to test on large distance...
If you read the WARNING already present on 6.13rc20 download page, you know this problem.
But you have read the WARNING, true?

rextended · Mon May 19, 2014 7:15 pm

Normis, and the following is a bug in the log or not?

Follow exactly this passage:

Set timezone to manual,
click apply,
DO NOT USE "UNDO"
set the right time zone,
reboot,
done.

[this problem appear only if the 6.8+ are installed by updating old RouterOS and not appear if you use netinstall to install directly 6.8+]

rextended · Mon May 19, 2014 7:20 pm

Apparently there is a international timezone called "GMT-3" which sets your offset to +3.

I hope this clarifies everything permanently.
ha-ha, thanks, guys! at last, after years, I got an answer

"In order to conform with the POSIX style, those zone names beginning with "Etc/GMT" have their sign reversed from what most people expect."

Oh, finally

wa4zlw · Mon May 19, 2014 7:31 pm

hi folks...I was running 6.12 and switched to 6.13 and I think that was a mistake. I am using IPSec tunnels to my other sites and it seems OSPF no longer propagates across the tunnels. I have a Watchguard XTM520 running latest released code which is my hub for all tunnels.

Anyone else having similar issues? I also noticed that you have to flush manually; if a rekey comes in itjust adds them to the list and doesnt get rid of the old ones.

I need this working.

Mikrotik really needs to stop adding "features" and needs to concentrate on fixing bugs.

Leon

_saik0 · Mon May 19, 2014 7:38 pm

OSPF failing for me is just the result of L2TP tunnel terminating again as a result of IPSec failing.
Anyhow, as i described above, similar issues - IPSec indeed seems very unstable in the whole 6.x release. In fact it's getting worse with every new version.

Fully support you on that one mate, MT stop adding new stuff, fix CORE issues.

visalink · Mon May 19, 2014 9:17 pm

Normis, and the following is a bug in the log or not?
Follow exactly this passage:

Set timezone to manual,
click apply,
DO NOT USE "UNDO"
set the right time zone,
reboot,
done.

[this problem appear only if the 6.8+ are installed by updating old RouterOS and not appear if you use netinstall to install directly 6.8+]

rextended

The correct procedure would be:

Set timezone to manual,
Set manual time zone to +00:00,
click apply,
set the right time zone,
no reboot necessary,
done.

Regards,
Dzieva.

rextended · Tue May 20, 2014 1:57 am

Normis, and the following is a bug in the log or not?
Follow exactly this passage:

Set timezone to manual,
click apply,
DO NOT USE "UNDO"
set the right time zone,
reboot,
done.

[this problem appear only if the 6.8+ are installed by updating old RouterOS and not appear if you use netinstall to install directly 6.8+]
rextended

The correct procedure would be:

Set timezone to manual,
Set manual time zone to +00:00,
click apply,
set the right time zone,
no reboot necessary,
done.

Regards,
Dzieva.

I'm happy to have hinted you

Trema · Tue May 20, 2014 10:42 am

I think there is a bug in the DHCPv6 client, that was introduced in 6.12. I observed this behaviour on both a 2011UiAS-2HnD and a 750GL.

When the option "Use Peer DNS" in the DHCPv6 client is checked, the v6 DNS addresses are added to the list of DNS servers, instead of being refreshed, each time the lease is renewed.

Schermafdruk-dhcpv6-client.png

This results in an ever growing list of (v6) DNS servers:

Schermafdruk-dns.png

Beeski · Tue May 20, 2014 12:06 pm

Are the wireless improvements in 6.13rc23 included in the official 6.13 release?
Can you give us a description of these improvements?
thanks,
Mike

Tue May 20, 2014 12:21 pm

Are the wireless improvements in 6.13rc23 included in the official 6.13 release?
Can you give us a description of these improvements?
thanks,
Mike

Everything in the last RC is included in the final release.

Which wireless improvements do you mean? This is the official changelog:

What's new in 6.13 (2014-May-15 16:03):

*) console - comments are now accepted where new command can start, that is,
where '/' or ':' characters can be used to start new command, e.g.
/interface { # comment until the end of the line
print
}
*) backup - backups by default are encrypted now (with user password).
To use backup on older versions, you should disable encryption with dont-encrypt
flag when creating it;
*) files with '.sensitive.' in the filename require 'sensitive'
permission to manipulate;
*) lcd - reduce CPU usage when displaying static screens;
*) l2tp - fixed occasional server lockup;
*) pptp - fixed memory leak;
*) sstp - fixed crashes;

Beeski · Tue May 20, 2014 12:35 pm

Everything in the last RC is included in the final release.
Looking good with first tests

Which wireless improvements do you mean?

Please contact the support@mikrotik.com <mailto:support@mikrotik.com> to
get access to the v6.13rc versions so you could try out the new test
RouterOS version which has some improvements for the wireless driver
including the new wireless-fp package.

bobxiao · Tue May 20, 2014 12:36 pm

routing mark can not use

klaparp · Tue May 20, 2014 1:40 pm

After upgrade to 6.13 i get "ERROR: Internal Server Error" when trying to log in.

I have tested on 2 different routers. one 750 and onte 2011 with same result. '

Suggestions?

rextended · Tue May 20, 2014 1:48 pm

Login from where?

jsahoka · Tue May 20, 2014 3:08 pm

Maybe its just me,

I am on 6.13 and every time I click Quick Set in winbox it shows my passwords in plain text even though the "Hide Passwords" option is checked in winbox, the "hide" box next to the WiFi Passwords keeps getting unchecked and the password is shown in plaintext.

bartjoo · Tue May 20, 2014 3:38 pm

Hello, is there a release list (features, fixes, bues) for Capsman to?

Tue May 20, 2014 3:49 pm

Maybe its just me,

I am on 6.13 and every time I click Quick Set in winbox it shows my passwords in plain text even though the "Hide Passwords" option is checked in winbox, the "hide" box next to the WiFi Passwords keeps getting unchecked and the password is shown in plaintext.

Looks like a bug, I can repeat

jsahoka · Tue May 20, 2014 4:00 pm

Maybe its just me,

I am on 6.13 and every time I click Quick Set in winbox it shows my passwords in plain text even though the "Hide Passwords" option is checked in winbox, the "hide" box next to the WiFi Passwords keeps getting unchecked and the password is shown in plaintext.
Looks like a bug, I can repeat

Hey Normis,
Thanks for the reply, do i need to submit an official bug report to try and get it fixed or is this sufficient?

Tue May 20, 2014 4:01 pm

Maybe its just me,

I am on 6.13 and every time I click Quick Set in winbox it shows my passwords in plain text even though the "Hide Passwords" option is checked in winbox, the "hide" box next to the WiFi Passwords keeps getting unchecked and the password is shown in plaintext.
Looks like a bug, I can repeat
Hey Normis,
Thanks for the reply, do i need to submit an official bug report to try and get it fixed or is this sufficient?

I have already submitted a bug report, so you don't need to do that

jsahoka · Tue May 20, 2014 4:07 pm

I have already submitted a bug report, so you don't need to do that

Thanks !

wa4zlw · Tue May 20, 2014 4:24 pm

what about the OSPF/IPSec issues which also seem to be in 5.26?

klaparp · Tue May 20, 2014 5:50 pm

After upgrade to 6.13 i get "ERROR: Internal Server Error" when trying to log in.

I have tested on 2 different routers. one 750 and onte 2011 with same result. '

Suggestions?

Get same problem with both webgui and winbox.

2-3 restarts solved the problem on my testing router, I´ll get back later tonight if it works for the production router.

/HW

Beeski · Tue May 20, 2014 7:13 pm

Everything in the last RC is included in the final release.

Which wireless improvements do you mean? This is the official changelog:

What's new in 6.13 (2014-May-15 16:03):

*) console - comments are now accepted where new command can start, that is,
where '/' or ':' characters can be used to start new command, e.g.
/interface { # comment until the end of the line
print
}
*) backup - backups by default are encrypted now (with user password).
To use backup on older versions, you should disable encryption with dont-encrypt
flag when creating it;
*) files with '.sensitive.' in the filename require 'sensitive'
permission to manipulate;
*) lcd - reduce CPU usage when displaying static screens;
*) l2tp - fixed occasional server lockup;
*) pptp - fixed memory leak;
*) sstp - fixed crashes;

is wireless-fp included in the official 6.13 release?
I am guessing no based on comments in the thread below:

http://forum.mikrotik.com/viewtopic.php ... ireless+fp

Tue May 20, 2014 10:21 pm

No it isn't. You have to download it separately. But it is obvious on download webpage...

jondavy · Wed May 21, 2014 2:58 am

I did upgrade from the menu /system upgrade and were not installed some packages defaults as MPLS, then I had to manually reinstall the MPLS package

patrickmkt · Wed May 21, 2014 7:53 am

Something I just noticed after updating two routers to 6.13:

When I connect via SSTP from one ROS client to one ROS Server, the firewall rule associated with this SSTP connection stays red on the server after connection.
It was working before on previous versions. Is there a new parameter that I didn't see in the update that I need to change or is it a regression?

rextended · Wed May 21, 2014 9:17 am

I know that is off-topic but please someone of MikroTik staff, or another one ingeneer, reply on this post:

http://forum.mikrotik.com/viewtopic.php ... 95#p427295

brosky · Wed May 21, 2014 11:28 am

Just upgraded from 6.12 to 6.13 (x86).

From the Winbox -> Queues -> Simple Queues - the list is empty.
In the terminal/ssh it's ok.

Anybody with the same problem?
I confirm this issue on RB1200.

Simple queues are gone in both winbox and webfig. Sometimes the queue are there but frozen in time. After a reboot, simple queues display properly in winbox for sometime before freezing or disappearing. This appears to be a display issue because the simple queues are still functional and limiting clients though.

Same here, 6.13 on RB1100AHx2, queues are not displayed and if I restore from backup they are displayed for a while and then they are missing again.

How can I downgrade to a previous version ? Where is a repository with older versions ?

Chupaka · Wed May 21, 2014 12:13 pm

I did upgrade from the menu /system upgrade and were not installed some packages defaults as MPLS, then I had to manually reinstall the MPLS package

check /log

DjM · Wed May 21, 2014 1:15 pm

Something I just noticed after updating two routers to 6.13:

When I connect via SSTP from one ROS client to one ROS Server, the firewall rule associated with this SSTP connection stays red on the server after connection.
It was working before on previous versions. Is there a new parameter that I didn't see in the update that I need to change or is it a regression?

Try to double-click on firewall rule, reselect again the same sstp-server interface, clisk OK and it should be fine. It seems that there were some changes in sstp-server interface indexing in 6.13 .

klaparp · Wed May 21, 2014 1:44 pm

After upgrade to 6.13 i get "ERROR: Internal Server Error" when trying to log in.

I have tested on 2 different routers. one 750 and onte 2011 with same result. '

Suggestions?
Get same problem with both webgui and winbox.

2-3 restarts solved the problem on my testing router, I´ll get back later tonight if it works for the production router.

/HW

Didnt help on the production router so I downgraded to 5.26. Also noticed that BCP wont work on 6.12 or 6.13 so its pretty useless. Something major is broken I think.

dg8ngn · Wed May 21, 2014 3:49 pm

dg8ngn - can you go into more detail about what is broken with BCP in 6.13? I need to upgrade (apparently L2TP is broken in 6.12, which is half the reason I upgraded...), but rely pretty heavily on BCP.

Yes, on the L2TP-Server side the packets are not bridged anymore. Downgrade to 6.12 fixed it.

rdhw · Wed May 21, 2014 6:11 pm

on the L2TP-Server side the packets are not bridged anymore. Downgrade to 6.12 fixed it.

Yes, I see the same server-side BCP failure over PPTP. Client-side BCP OK on 6.13, but server-side BCP did not work at 6.13 and required to be down-graded to 6.12.

Clauu · Wed May 21, 2014 6:29 pm

winbox still crash when conected via pptp
Same here.

Same..

patrickmkt · Wed May 21, 2014 7:45 pm

Something I just noticed after updating two routers to 6.13:

When I connect via SSTP from one ROS client to one ROS Server, the firewall rule associated with this SSTP connection stays red on the server after connection.
It was working before on previous versions. Is there a new parameter that I didn't see in the update that I need to change or is it a regression?
Try to double-click on firewall rule, reselect again the same sstp-server interface, clisk OK and it should be fine. It seems that there were some changes in sstp-server interface indexing in 6.13 .

I already did that but still same behavior on the next reconnection. The firewall rule stays red and I have to reselect it manually again.

DjM · Wed May 21, 2014 11:19 pm

In this case I can see two options:

1) Wait for 6.14 release, there are also some SSTP bug fixes
2) Generate support file and send it to MK support

The third possible option can be selecting a different interface in firewall rule, apply configuration, then select back the sstp-server interface and apply configuration. I haven't tested this step so I'm not sure if it will help, but you can share you experience

skibi82 · Wed May 21, 2014 11:26 pm

I think 6.13 is as good firmware.
1. Do Not hangs when l2tp/sstp/pptp all versions greater than 6.7 unfortunately had fucked implementations of these tunnels (with 200 tunnels and a large movement of the device had to August reboot using the watchdog)
Apart from the odd behavior of a process and a higher load mgnt far knock on wood work

2. IPSEC for me is ok about 300 tunnels after the upgrade works correct

I can honestly say I see a light in the tunnel

gl4ck · Wed May 21, 2014 11:34 pm

In this case I can see two options:

1) Wait for 6.14 release, there are also some SSTP bug fixes
2) Generate support file and send it to MK support

The third possible option can be selecting a different interface in firewall rule, apply configuration, then select back the sstp-server interface and apply configuration. I haven't tested this step so I'm not sure if it will help, but you can share you experience

Got the same problem with L2TP interfaces, unfortunately changing the rule and changing it back to the l2tp-interface did solve the problem only until the next reconnect.
For the time being I went back to 6.12...

patrickmkt · Thu May 22, 2014 12:07 am

In this case I can see two options:

1) Wait for 6.14 release, there are also some SSTP bug fixes
2) Generate support file and send it to MK support

The third possible option can be selecting a different interface in firewall rule, apply configuration, then select back the sstp-server interface and apply configuration. I haven't tested this step so I'm not sure if it will help, but you can share you experience

Option 3 does not work either. The only way so far was to disable/enable the rule again.
I can maybe write an ugly workaround script to check the status of the sstp and then enable the firewall rules, but that's really a waste of resources.

DjM · Thu May 22, 2014 9:40 am

Workaround with script is not a good option, I think. Ask for 6.14rc version, where are sstp-server interface bindings fixed (related to another SSTP issue), and if the issue will be active also in the RC version, then the last step is to generate support file and send it to MK, I think. For me are the interface bindings working fine in firewall rules (ROS 6.13), but it doesn't mean that there is no other issue.

nick3dos · Thu May 22, 2014 10:05 am

I just recieved my new CCR1009-8G-1S
Upadate to 6.12 -> OK
Upadate to 6.13 -> Problem !!!

soulflyhigh · Thu May 22, 2014 10:22 am

A "cosmetic" bug in winbox, "full duplex" value isn't shown :

mangust · Thu May 22, 2014 11:13 am

I just recieved my new CCR1009-8G-1S
Upadate to 6.12 -> OK
Upadate to 6.13 -> Problem !!!

PPC package for CCR box ?

Thu May 22, 2014 11:17 am

I just recieved my new CCR1009-8G-1S
Upadate to 6.12 -> OK
Upadate to 6.13 -> Problem !!!
PPC package for CCR box ?

No, you must use the Tile packge for CCR

nick3dos · Thu May 22, 2014 11:21 am

I just recieved my new CCR1009-8G-1S
Upadate to 6.12 -> OK
Upadate to 6.13 -> Problem !!!
PPC package for CCR box ?
No, you must use the Tile packge for CCR

I downloaded Tile package
I tried to updated again.
with the second try, all is ok now...
I don't know what happened the first time (imported the same files !!!)

Rudios · Thu May 22, 2014 1:04 pm

A "cosmetic" bug in winbox, "full duplex" value isn't shown :

There is a second entry of "Full Duplex" in the available columns. Also there is a "Speed" column which is blank.
But the "Rate", as you see, is filled. If you choose the other "Full Duplex" column, it will also be filled.

morf · Fri May 23, 2014 5:43 pm

After the broadcast storm, stop working Vlan on CCR.
Please check.

bartjoo · Fri May 23, 2014 9:32 pm

Hello everyone, i use al lot off hotspots in my network.
but when i upgrade my routers to the latest version, the upgrade have some problems.

The ip addresses are active the bridge and change after the upgrade to a WLAN connection
OR
The dhcp server was active on the bridge and after the upgrade its on a WLAN connection.

So after the upgrade, some functions didnt work well anymore. I had to change it by meself.

Erastus · Sat May 24, 2014 5:18 am

Upgraded 5 493 two are hanging. Need to drive to reboot.
The old ways are bagain.
Carefull before you upgrade. Systems do hang when you upgrading to Ver 13

steen · Sat May 24, 2014 12:33 pm

Hello Folks!

At RoS6.7 L2TP was working properly, after RoS6.7 L2TP stops transferring data after anything between 70Kb to 13MB.
The same goes for V6.13.

It is for all RB platforms, RB411, RB433, RB333, RB600, RB2011UAS, RB750 and so on.

Anyone else who had experiense of problem with L2TP ?

I will put a supout and make ticket to MT about it, we are now stuck to RoS6.7 and have to rollback entire environment, nothing works...

bartjoo · Sat May 24, 2014 12:36 pm

Hello Folks!

At RoS6.7 L2TP was working properly, after RoS6.7 L2TP stops transferring data after anything between 70Kb to 13MB.
The same goes for V6.13.

It is for all RB platforms, RB411, RB433, RB333, RB600, RB2011UAS, RB750 and so on.

Anyone else who had experiense of problem with L2TP ?

I will put a supout and make ticket to MT about it, we are now stuck to RoS6.7 and have to rollback entire environment, nothing works...

Its look like, i have the same problem, but with PPTP. Yesterday i upgraded my hotspots from 6.7 to 6.13 but the connections were bad. So i downgrade everything to 6.7 again and no problems anymore.

steen · Sat May 24, 2014 1:17 pm

Hello Folks!

At RoS6.7 L2TP was working properly, after RoS6.7 L2TP stops transferring data after anything between 70Kb to 13MB.
The same goes for V6.13.

It is for all RB platforms, RB411, RB433, RB333, RB600, RB2011UAS, RB750 and so on.

Anyone else who had experiense of problem with L2TP ?

I will put a supout and make ticket to MT about it, we are now stuck to RoS6.7 and have to rollback entire environment, nothing works...

Its look like, i have the same problem, but with PPTP. Yesterday i upgraded my hotspots from 6.7 to 6.13 but the connections were bad. So i downgrade everything to 6.7 again and no problems anymore.

I put a supout file to Support, and did a test transfering a file in the tunnel and at same time I was logged in to the target without tunnel making that supout and took some screenshots on what is going on there.

I saw the rb device become "busy" for some time right after traffic stopped, and it simply throws out the tunneled winbox session.

I did downgrade the device to RoS6.7 again but left firmware to 3.14, it then stabilized. Shortly after whole l2tp infrastructure collapsed and I did downgrade all of them.

pcunite · Sun May 25, 2014 4:13 pm

v6.13 has been good for me so far, ping replies for example are 10ms quicker. Overall seems to be a faster release.

theprism · Mon May 26, 2014 3:19 am

6.13 BAD in my case!

firewall rules that are written for an inactive L2TP are not active - red, which is normal.
But once L2TP became connected the firewall rule isn't activated, so it remains red and logs are filled with input errors on the l2tp interface.
If I recreate the same rule, the old one became active now - until the next reboot.

Downgrading...

6.12 was bad for my RB951G-2HnD. It just keeps crashing the kernel on the Routerboard for unknown reasons. Even access to the routerboard dissapears and any accesses outside it became dead. 6.12 works just fine on my RB751G-2HnD and RB751U-2HnD.
Won't try 6.13 there.

Haven't tried 6.11-6.2.
6.1 was a nightmare for L2TP+IPSEC.

Downgrading back to stable 6.0 (for me).

CStrauch · Mon May 26, 2014 3:51 pm

Anyone tried 802.3ad (trunk) on CRS?

With 2 CRS-125 it worked fine but when trying to connect with a HP V1910 (Link Aggregation / Dynamic) I can't get the ports to sync using the new "trunk" option on CRS.
Also I tried to sync CRS's trunk with CCR's bond with no luck.

The BOND feature syncs up with V1910 Link Aggregation, but on CRS-125 it maxes the CPU up.

machack · Mon May 26, 2014 4:27 pm

In Winbox, interface shows traffic intermittently. from snmp has the same problem.

Belgarion186 · Mon May 26, 2014 4:30 pm

Anyone tried 802.3ad (trunk) on CRS?

With 2 CRS-125 it worked fine but when trying to connect with a HP V1910 (Link Aggregation / Dynamic) I can't get the ports to sync using the new "trunk" option on CRS.
Also I tried to sync CRS's trunk with CCR's bond with no luck.

The BOND feature syncs up with V1910 Link Aggregation, but on CRS-125 it maxes the CPU up.

You'll need to set it to Static Link Aggregation. Dynamic doesn't work with non MT devices (tested with Cisco & Intel).

CStrauch · Mon May 26, 2014 8:23 pm

Anyone tried 802.3ad (trunk) on CRS?

With 2 CRS-125 it worked fine but when trying to connect with a HP V1910 (Link Aggregation / Dynamic) I can't get the ports to sync using the new "trunk" option on CRS.
Also I tried to sync CRS's trunk with CCR's bond with no luck.

The BOND feature syncs up with V1910 Link Aggregation, but on CRS-125 it maxes the CPU up.
You'll need to set it to Static Link Aggregation. Dynamic doesn't work with non MT devices (tested with Cisco & Intel).

So, the trunk is NOT 802.3ad.

BTW, I have a Dynamic Link Aggregation from a HP V1910 and a Planet Switch working just fine... And V1910 and MT Bond.

Thanks!

mmc1800 · Mon May 26, 2014 8:50 pm

Problem: After upgrade to 6.13 WinBox session will not stay open.

After upgrading an OmniTIK from I think it was 5.19 that I access over an L2TP VPN I cannot keep a WinBox session open.

It drops the session and disconnects after a few seconds. I did a firmware upgrade after I did the 6.13 upgrade, so now it is at firmware 3.13.

I can keep connected to the OmniTIK just fine with SSH, and I do not lose connections to older versions of winbox that are connected through that OmnitTIK, so there is no problem with the link, or the VPN or anything connection related, the problem is only with keeping a WinBox session active to the newly upgraded device.

This is not mission critical as I can manage the OmniTIK through the ssh interface, but it is annoying, and I am not going to upgrade anything else until I can figure out how to resolve this issue.

Thanks for any ideas!

Belgarion186 · Tue May 27, 2014 3:30 am

So, the trunk is NOT 802.3ad.

BTW, I have a Dynamic Link Aggregation from a HP V1910 and a Planet Switch working just fine... And V1910 and MT Bond.

Thanks!

I believe 802.3ad specs cover both SLA and dynamic. Took me quite a while of trial and error before I got my Teaming on my Intel Pro/1000 PT Dual Port to work with the CRS226.

BTW, setting the Intel to 802.3ad Dynamic will literally crash the CRS226 needing a reboot. Setting Cisco SG200 switch to Dynamic will not crash CRS226 but the link doesn't pass traffic.

I'm just surprised why Interface-Bonding is 802.3ad Dynamic but Port-Trunking is static.

kei888 · Tue May 27, 2014 4:48 am

Hi folks!

Can someone teach me how I can convert one Ethernet port of Mikrotik 1100AH PPC router to be a TRUNK port and pass tagged VLANS?

Desired set-up is:

1100AH's ether2 --- trunk ---- Cisco switch.

Thank you.

onnoossendrijver · Tue May 27, 2014 11:18 am

You can add a 'vlan interface' to a physical port and give that interface a specific vlan tag.
Then you can do whatever you want with that interface. All traffic from that interface will have the specified vlan tag on the underlying physical interface.

kei888 · Tue May 27, 2014 11:53 am

So it's not anymore necessary to create a bridge interface then add the port of ether2(trunk) and the vlans?

Thanks.

SwissWISP · Tue May 27, 2014 12:35 pm

So it's not anymore necessary to create a bridge interface then add the port of ether2(trunk) and the vlans?

It depends on what you want to do. In general, this isn't necessary and it wasn't before.

CStrauch · Tue May 27, 2014 5:18 pm

So, the trunk is NOT 802.3ad.

BTW, I have a Dynamic Link Aggregation from a HP V1910 and a Planet Switch working just fine... And V1910 and MT Bond.

Thanks!
I believe 802.3ad specs cover both SLA and dynamic. Took me quite a while of trial and error before I got my Teaming on my Intel Pro/1000 PT Dual Port to work with the CRS226.

BTW, setting the Intel to 802.3ad Dynamic will literally crash the CRS226 needing a reboot. Setting Cisco SG200 switch to Dynamic will not crash CRS226 but the link doesn't pass traffic.

I'm just surprised why Interface-Bonding is 802.3ad Dynamic but Port-Trunking is static.

I actually tried setting the V1910 for static and it seemed to work. The problem is when it's static and something happens you loose conectivity (at least in my case, since the interfaces are not directly connected to each other).

I think the port trunking still needs a little work....

rextended · Tue May 27, 2014 8:36 pm

PLEASE ADD THIS FEATURE ON ROUTEROS:

STOP (by option on/off) TO SEND THESE F(self-censored) HTTP HEADERS FROM WEB-PROXY:

HTTP_VIA 1.1 <public_ip_on_mikrotik_wan> (Mikrotik HttpProxy)
HTTP_X_PROXY_ID <routerboard_serial_number>
HTTP_X_FORWARDED_FOR <true_ip_of_the_computer>

OR EVERYTIME WE GET THIS ON ALL COMPUTER OF ALL USERS:

rextended · Tue May 27, 2014 9:12 pm

OHHHHHHHHHHHHHHHHHHHHHHH!!!!

YOU READ ON MY MIND???

*) proxy - added 'anonymous' option which will skip adding X-* and Via headers;

RouterOS +100

THANKS!!!

dohmniq · Tue May 27, 2014 10:39 pm

Anyone else using SNMP to query registration table entries?

I've raised Ticket#2014051566000304 asking for tx-ccq, rx-ccq and distance to be made available (ap would be nice too). Not sure what the procedure is for people to add support to a new feature but here's at least a ticket number you can quote.

Also, some SNMP responses, specifically MAC addresses in strings, could be normalized to reduce coding/processing load:

MIKROTIK-MIB::mtxrWlApBssid.2 = STRING: 0:c:42:f1:c4:95

It would help me if this had leading zeros like %02d:

MIKROTIK-MIB::mtxrWlApBssid.2 = STRING: 00:0c:42:f1:c4:95

Case conversion isn't an issue but parsing mac-addresses just to pad with leading zeros is more code I could do without writing - especially if it just mean editing a printf-style format string by MikroTik. Would this be of any use to anyone else OR would this actually BREAK anyone's existing production setup? (I'd guess not)

Thanks

Chupaka · Tue May 27, 2014 10:59 pm

PLEASE ADD THIS FEATURE ON ROUTEROS:

STOP (by option on/off) TO SEND THESE F(self-censored) HTTP HEADERS FROM WEB-PROXY:

OR EVERYTIME WE GET THIS ON ALL COMPUTER OF ALL USERS:

how do those headers affect that behaviour? we don't use proxy, but some users face this situation if there are many users NATted behind one ip address

rextended · Tue May 27, 2014 11:27 pm

PLEASE ADD THIS FEATURE ON ROUTEROS:

STOP (by option on/off) TO SEND THESE F(self-censored) HTTP HEADERS FROM WEB-PROXY:

OR EVERYTIME WE GET THIS ON ALL COMPUTER OF ALL USERS:
how do those headers affect that behaviour? we don't use proxy, but some users face this situation if there are many users NATted behind one ip address

I have behind nat some "low profile" users, ~200 user, never have this problem, but when I use web-proxy only for 10, 20 users, this problem come out, because Google not like proxy... is assuming that are bots if see HTTP_VIA header...

rextended · Tue May 27, 2014 11:29 pm

Anyone else using SNMP to query registration table entries?

I've raised Ticket#2014051566000304 asking for tx-ccq, rx-ccq and distance to be made available (ap would be nice too). Not sure what the procedure is for people to add support to a new feature but here's at least a ticket number you can quote.

Also, some SNMP responses, specifically MAC addresses in strings, could be normalized to reduce coding/processing load:
Code: Select all
MIKROTIK-MIB::mtxrWlApBssid.2 = STRING: 0:c:42:f1:c4:95
It would help me if this had leading zeros like %02d:
Code: Select all
MIKROTIK-MIB::mtxrWlApBssid.2 = STRING: 00:0c:42:f1:c4:95
Case conversion isn't an issue but parsing mac-addresses just to pad with leading zeros is more code I could do without writing - especially if it just mean editing a printf-style format string by MikroTik. Would this be of any use to anyone else OR would this actually BREAK anyone's existing production setup? (I'd guess not)

Thanks

In my opinion the right behavior must be: all you can obtain (read) from API or script, you must also able to read from OID...
Not only some specific set of values.

dohmniq · Wed May 28, 2014 12:36 am

In my opinion the right behavior must be: all you can obtain (read) from API or script, you must also able to read from OID...
Not only some specific set of values.

That would be the ideal situation, I agree!

Also, some SNMP responses, specifically MAC addresses in strings, could be normalized to reduce coding/processing load:
Code: Select all
MIKROTIK-MIB::mtxrWlApBssid.2 = STRING: 0:c:42:f1:c4:95
It would help me if this had leading zeros like %02d:
Code: Select all
MIKROTIK-MIB::mtxrWlApBssid.2 = STRING: 00:0c:42:f1:c4:95
Case conversion isn't an issue but parsing mac-addresses just to pad with leading zeros is more code I could do without writing - especially if it just mean editing a printf-style format string by MikroTik. Would this be of any use to anyone else OR would this actually BREAK anyone's existing production setup? (I'd guess not)

Thanks

Ignore my above rubbish - the formatting was done by snmpwalk using the SNMPv2-TC MIB but when I actually get the data it's an octet string so I can format it how I like - d'oh!

skibi82 · Wed May 28, 2014 10:16 am

Request a positive one variable informing about the configuration sequence number.
This variable should be automatically updated eg enum or something similar, gets called reconfiguration. It was applicable to the case of automatic copy.
Very useful for synchronizing devices in the cluster.

enc · Wed May 28, 2014 12:47 pm

In this case I can see two options:

1) Wait for 6.14 release, there are also some SSTP bug fixes
2) Generate support file and send it to MK support

The third possible option can be selecting a different interface in firewall rule, apply configuration, then select back the sstp-server interface and apply configuration. I haven't tested this step so I'm not sure if it will help, but you can share you experience
Option 3 does not work either. The only way so far was to disable/enable the rule again.
I can maybe write an ugly workaround script to check the status of the sstp and then enable the firewall rules, but that's really a waste of resources.

@MikroTik: is this fixed in 6.14?

DjM · Thu May 29, 2014 12:09 pm

In this case I can see two options:

1) Wait for 6.14 release, there are also some SSTP bug fixes
2) Generate support file and send it to MK support

The third possible option can be selecting a different interface in firewall rule, apply configuration, then select back the sstp-server interface and apply configuration. I haven't tested this step so I'm not sure if it will help, but you can share you experience
Option 3 does not work either. The only way so far was to disable/enable the rule again.
I can maybe write an ugly workaround script to check the status of the sstp and then enable the firewall rules, but that's really a waste of resources.
@MikroTik: is this fixed in 6.14?

@enc:
I was able to replicate the issue with firewall rules only in case that sstp-client binded into sstp-server interface and sstp-interface used in firewall rule was ROS in versions range 6.8 - 6.13 (including). SSTP clients 6.7, 5.26, 6.14rc10 are working fine for me, SSTP server used in testing environment was 6.13 and 6.14rc10. Based on this it seems that it should be connected to confirmed SSTP bug in versions 6.8 - 6.13 (including).

Which ROS version are your SSTP clients using?

enc · Thu May 29, 2014 1:16 pm

@MikroTik: is this fixed in 6.14?
@enc:
I was able to replicate the issue with firewall rules only in case that sstp-client binded into sstp-server interface and sstp-interface used in firewall rule was ROS in versions range 6.8 - 6.13 (including). SSTP clients 6.7, 5.26, 6.14rc10 are working fine for me, SSTP server used in testing environment was 6.13 and 6.14rc10. Based on this it seems that it should be connected to confirmed SSTP bug in versions 6.8 - 6.13 (including).

Which ROS version are your SSTP clients using?

ROS 6.13

the situation is even worst with the stability of my windows sstp clients. the connection drops after some 2-5min.

we use our ccr since 4 month in production now and i have to say the vpn impl is absolutely crap. bugs over bugs an no stable release in sight. how can this be...

DjM · Thu May 29, 2014 1:49 pm

@MikroTik: is this fixed in 6.14?
@enc:
I was able to replicate the issue with firewall rules only in case that sstp-client binded into sstp-server interface and sstp-interface used in firewall rule was ROS in versions range 6.8 - 6.13 (including). SSTP clients 6.7, 5.26, 6.14rc10 are working fine for me, SSTP server used in testing environment was 6.13 and 6.14rc10. Based on this it seems that it should be connected to confirmed SSTP bug in versions 6.8 - 6.13 (including).

Which ROS version are your SSTP clients using?
ROS 6.13

the situation is even worst with the stability of my windows sstp clients. the connection drops after some 2-5min.

we use our ccr since 4 month in production now and i have to say the vpn impl is absolutely crap. bugs over bugs an no stable release in sight. how can this be...

@enc:
Try to use SSTP client with version out of range 6.8 - 6.13 (included) if the issue will still persist on SSTP server, where the firewall rules are implemented.

Just to give a full information, I'm providing you my test environment results:
sstp-server1 - 6.14rc10, firewall rules with sstp-server interface binding used
sstp-server2 - 6.13, firewall rules with sstp-server interface binding used
sstp-client1 - 6.14rc10
sstp-client2 - 6.13
sstp-client3 - 6.7

I have recorded issue with invalid firewall rules only in case, that one or more sstp-server interfaces was used with client 6.13, other firewall rules, where sstp-server interfaces with clients 6.14rc10 and 6.7 were used I haven't seen the issue. Reported issue with Windows 7 SSTP client I'm not able to replicate as I don't have Windows 7 test client at this moment.

Neilson · Thu May 29, 2014 2:37 pm

Is anyone else seeing their Winbox crash when they double click to open the WLAN1 interface on the interfaces tab on the first tab of the Wireless window?

I have noticed this on some Sextant G models when updated to ROS6.13 and with the latest firmware for it.

Looking to see if others have experienced it or if not I will submit some supouts to try trace it down.

I have experienced it on radios doing both 802.11 and NV2.

I don't experience the crash when I am looking at individual radio connection details by double clicking on the registered clients.

This has been tried both on Mac under latest wine 1.7.19 and also on a windows 7 VM (up to date) with winbox running.

Under wine it locks up the window totally, on windows it just crashed the whole winbox and drops you back to the desktop.

Regards
Alexander

DjM · Thu May 29, 2014 2:45 pm

Is anyone else seeing their Winbox crash when they double click to open the WLAN1 interface on the interfaces tab on the first tab of the Wireless window?

I have noticed this on some Sextant G models when updated to ROS6.13 and with the latest firmware for it.

Looking to see if others have experienced it or if not I will submit some supouts to try trace it down.

I have experienced it on radios doing both 802.11 and NV2.

I don't experience the crash when I am looking at individual radio connection details by double clicking on the registered clients.

This has been tried both on Mac under latest wine 1.7.19 and also on a windows 7 VM (up to date) with winbox running.

Under wine it locks up the window totally, on windows it just crashed the whole winbox and drops you back to the desktop.

Regards
Alexander

Winbox -> Wireless -> Interfaces -> wlan1 => no issue, no crash
ROS 6.13, RB433UAH, Windows XP

Try to connect without "Load Previous Session" checked in winbox login dialog.

Chupaka · Thu May 29, 2014 2:51 pm

ROS 6.13

the situation is even worst with the stability of my windows sstp clients. the connection drops after some 2-5min.

seems like it's known problem:

What's new in 6.14rc10 (2014-May-26 14:52):

*) sstp - fixed problem where session was closed every 2min;

Neilson · Thu May 29, 2014 2:52 pm

Thanks, that helped when loading directly through winbox, do I need to open and close it successfully a few times to clear the "last session" information from loading it in the dude?

I am getting this as the error when in Wine:

ACCESS VIOLATION EXCEPTION at address: 6249cfda

eip=6249cfda eflags=210202
edi=0 esi=adcde0 ebp=adcdb8 esp=adcd80
eax=adcde0 ebx=adcd98 ecx=0 edx=0

backtrace:
6249d035 624a72ab 624ac3f0 624ade46 624a4907 624508cf 624b077b 6252adaf 6249b396 6249b47a 624592ad 62498c5e 6246d5cb 6246eb1a 6246ee26 426b207a 426b32db 426b3493 42682d98 6246af63 6246afcc 624cc6ca 5037ca

stack:
8 0 0 0 2 240000 0 0 adcdf8 16e7240 adcdc8 adcde0 6258664c 16e7240 adcdc8 6249d035

linked to KERNEL32!
modules:
400000 dude.exe
7bc10000 ntdll.dll
7b810000 KERNEL32.dll
68dc0000 libcairo-2.dll
64f80000 libfontconfig-1.dll
68f40000 libexpat-1.dll
42410000 msvcrt.dll
64880000 libfreetype-6.dll
65300000 libpixman-1-0.dll
69c80000 libpng12-0.dll
424b0000 gdi32.dll
425b0000 advapi32.dll
404f0000 msimg32.dll
42620000 user32.dll
420d0000 version.dll
65340000 libgdk_pixbuf-2.0-0.dll
6d580000 libgio-2.0-0.dll
685c0000 libglib-2.0-0.dll
43010000 ole32.dll
42770000 rpcrt4.dll
43130000 shell32.dll
43370000 shlwapi.dll
43510000 ws2_32.dll
6dd00000 libgmodule-2.0-0.dll
63a40000 libgobject-2.0-0.dll
6c240000 libjpeg-62.dll
6d140000 libnetsnmp-15.dll
68e40000 librsvg-2-2.dll
66440000 libcroco-0.6-3.dll
70f40000 libxml2-2.dll
65580000 libpango-1.0-0.dll
6d4c0000 libpangocairo-1.0-0.dll
6d700000 libpangoft2-1.0-0.dll
6b280000 libpangowin32-1.0-0.dll
43540000 comctl32.dll
43640000 comdlg32.dll
43740000 winspool.drv
43780000 iphlpapi.dll
437b0000 winmm.dll
43860000 msacm32.dll
43b40000 winemac.drv
452a0000 uxtheme.dll
62440000 roteros.dll
46370000 wininet.dll
467e0000 mpr.dll
65f80000 advtool.dll
6ac40000 dhcp.dll
67180000 ipv6.dll
69040000 mpls.dll
6e0c00
local msgs:
wnd=1101cc msg=203, wParam=1, lParam=4a003e
wnd=1101cc msg=20, wParam=1101cc, lParam=2010001
wnd=40206 msg=21, wParam=40206, lParam=2010001
wnd=201a2 msg=21, wParam=40206, lParam=2010001
wnd=1501ca msg=21, wParam=40206, lParam=2010001
wnd=1101cc msg=21, wParam=40206, lParam=2010001
wnd=40206 msg=210, wParam=201, lParam=de00cb
wnd=201a2 msg=210, wParam=201, lParam=c20043
wnd=1501ca msg=210, wParam=201, lParam=970043
wnd=1101cc msg=84, wParam=0, lParam=10b00cf
wnd=40206 msg=f, wParam=0, lParam=0
wnd=20166 msg=113, wParam=121, lParam=0
wnd=1101cc msg=215, wParam=0, lParam=0
wnd=1101cc msg=202, wParam=0, lParam=4a003e
wnd=24019a msg=d, wParam=10, lParam=ccb348
wnd=24019a msg=e, wParam=0, lParam=0

code:
8b 07 89 5c 24 04 89 14 24 ff 50 0c 50 50 8b 45

DjM · Thu May 29, 2014 3:13 pm

ROS 6.13

the situation is even worst with the stability of my windows sstp clients. the connection drops after some 2-5min.
seems like it's known problem:
Code: Select all
What's new in 6.14rc10 (2014-May-26 14:52):

*) sstp - fixed problem where session was closed every 2min;

Yes, it's confirmed bug - SSTP is broken in ROS 6.8 - 6.13 (including), if there is sstp-server interface binding used. Some details are in http://forum.mikrotik.com/viewtopic.php?f=2&t=82578

DjM · Thu May 29, 2014 3:20 pm

Thanks, that helped when loading directly through winbox, do I need to open and close it successfully a few times to clear the "last session" information from loading it in the dude?

K+1?

For loading from Dude I'm not sure how is the winbox internally called from this inteface, but you can try

Another option can be to delete temporary files for winbox stored "Application Data" directory (try to find folder with name "Mikrotik" on system drive in Windows) and delete everything except winbox.cfg to do not lost configured hosts. Option "last session" is probably deleting data on ROS side, second described option on client's side. If anything of this will not help, then try to get in contact with MK support how to clean the session data on both sides.

TymeWyse · Fri May 30, 2014 12:03 am

Yes, after updating (x86) to 6.13, simple queues disappear. Re-boot, they show up and slowley disappear again.

Jeff
-----------------

Just upgraded from 6.12 to 6.13 (x86).

From the Winbox -> Queues -> Simple Queues - the list is empty.
In the terminal/ssh it's ok.

Anybody with the same problem?

Chupaka · Fri May 30, 2014 10:00 am

also known problem:

What's new in 6.14rc10 (2014-May-26 14:52):

*) fixed - simple queues could sometimes crash router;
*) fixed - simple queue stats freeze (empty winbox queue window);

kapode · Fri May 30, 2014 10:30 am

The same problem with queues on my CCR router.
They r invisible.

Vitokhv · Fri May 30, 2014 4:09 pm

How to split speed client access to the global network and local resources provider:
http://asp24.com.ua/blog/mikrotik-queue ... lno-video/
http://asp24.com.ua/blog/kak-razdelit-s ... rovajdera/

It works on version 5 why does not work on version 6

I need to divide the network speed from 0 to 100 Mbps and a separate share of internet traffic from 0 to 5 Mbit

Internet to 5Mbit (0.0.0.0/0)
Local resources to 100Mbps (10.0.0.0/8; 172.0.0.0/21)

Mikrotik RB751U

gondim · Fri May 30, 2014 4:29 pm

also known problem:

What's new in 6.14rc10 (2014-May-26 14:52):

*) fixed - simple queues could sometimes crash router;
*) fixed - simple queue stats freeze (empty winbox queue window);

Hi,

Can you send a link for download?

Thanks and best regards

Fri May 30, 2014 4:30 pm

this version only available to developer testing program members

Spirch · Fri May 30, 2014 7:05 pm

is it possible to get current changelog of 6.14, all rc?

patrikg · Fri May 30, 2014 7:12 pm

http://www.mikrotik.com/download/share/ ... evnote.txt

Mibsbe
http://www.mikrotik.com/download/share/ ... latest.npk

Tile
http://www.mikrotik.com/download/share/ ... latest.npk

x86
http://www.mikrotik.com/download/share/ ... latest.npk

PPC
http://www.mikrotik.com/download/share/ ... latest.npk

Spirch · Fri May 30, 2014 8:29 pm

http://www.mikrotik.com/download/share/ ... evnote.txt

thanks

ndbjorne · Fri May 30, 2014 11:02 pm

Hello,
what about adding a

Known Issues:
*) ...

section to the

What's new in x.yy (%Y-%b-%d %H:%M):
*) ...

release log just to make not waste time to upgrade systems that would surely require an almost immediate downgrade?
And to avoid all the "still exists..." posts, too?

Regards

Erastus · Sat May 31, 2014 8:12 am

Upgraded 5 493 two are hanging. Need to drive to reboot.
The old ways are bagain.
Carefull before you upgrade. Systems do hang when you upgrading to Ver 13

Had the same problem

Shiro · Sat May 31, 2014 11:17 pm

Is it normal for PPP/L2TP/SSTP connection to drop and reconnect after adding a comment to the interface? can't remember seeing this behavior on older version of routeros.

rextended · Sun Jun 01, 2014 12:50 am

STILL EXIST ON 6.13

http://forum.mikrotik.com/viewtopic.php ... 88#p416454
BUG SIGNALED FROM 6.10 AND STILL NOT FIXED???

Opened another ticket for that: [Ticket#2014041566000226] 6.12 UNFIXED BUG: user-manager profile limitation

I wait again the fix on 6.14....

Still no fixed on 6.14rc25

Chupaka · Sun Jun 01, 2014 1:56 pm

It works on version 5 why does not work on version 6

I need to divide the network speed from 0 to 100 Mbps and a separate share of internet traffic from 0 to 5 Mbit

Internet to 5Mbit (0.0.0.0/0)
Local resources to 100Mbps (10.0.0.0/8; 172.0.0.0/21)

'target' value is now mandatory, set it at least to your LAN interface, and then add 'Dst' with 10.0.0.0/8 and 172.0.0.0/21 values

theprism · Sun Jun 01, 2014 8:19 pm

Format USB doesn't work in v6.13 through GUI. Through CLI is OK.

paoloaga · Sun Jun 01, 2014 8:45 pm

Is it normal that users can connect with MPPE even if it's not enabled in the PPP profile?

> /ppp profile export
# jun/01/2014 19:42:38 by RouterOS 6.13
/ppp profile
add change-tcp-mss=yes dns-server=***.***.***.***,***.***.***.*** local-address=\
***.***.***.*** name=pppoe-data only-one=yes remote-address=intercom-data \
use-compression=no use-encryption=no use-ipv6=no use-mpls=no \
use-vj-compression=no

> /interface pppoe-server print detail where user="*******"
Flags: X - disabled, D - dynamic, R - running
21 DR name="<pppoe-********>" user="********" service="intercom-broadband" mtu=1492 mru=1500
remote-address="**:**:**:**:**:**" encoding="MPPE128 stateless/none" uptime=5m57s

It happens with v6.13, didn't try with earlier versions. If it is a bug, should I open a ticket or is this post enough?

Thanks.

yozz · Sun Jun 01, 2014 10:20 pm

hello!

not working ROUTING-BGP-PEERS-TCP MD5 KEY

payday · Mon Jun 02, 2014 8:03 pm

Is it normal for PPP/L2TP/SSTP connection to drop and reconnect after adding a comment to the interface? can't remember seeing this behavior on older version of routeros.

Unfortunately yes, it's normal. It happens also on other interfaces (i. e. wireless disconnects all clients when you change it's comment).

morf · Mon Jun 02, 2014 9:12 pm

Hello! Have CCR 1036.
Unexpectedly lost all BGP settings.
Helped restart. Ticket # 2014060266000809

napismizpravu · Mon Jun 02, 2014 9:37 pm

Problém se zobrazením address listu ve WinBoxu při záznamu více jak 166 000 IP adres z několikadenního DDoS útoku, CPU 100%. Taky je problém s exportem toho seznamu na USB. (RB433UAH RouterOS 6.14rc10). Poslední funkční export a zobrazení byl při 144 000 záznamech. (czech).

WinBox problem with displaying big Address List 166 000 records - CPU 100%, export address list to USB CPU 100% (graf)

rextended · Fri Jun 06, 2014 1:00 am

Tomorrow when 6.14 go out, i add again this on the new topic about 6.14:

STILL EXIST ON 6.14

http://forum.mikrotik.com/viewtopic.php ... 88#p416454
BUG SIGNALED FROM 6.10 AND STILL NOT FIXED???

Opened another ticket for that: [Ticket#2014041566000226] 6.12 UNFIXED BUG: user-manager profile limitation

I wait again the fix on 6.15... I'm waiting the fix from 6.10...

steen · Fri Jun 06, 2014 8:46 am

Hello Folks!

We successfully updated whole wireless infrastructure today containing of RB411, RB600, SXT and SEXTANT.
No problems observed so far, also routing tags etc. is working.

Yet we have fully failed to upgrade our VPN routers and VPN infrastructure which is using L2TP, devices does not matter.
L2TP tunnels simply stop passing traffic after some few megabytes has passed through tunnels. It was working till RoS6.7 after that not working anymore. We have put supout files to MT, and got some advices doing more logging, yet no time to do that experiment due to the badwill it caused among customers last time.

CRS using vlans and trunks is now working to a satisfaction level, we have a minor notification that mt has fixed in some pre-release, autosupout file is generated when you reboot

so we will wait put them in production till the fix is official in nextcoming RoS version.

nicklowe · Fri Jun 06, 2014 10:25 am

Please do not release 6.14 with SSTP still broken. 6.13 has terrible performance/throughput compared to 6.12 and this issue has persisted to the pre-releases of 6.14.

http://forum.mikrotik.com/viewtopic.php?f=1&t=85568

It would also make an abundance of sense to update OpenSSL to fix the recently disclosed vulnerabilities.

voxframe · Fri Jun 06, 2014 4:01 pm

Agreed!

DO NOT RELEASE 6.14 UNTIL THE ISSUES IN THIS THREAD ARE FIXED!!!!!!

Take as much time as you need to test and sort these issues, do not make US the testers!

fernandolcx · Fri Jun 06, 2014 5:28 pm

Found a problem with Proxy ARP.
RouterOS is delivering his own MAC Address to clients even when it does not know the route to the destination.

Issue described here: http://forum.mikrotik.com/viewtopic.php?f=13&t=85054

kobuki · Fri Jun 06, 2014 10:30 pm

I've upgraded an RB2011L-IN to 6.13 a few days ago, and I' observing a strange CPU behaviour. The average CPU usage is higher by about 2-3% and there are randomly repeating very short 100% usage spikes, without any significant traffic or other measurable activity (in the middle of the night for example).

I'm not very concerned about the 2-3% additional average load, but the spikes are weird. Did anyone experience anything similar?

yozz · Sat Jun 07, 2014 7:51 pm

yesterday when i came home, my ros was be destroyed!!! statisic on usb flash drive was removed from 1 yer!!!! and menu ppp was removed form winbox! when i do system restore from backup it was not worked. than i do system hardware reset - out reset button and power on. but also menu ppp not resored! fucK!

yozz · Sat Jun 07, 2014 11:49 pm

Problém se zobrazením address listu ve WinBoxu při záznamu více jak 166 000 IP adres z několikadenního DDoS útoku, CPU 100%. Taky je problém s exportem toho seznamu na USB. (RB433UAH RouterOS 6.14rc10). Poslední funkční export a zobrazení byl při 144 000 záznamech. (czech).

WinBox problem with displaying big Address List 166 000 records - CPU 100%, export address list to USB CPU 100% (graf)

people its not DDOS maby on my royter in was DNS attack..and after then was crashed firmware and stat form 1 YEARS!!!! drop all tcp and udp traffik on inbound interface (from inet) on 53 port.!!!

yozz · Sat Jun 07, 2014 11:54 pm

What is new at 6.13

What's new in 6.13 (2014-May-15 16:03):

*) console - comments are now accepted where new command can start, that is,
where '/' or ':' characters can be used to start new command, e.g.
/interface { # comment until the end of the line
print
}
*) backup - backups by default are encrypted now (with user password).
To use backup on older versions, you should disable encryption with dont-encrypt
flag when creating it;
*) files with '.sensitive.' in the filename require 'sensitive'
permission to manipulate;
*) lcd - reduce CPU usage when displaying static screens;
*) l2tp - fixed occasional server lockup;
*) pptp - fixed memory leak;
*) sstp - fixed crashes;
Simply click “Check for updates” in QuickSet, Webfig or Winbox packages menu. If you run v5 or older, download the newest NPK package from our webpage, upload to your router, and reboot: http://www.mikrotik.com/download

NOTE:
We plan to have full release of CAPsMAN for general public in RouterOS v6.14!
Please, make all additional tests and report all persisting problems in wireless-fp package.

people why my message killed???? I WAS WROTE!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! that TCP MD 5 KEYS in ROUTING -BGP NOT WORKING!!!!!!!!!!!!!!!!!!

rextended · Sun Jun 08, 2014 1:17 am

hello!

not working ROUTING-BGP-PEERS-TCP MD5 KEY

I see the message, the post are not deleted.

Are you blind?

Stop writing on this hurting way.

rextended · Mon Jun 09, 2014 4:03 pm

I'm new to MT, so maybe it was somethere on the forum...

I got and installed ROS 6.12RC20.
Now after pressed Download and Upgrade it Download and Reboot. After reboot still 6.13RC20.
Download manualy to Files -> Reboot. Still 6.13RC20.

How to upgrade to ROS 6.13 final ?

Best
Its very simple first download your ROS 6.13 from http://mikrotik.com/download whatever device u are using. then open your router using winbox>file. and just copy your download .npk file and paste in file tab.

How ever there are variuos ways to upgrade of ROS.

Hope it will help you.

You suggest wrong solution, and this user already have received one answer with correct procedure:

NOTE 2:
In some rare cases it might be impossible to do any package managment
(upgrade/downgrade/enable/disable/uninstall/install) in 6.13rc13+ rc versions,
board will simply reboot itself staying on previous version
This is fixed for the release version of 6.13 (2014-May-15 14:33) and newer:
Please, use Netinstall for upgrade/downgrade