MikroTik

What's new in 6.32.2 (2015-Sep-17 15:20):

*) cerm - guard template from parallel use
*) mipsle - fixed missing second level menu in CLI;
*) sstp - avoid routing loops on client when adding default route;
*) sstp - fixed problem where sometimes sstp ip addresses were invalid;
*) switch - fixed bogus log messages about excessive broadcasts/multicasts on master-port;
*) tftp - fix request file name reading from packet
*) pptp encryption - better handling for out-of-order packets;
*) ethernet - added support for new ASIX USB Ethernet dongles;
*) CAPsMAN - fix 100% CPU usage when trying to upgrade RouterOS on CAP;
*) upgrade - fixed default configuration export;
*) ppp - fixed ppp interface stuck in not running state;
*) ipsec - fixed kernel failure when packets were not ordered on first call;
*) upnp - randomize action urls to fix "filet-o-firewall" vulnerability;
*) RB532/RB564 - fixed no link after ethernet disable/enable;
*) romon - fixed default configuration export;
*) tile - fixed occasional deadlock on module unload;
*) mesh - fix router lock-up when interface is added/removed;
*) ipsec - fix sockaddr buf size on id generation for ipv6 address;
*) health - show correct voltage for CRS109,CRS112,CRS210 when powered through PSU and show voltage up to 27V when powered through PoE;
*) email - resolve server address;
*) snmp - show firmware upgrade info;
*) upgrade - report status in check-for-updates.

Is that a bugfix release?

At least for now it is current release.

So there are new features added in this version since 6.32?

There are no new features, but it is not replacing the 6.30.4 as the "stable" one yet. When it will be polished enough, it might replace it, or maybe the next one

In CRS212-1G-10S-1S+ after update does't working SFP DDM info siganl

OK. I thought XX.YY is the current release and XX.YY.ZZ is bugfix.

No, number does not indicate version type. We manually choose which chain it belongs to.

Ok, going to install it in my lab as the noted SSTP problems and RB532/564 interface problems were corrections to my tickets.

Just to be sure, is the re-introduced bug in 6.33rc12 (the dynamic DNS server vanishing) solved in 6.32.2 or not?

what is this exactly -> email - resolve server address;
i was happy when i saw it because i believe that finally i could write at tools-email-server something like "mail.blablabla.com" and not only ip address. this is useful because ip address of mail outgoing server change. now i am using a script to resolve-update every hour the tools-email-server

I have still a problem on CRS212-1G-10S-1S+ - SFP modules (OEM) temperatures are not available, Wavelength is also not correct. It was working with 6.31.

The good thing is that this update has resolved full log of "excessive broadcast" (problem in 6.32.1, was fine in 6.31)

what is this exactly -> email - resolve server address;
i was happy when i saw it because i believe that finally i could write at tools-email-server something like "mail.blablabla.com" and not only ip address. this is useful because ip address of mail outgoing server change. now i am using a script to resolve-update every hour the tools-email-server

I was also inquisitive about this, turns out it will just resolve it once off to an IP when using a FQDN as the address in terminal, winbox (v2) doesnt allow the FQDN to be entered. (not sure about winbox v3).
MY tests were on on a RB951G-2HnD v6.32.2 (firmware v3.24).

So I dont think its 100% what you (we) were hoping for.
would be great if it allowed it to remain as an FQDN and only resolved to an IP when triggered.

Broke mesh, there goes the parasitic traffic that clogs the whole channel and the processor on all points, if one of the participants has been updated to 6.32.2.
What are the necessary data that would help fix?

Сломали mesh, в нём идёт паразитный трафик, который забивает весь канал и процессор на всех точках, если хотя бы один из участников обновился до 6.32.2.
Какие необходимы данные что бы помочь исправить?

TerAnYu - Please write to support@mikrotik.com. Generate supout file/s after problem has occurred and attach it/them to your problem description.

Broken tftp file download - seq wrapping detected.

what is this exactly -> email - resolve server address;
i was happy when i saw it because i believe that finally i could write at tools-email-server something like "mail.blablabla.com" and not only ip address. this is useful because ip address of mail outgoing server change. now i am using a script to resolve-update every hour the tools-email-server

I was also inquisitive about this, turns out it will just resolve it once off to an IP when using a FQDN as the address in terminal, winbox (v2) doesnt allow the FQDN to be entered. (not sure about winbox v3).
MY tests were on on a RB951G-2HnD v6.32.2 (firmware v3.24).

So I dont think its 100% what you (we) were hoping for.
would be great if it allowed it to remain as an FQDN and only resolved to an IP when triggered.

good news everyone FQDN use for e-mail was brought back to developers and it will attempt to resolve an IP address on each send attempt, not just first one to save an IP address in the settings.

We are beginning to flash a few 100 routers here for shipment.
Would you recommend 6.32.2 or 6.30.4

Rudy

We are beginning to flash a few 100 routers here for shipment.
Would you recommend 6.32.2 or 6.30.4

Rudy

I would recommend the stable bugfix. But you should make your own tests before in both cases and don't rely only on someone's recommendation. The eventual failure will be only your fault. If there is no need for update,don't do it.

I recommend 6.30.4 too

IPSEC fails on phase1 with both xauth and l2tp for me on this version. 6.27 worked fine. Any ideas why that could be?

Nevermind, it was a faulty Internet connection apparently -- it just happened to start acting up when I updated.

There are no new features, but it is not replacing the 6.30.4 as the "stable" one yet. When it will be polished enough, it might replace it, or maybe the next one

Glad to hear that....we have been recommending 6.30.4 to our clients for pretty much all of the production networks we are involved with across the world. All of our new greenfield network designs are starting with bugfix code at 6.30.4 and it seems to be incredibly stable for the features we use most often which include at a high level:

MPLS
BGP
OSPF
Queues / QoS
Firewall
LACP
VLANs

We would like to see 6.30.x stay around for 6 months at a minimum with continual bugfixes as this has been a huge improvement in stability and the response from customers has been extremely positive followed by questions about how long 6.30.x will be around before moving to the next bugfix release.

+1 to keep the bugfix version around for a long period of time before incrementing.

Are there any news regarding the fix for the OpenVPN?

Since v6.31 we were unable to connect to ovpn server on Mikrotik.
The log gives the following message:

The last working version is v6.30.4.

Thank you.

what is this exactly -> email - resolve server address;
i was happy when i saw it because i believe that finally i could write at tools-email-server something like "mail.blablabla.com" and not only ip address. this is useful because ip address of mail outgoing server change. now i am using a script to resolve-update every hour the tools-email-server

I was also inquisitive about this, turns out it will just resolve it once off to an IP when using a FQDN as the address in terminal, winbox (v2) doesnt allow the FQDN to be entered. (not sure about winbox v3).
MY tests were on on a RB951G-2HnD v6.32.2 (firmware v3.24).

So I dont think its 100% what you (we) were hoping for.
would be great if it allowed it to remain as an FQDN and only resolved to an IP when triggered.

good news everyone FQDN use for e-mail was brought back to developers and it will attempt to resolve an IP address on each send attempt, not just first one to save an IP address in the settings.

VERY NICE!!!

what is this exactly -> email - resolve server address;
i was happy when i saw it because i believe that finally i could write at tools-email-server something like "mail.blablabla.com" and not only ip address. this is useful because ip address of mail outgoing server change. now i am using a script to resolve-update every hour the tools-email-server

I was also inquisitive about this, turns out it will just resolve it once off to an IP when using a FQDN as the address in terminal, winbox (v2) doesnt allow the FQDN to be entered. (not sure about winbox v3).
MY tests were on on a RB951G-2HnD v6.32.2 (firmware v3.24).

So I dont think its 100% what you (we) were hoping for.
would be great if it allowed it to remain as an FQDN and only resolved to an IP when triggered.

good news everyone FQDN use for e-mail was brought back to developers and it will attempt to resolve an IP address on each send attempt, not just first one to save an IP address in the settings.

From 6.33rc15: *) email - allow server to be specified as fqdn which is resolved on each send;

But could it be treated the same way in SNTP client? For example if I use any of the pools (*.pool.ntp.org), the dns name is resolved only at the time of insert and this ip stays in the configuration forever.

Then you should simply do this:
":put [/resolve xxxxxxxxx]"

It will give you an IP address which you should use.

Then you should simply do this:
":put [/resolve xxxxxxxxx]"

It will give you an IP address which you should use.

For BFU it's not so simple but yes, it solves the problem

Either you need to resolve it every time or just once.

More useful option is to resolve it every time. If address changes, then you do not need to worry about it and if you want to use static address, then you need to resolve address only one manually.

I have an upgrade server that I use to push updates out on my network. On a hAP lite in winbox, I select System --> Auto Upgrade --> Refresh. It shows mipsbe packages and not the smips packages. I have the smips package on my update server.

I have an upgrade server that I use to push updates out on my network. On a hAP lite in winbox, I select System --> Auto Upgrade --> Refresh. It shows mipsbe packages and not the smips packages. I have the smips package on my update server.

hAP_lite_upgrade.jpg

Looks like AutoUpdate feature bug. Thanks, we will look into it.

i have RB951Ui-2HnD mikrotik version 6.32.2 system health voltage not showing , what can i do to fix this problem ?

The same health/voltage issue goes also for 750, including 6.30.4.

Please check your router at routerboard.com

Neither 951 or 750 have anything to show in Health menu:

Voltage Monitor No
CPU temperature monitor No
PCB temperature monitor No
Current Monitor No

http://routerboard.com/RB951Ui-2HnD
http://routerboard.com/RB750

I have a log full of "excessive broadcasts/multicasts" with 6.32.2 on RB951-2HnD. Looks like the fix for this problem is not effective.

Please check your router at routerboard.com

Neither 951 or 750 have anything to show in Health menu:
...

Thank you for the reminder...
But wouldn't it make sense to have a "N/A" in that field instead of "0.0V" to avoid any confusion.
This should be a trivial string replacement, since a single power supply board doesn't work with 0V, a reading of 0V means no sensor...

The same goes for the fan control and active fan on the NetMetal which should not be there.

whats mipsbe architecture lte package for?? documentation about it ??

Please check your router at routerboard.com

Neither 951 or 750 have anything to show in Health menu:

Voltage Monitor No
CPU temperature monitor No
PCB temperature monitor No
Current Monitor No

http://routerboard.com/RB951Ui-2HnD
http://routerboard.com/RB750

Showing fields even with zero value if the device doesn't have health sensors is definitely a bug that repetitively emerges version to version. I would rather see the health menu inactive or even missing than showing any (even zero) values.

Aligner - Very often people are confusing bug with real error messages. In this case I see that these errors are on your WAN interface. Fix was for errors on switch group. In any case that is not the same situation. I suggest that at first you check with packet Sniffer if there really are no excessive broadcast/multicast packets. If you do not have them, then generate supout file and send it to support@mikrotik.com

For rest of you - we will remove Voltage monitor 0.0V value if monitor itself is not present in upcoming versions.

chechito - RouterOS also supports 4G LTE cards:
http://wiki.mikrotik.com/wiki/Supported ... _LTE_cards

Aligner - Very often people are confusing bug with real error messages. In this case I see that these errors are on your WAN interface. Fix was for errors on switch group. In any case that is not the same situation. I suggest that at first you check with packet Sniffer if there really are no excessive broadcast/multicast packets. If you do not have them, then generate supout file and send it to support@mikrotik.com

Thank you, will check it with packet sniffer. For now problem gone, but who know will it ever return or not. My internet provider insist that there no any loops or excessive multicasts in the network. For a future, if this is a provider problem - is there any way to prevent this message from logging? Also, what is a conditions for this message?

myke1124 - We managed to reproduce your problem and we will fix it as soon as possible;
Aligner - You can remove these log messages by accessing logging settings. There you should change topics from topics=warning to topics=warning,!interface

For ppc(rb850gx2) we don't longer have resources-sector writes?

BLAKE12 - Which one/s are you understanding as new features?

I upgraded my Mikrotik RB951G-2HnD to 6.32.2 from 6.32.1 and now I cannot set my wireless to channel 11 frequency (2.462). Before I updated I was able to set my frequency to that channel. I am trying to setup multiple access points on my network, and I need the be able to have 1,6, and 11 for wireless, non-overlapping channel setup. I tried downgrading to 6.30.4, but I am still unable to set my wifi to channel 11. Is this a known bug, or is there a setting I'm missing? Please advise.

Screenshot here (the forum IMG tags are not working for me): https://www.dropbox.com/s/lk4yllgd0vemx ... 1%3A58.png

I upgraded my Mikrotik RB951G-2HnD to 6.32.2 from 6.32.1 and now I cannot set my wireless to channel 11 frequency (2.462). Before I updated I was able to set my frequency to that channel. I am trying to setup multiple access points on my network, and I need the be able to have 1,6, and 11 for wireless, non-overlapping channel setup. I tried downgrading to 6.30.4, but I am still unable to set my wifi to channel 11. Is this a known bug, or is there a setting I'm missing? Please advise.

Screenshot here (the forum IMG tags are not working for me): https://www.dropbox.com/s/lk4yllgd0vemx ... 1%3A58.png

It's because you have the extension channel set to Ce which puts it above channel 11 and outside the valid range. You need eC.

I upgraded my Mikrotik RB951G-2HnD to 6.32.2 from 6.32.1 and now I cannot set my wireless to channel 11 frequency (2.462). Before I updated I was able to set my frequency to that channel. I am trying to setup multiple access points on my network, and I need the be able to have 1,6, and 11 for wireless, non-overlapping channel setup. I tried downgrading to 6.30.4, but I am still unable to set my wifi to channel 11. Is this a known bug, or is there a setting I'm missing? Please advise.

Screenshot here (the forum IMG tags are not working for me): https://www.dropbox.com/s/lk4yllgd0vemx ... 1%3A58.png

It's because you have the extension channel set to Ce which puts it above channel 11 and outside the valid range. You need eC.

Thanks for the prompt response

There seems to be an issue with SNTP getting synced in this version.
I use a script to periodically set my server IPs to us.pool.ntp.org and time.nist.gov
script works fine but router simply wont sync any more after upgrading from 6.30 to 6.32.2.

There seems to be an issue with SNTP getting synced in this version.
I use a script to periodically set my server IPs to us.pool.ntp.org and time.nist.gov
script works fine but router simply wont sync any more after upgrading from 6.30 to 6.32.2.

Well, turns out I forgot to do the Routerboad fw updated after the package update. now its working like it should.

For ppc(rb850gx2) we don't longer have resources-sector writes?

??

Problem with router cache owerflow caused by sending traffic over IPsec tunnels (even if the router cache is turned off in the settings) still persists in this release

Ticket number #2015081766000633

My RB1100 rebooted itself every roughly middle of the night with error Ticker #2015100966000564

Hello, I excuse my English, I am using the google translator.
I want help, I have a RB1100Hx2 v6.32.2 and today and I can not enter Winbox is very strange because I've always been able to enter without problems, can not enter the web.
When I want to enter the IP Winbox shows me a message "ERROR: could not connect to" ip "
When I want to enter the MAC winbox shows me the message "Incorrect user or password" is strange because no one has changed the user or password.
I also rebooted the router to test whether solved but I can not connect, there is a way to reset the settings or to enter?

My winbox is v3.0

Hola, disculpen mi inglés, estoy usando el traductor de google.
Quiero pedir ayuda, tengo un RB1100Hx2 v6.32.2 y el día de hoy ya no puedo entrar por Winbox, es muy extraño ya que siempre he podido ingresar sin problemas, tampoco puedo entrar por la web.
Cuando quiero ingresar por Winbox con la IP me muestra un mensaje “ERROR: could not connect to "ip"
Cuando quiero ingresar por winbox con la MAC me muestra un mensaje de “user or password incorrectos”, es raro porque nadie ha cambiado el usuario ni password.
También he reiniciado el router para probar si se soluciona pero no logro conectarme, hay algún modo de restablecer la configuración o poder ingresar?

When I want to enter the IP Winbox shows me a message "ERROR: could not connect to" ip "

can you ping that IP? is http://IP working?

When I want to enter the MAC winbox shows me the message "Incorrect user or password" is strange because no one has changed the user or password.

try empty password (and 'admin' user with empty password)

*) vrrp - fix arp=reply-only;
*) vrrp - do not warn about version mismatch if VRID does not match;
*) vrrp - allow VRRP to work behind firewall and NAT rules;

How this feature actually works?

Sí, el ping a la ip funciona, mi router tiene una IP privada y una pública a ambas el ping funciona correcto, pero con ninguna de las 2 puedo conectarme.
He probado todas las combinaciones posibles de usuario y password con la MAC pero siempre me indica que está equivocado.
Lo raro es que hasta la semana pasada he podido conectarme sin problemas, tengo backups diarios de mi configuración pero no sé cómo restaurarlos si no puedo conectarme al mikrotik.


Yes, the ping ip works, my router has a private and a public IP ping works both correct, but with none of the two can connect.
I tried every possible combination of username and password to MAC but I always indicates that you are wrong.
The strange thing is that until last week was able to connect smoothly, I have daily backups of my setup but do not know how to restore them if I can not connect to mikrotik.

Problem with router cache owerflow caused by sending traffic over IPsec tunnels (even if the router cache is turned off in the settings) still persists in this release

Ticket number #2015081766000633

Just for the record - I finally managed to resolve this issue on router with ~20 IPsec tunnels (same router, same configuration with disabled tunnel never encountered this problem). tl;dr version: Guess what... by turning the cache back on.

This settings had no effect in version 6.30.2. - when I opened the ticket back then, I got the advice from MT support : "turn the IP cache feature off", but it had no effect and the setting stayed there.

But turning on the route cache ( /ip settings set route-cache=yes) in 6.32.1 (I did not test the next versions yet) actually force the cache to work as it should. However if you change it on running system, this change affect the cache records from that point only - cache entries created prior to the point you turn the route cache feature on, stays there forever, until the router is restarted.

It almost looks like IPsec tunnels use router cache regardless the cache on/off settings, but if the case is turned off in IP settings, no one cares about the records in cache any more, so it will overflow in the end, causing all IPv4 traffic to stop. Turning on the cache feature forces all records to be managed by regular cache algorithm, so it works as it should.

At least... two months of random outages and half year of daily reboots is finally over.

DYNAMIC DNS PROBLEM IN BUGFIX 6.32.3 VERSION

Is any1 else having problems while using the BUGFIX VERSION: 6.32.3 that sometimes/often the Dynamic DNS does not get recieved or registered?!

Meaning that the customer does not get online, cause the DNS does not work for them!

The problem seems to be fixed in ROS v6.33.5 but as 6.32.3 is supposed to be a BUGFIX version, i would STRONGLY suggest that this fix is also applied and put into a "Bugfix 6.32.4 " version ASAP!


PS. For easier use of forums:

Would it not also be better to have a STICKY/PERMANENT forumthread for whatever is the present version of: CURRENT, BUGFIX and RC version. And that had the name in forum like f.ex: "ROS "BUGFIX" VERSION THREAD(Currently bugfix v6.32.3)" and "ROS "CURRENT" VERSION THREAD(Currently: Current v6.33.5)" or something similar??!!.

Tried to search for a thread that concerned only Bugfix v6.32.3, I couldnt find it, probably just me who screwed it up, but if one can probably others too!

Cheers

I had this problem with older versions but it is working correctly for all my devices running 6.32.3 so far.