MikroTik

Posted: **Wed Apr 18, 2018 11:41 am**

*) fetch - added "output" option for all modes in order to return result to file, variable or ignore it;

I have failed to write in variable. Any way.

Posted: **Wed Apr 18, 2018 11:48 am**

Guys what this means:
*) capsman - added support for "interface-list" in Access List and Datapath entries;

In CLI, you can see interface-lists in "interface=" values when you press Tab.

Posted: **Wed Apr 18, 2018 11:51 am**

Chupaka Ahh waitng when it be available on Winbox.

Posted: **Wed Apr 18, 2018 11:55 am**

My rb850gx2 (early - non ipsec version) went into reboot loop for an hour ... then managed to stabilise itself and seems to be running stable now for more than an hour. Log reports: kernel failure in previous boot. Upgraded from 6.41.3.

Posted: **Wed Apr 18, 2018 12:20 pm**

Chupaka Ahh waitng when it be available on Winbox.

Yeah, funny, WinBox shows "Unknown interface" when you add it via CLI with interface-list

Posted: **Wed Apr 18, 2018 12:21 pm**

Hello,

I get a some issues with bonding 802.3ad interfaces. First between BR3011 and CRS326. There is no traffic if bonding is configured with two activated interfaces. If I release one interface, traffic will pass the bonding interface and all is fine. When I reconnect the second interface, traffic is interrupted again. Within the bonding interface, a bunch of VLANs are configured.

On the CRS326 is a second bonding interface is configured without some VLANs. On this interface bonding works and balanced the traffic as expected, but the speed is not faster then 100 MBit/s per link. With RouterOS 6.42rc39 I get full GBit/s speed on this link.

As already mentioned the CRS326 was upgraded from 6.42rc39, the RB3011 from 6.41.4.

Many thanks

Posted: **Wed Apr 18, 2018 2:52 pm**

Running fine on a RB941-2nD (hAP-lite) as SOHO (L2TP + IPSec).

Posted: **Wed Apr 18, 2018 2:55 pm**

Hi,

After upgrading from 6.41.3, RB850Gx2 started rebooting in a loop, reseting did help but only after 4th time, and still wasn't able to restore from backup, restoring from backup it started rebooting again.

with 6.41.4 IPTV (multicast) traffic started glitching, I had to switch back to 6.41.3...

Same issue, can't generate supout. Even after Netinstall, RouterBOOT @ 3.24, and 6.41.3 before web upgrade.

Posted: **Wed Apr 18, 2018 3:59 pm**

Hello,

I get a some issues with bonding 802.3ad interfaces. First between BR3011 and CRS326. There is no traffic if bonding is configured with two activated interfaces. If I release one interface, traffic will pass the bonding interface and all is fine. When I reconnect the second interface, traffic is interrupted again. Within the bonding interface, a bunch of VLANs are configured.

On the CRS326 is a second bonding interface is configured without some VLANs. On this interface bonding works and balanced the traffic as expected, but the speed is not faster then 100 MBit/s per link. With RouterOS 6.42rc39 I get full GBit/s speed on this link.

As already mentioned the CRS326 was upgraded from 6.42rc39, the RB3011 from 6.41.4.

Many thanks

The performance issue is no longer a problem and has been solved. The RB3011 <- bond -> CRS326 is still open.

Posted: **Wed Apr 18, 2018 6:04 pm**

LTE Interface does not work in 6.42

Posted: **Wed Apr 18, 2018 6:41 pm**

When upgrade packages to v6.42 in CHR router (Hyper-V image), after reboot have:
Loading system with initrd
ERROR: could not find disk!
Please attach it somewhere else.

System hangs up.

When loading blank chr-6.42.vhdx, the same error.
Before v6.42 all works fine.
Something wrong with .vhdx image?

Posted: **Wed Apr 18, 2018 7:29 pm**

I have hap lite /smips/ with 6.41.2 . With 6.42 in station mode and WEP security /40bit/not work - can not connect. I downgrade to bugfix 6.40.7 and everything is fine - the routrboard is connected !

Posted: **Wed Apr 18, 2018 8:42 pm**

I am trying on the moment 6.42 with the new bridging setup. Had some problems to access websites and an other setup (NAT) than I used on 6.40.6 did solve that.

My L2TP/IPSEC is about 20 to 30 percent slower on 6.42 than on 6.40.6 and I will try 6.42 some more and try solve the slowness I experience in the new bridge setup. I have RTSP active and except for the slower VPN that drops from 90 Mbps to 70 Mbps.

Update: did some optimizing on the mangle and raw and I now get the same or more speed through L2TP/IPSEC than on 6.40.6.

Posted: **Wed Apr 18, 2018 9:50 pm**

Hi All,

For the last couple of times every time I try to upgrade a map lite it crashed and I need to reinstall it through netinstall.

By design those are powered only through poe switch, but that doesn't seem to work for upgrading.
This time I expressly connected them to a normal network connection and usb power.

Still no luck boot loop after upgrade.

I've already contacted support but until now didn't really get an answer about this issue ?

Anybody else having the same issues with map lite ?

Thanks

Posted: **Wed Apr 18, 2018 9:52 pm**

Nightmare upgrade.

Considering previous upgrades this one didn't go smooth at all.

I was able to upgrade my ccr's / csrs without a problem.

However I have a small hex poe light witch powers a test setup with a cap lite / 2 rb2011 / netmetal.
In the past upgrades went without a problem. this time each device stayed in a boot loop.

I've even tried a rb 2011 directy connected to power and still same issue

Might want to be carefull in production for this one..

Posted: **Wed Apr 18, 2018 10:27 pm**

When upgrade packages to v6.42 in CHR router (Hyper-V image), after reboot have:
Loading system with initrd
ERROR: could not find disk!
Please attach it somewhere else.

System hangs up.

When loading blank chr-6.42.vhdx, the same error.
Before v6.42 all works fine.
Something wrong with .vhdx image?

I have updated two CHR running under Hyper-V (2012R2) without any issues at all. Much appreciated additions for CHR in this release so it should work. Which version of Hyper-v are you using?

Posted: **Wed Apr 18, 2018 10:34 pm**

After an upgrade of multiple devices to 6.42:
1. /ip neighbor discovery - is not working anymore on dynamic (L2TP) interfaces

2. OSPF is broken. Massive messages:
2018-04-18T22:23:14.504801+03:00 kv-per37k7-rtr01.sysadm.in.ua route,ospf,info Ignoring Link State Acknowledgment packet: wrong peer state
2018-04-18T22:23:14.504801+03:00 kv-per37k7-rtr01.sysadm.in.ua route,ospf,info state=Init

Other side routers shows that incorrect source IP is being used for OSPF packets.
OSPF is running in VRF but source is from main table!!!!

3. Log window (in winbox 3.13) shows only one last line. Which is the biggest pain in the ass as it does not give a chance to debug issues

I think more will come.
UPGRADE TO PRODUCTION DEVICES IS NOT RECOMMENDED!!

Posted: **Wed Apr 18, 2018 11:23 pm**

2. OSPF is broken. Massive messages:
2018-04-18T22:23:14.504801+03:00 kv-per37k7-rtr01.sysadm.in.ua route,ospf,info Ignoring Link State Acknowledgment packet: wrong peer state
2018-04-18T22:23:14.504801+03:00 kv-per37k7-rtr01.sysadm.in.ua route,ospf,info state=Init

Resolved. Caused by wrong NAT rule

Posted: **Thu Apr 19, 2018 1:04 am**

Hard to tell is there any improvement. When running TCP bandwidt test on every client, 40-45 Mbit getting average 700Kbps per client. With less than 10 clients can get approx 50-55Mbps troughput.
There is bug in ping or in packet transmision. on every ping request there is 2 responses. packet loss -70%(minus)

Posted: **Thu Apr 19, 2018 2:08 am**

Seems to work ok for me in various devices

-tp

Posted: **Thu Apr 19, 2018 3:29 am**

What is the wireless XX option under Channel width?

Posted: **Thu Apr 19, 2018 3:45 am**

I get a some issues with bonding 802.3ad interfaces. First between BR3011 and CRS326. There is no traffic if bonding is configured with two activated interfaces. If I release one interface, traffic will pass the bonding interface and all is fine. When I reconnect the second interface, traffic is interrupted again. Within the bonding interface, a bunch of VLANs are configured.

I'm running into a similiar problem on RB3011. I have bonding configured on SFP and one Ethernet to CRS317 and a lot of VLANs on that bonding interface; RB3011 seems to be unable to handle traffice after booting for several hours. Bridging seems to be broken, downlink devices cannot get DHCP leases or cannot ping router. (but some devices still get some sort of network connectivity regardless of whether they can ping gateway.) Reboot usually solves the problem (not guaranteed).

Posted: **Thu Apr 19, 2018 4:36 am**

LTE Interface does not work in 6.42

Upgrade SXT LTE without any problems.

Posted: **Thu Apr 19, 2018 4:42 am**

Is it just me, or can anyone else not Torch any Ethernet interfaces in a bridge on a CRS device?

Specifically CRS210-8G-2S+?

I'm using 6.42 but I think this also applies to some earlier versions as well.

Winbox 3.13

When I Torch any Ethernet interface it only shows a blank screen.

I can torch the bridge, but it just shows broadcast traffic, nothing specific to the interfaces belonging to the bridge.

Posted: **Thu Apr 19, 2018 10:05 am**

I have updated two CHR running under Hyper-V (2012R2) without any issues at all. Much appreciated additions for CHR in this release so it should work. Which version of Hyper-v are you using?

Same issue here. Two CHRs: first on 2008R2 does not boot, second on 2012R2 is ok.
Please check.

Posted: **Thu Apr 19, 2018 10:18 am**

LTE Interface does not work in 6.42

I have upgraded 3 wAP and lte interface works, try to reset RB or downgrade to another stable and then upgrade again.

Posted: **Thu Apr 19, 2018 10:41 am**

Jamesits, csi -

Please write to support@mikrotik.com
We will need to see your network topology and your supout.rif from both devices.
There is something specific in your configuration or your setup that might be causing these issues.

Posted: **Thu Apr 19, 2018 10:58 am**

What is the wireless XX option under Channel width?

I am wondering too

Posted: **Thu Apr 19, 2018 1:30 pm**

I can't see any improvement on Nv2 PTMP, someone have some tests and can share AP and client configuration?
did you upgrade AP and stations, or only AP ?

I've upgraded AP and two of three stations connected, only RouterOS

Posted: **Thu Apr 19, 2018 1:33 pm**

*) fetch - added "output" option for all modes in order to return result to file, variable or ignore it;
I have failed to write in variable. Any way.

Same for me. Anybody should update the documentation in the wiki (and possibly add an example).

Posted: **Thu Apr 19, 2018 2:08 pm**

*) fetch - added "output" option for all modes in order to return result to file, variable or ignore it;
I have failed to write in variable. Any way.

How did you tried, I am also looking for this feature, thank you

Posted: **Thu Apr 19, 2018 3:59 pm**

I can't see any improvement on Nv2 PTMP, someone have some tests and can share AP and client configuration?
did you upgrade AP and stations, or only AP ?
I've upgraded AP and two of three stations connected, only RouterOS

It's been said somewhere in another tread, but the improvements are only noticeable in a P2MP network with more then lets say 15-20 connected CPE's.
We tried is on some with 28+ connected clients and indeed increased the total throughput by some 60-100% and each client also saw more speed, and the most importante feature, if one bad signal client consumes the 'airtime' of the AP a lot because of his low modulation rate and thus throughput and thus time needed for transport of data, under the new version ROS the bad client doesn't have no more effect.
I tried that last where we have had a -75dBm client that had only a couple of megabits of speed in a 28 client network where other had much more speed, some 30Mbps in a single client test, after the uprade not only could I run several clients at the same time with 30-60mbsp, a single client up to 80-90Mbps, and when I did start the bandwidth test of the 'bad' client he not only also went up from 1 to some 5Mbps but more important, it didn't have an effect on the other clients running their download bandwidth test.
It did have an disastrous effect while AP was still running the old software.

Only the AP needs the upgrade.

Posted: **Thu Apr 19, 2018 4:00 pm**

What is the wireless XX option under Channel width?

Indeed, that what I would like to know too.....

Posted: **Thu Apr 19, 2018 4:13 pm**

Ce - eC automatic selection? Just a guess.

Posted: **Thu Apr 19, 2018 5:27 pm**

@MikroTik:
Could you please add robustness to your DHCP and PPPoE clients. They are unreliable and I have had to add scripts to restore connectivity after a drop.

Right now, I've got a router with DHCP stuck in status "renewing..." and I have no default gateway. This should not happen. Monitor the DHCP request processes, with timeouts and follow-up actions.

I also have to deal with the same kind of crap with L2TP servers, whereby sometimes the service gets in a bad state and some clients are able to establish tunnels and some keep failing -- until I reboot the router that is the server.

Same goes for OSPF and some other routing/forwarding protocols sometimes misbehaving until a reboot... and on and on.

You are missing a lot of service health/status checks, problem detection and recovery routines in your code. Users shouldn't have to write so many scripts to recover operation of core functionality. This has pretty much been the routine over the past 6 years, for every routerboard line I have dealt with, and it gets frustrating.

We need greater stability.

And for any user who's going to respond about it all working great for them, keep it to yourself. You're not being helpful.

Thanks

Posted: **Thu Apr 19, 2018 5:52 pm**

Do you actually reported to support?
If it is on the same router, then for me it looks like problem with interface driver not separate processes.

Posted: **Thu Apr 19, 2018 5:57 pm**

DHCP client and server work as expected. PPPoE client also ... No problems on my RB2011 here

Posted: **Thu Apr 19, 2018 7:02 pm**

Problem with 6.42 on Whatsapp, if there more than 2 whatsapp connected the new ones will not connect will be offline, same problem with web.whatsapp can you guys check?

Posted: **Thu Apr 19, 2018 7:38 pm**

Rb2011 lost wireless option after the upgrade. I see the wireless 6.42 package is still in the files. I added the wireless package from the extras package and it won't load either. Many reboots and still no install.

Posted: **Thu Apr 19, 2018 9:08 pm**

upgraded one RB951G-2HnD in office (running as an AP only) and it went fine
just upgraded another RB951G-2HnD at home (also an AP only) and it immediately went into a bootloop... beeping every 7 seconds or so. I was only upgrading the software and using winbox.

power on reset does not help
can not get it to netinstall either. never had tu use it before but have read many sets of online instructions and pretty sure i am doing everything right.. the RB never even shows up in the netinstall window

anyone else had this problem and solve it ?(my kids are cross with me now .. no wifi

EDIT: managed to get it back in the end... netinstall to the rescue (change of cables and a switch between the MT and the PC got me there in the end)

Posted: **Thu Apr 19, 2018 10:18 pm**

Hello Folks!

I tried two devices, rb411, one worked and the other disappeared from network will visit it physically tomorrow trying to figure out what happened.
The first device took unusual long time come back on network.

Posted: **Thu Apr 19, 2018 11:12 pm**

no problems with update from 6.40.5 to 6.42 on hap lite with almost stock configuration

Posted: **Thu Apr 19, 2018 11:34 pm**

What is the wireless XX option under Channel width?

Maybe set your wifi channel to combined ce and ec!

Posted: **Fri Apr 20, 2018 12:35 am**

LTE Interface does not work in 6.42

I have upgraded 3 wAP and lte interface works, try to reset RB or downgrade to another stable and then upgrade again.

i have SXT LTE

Posted: **Fri Apr 20, 2018 1:17 am**

What is the wireless XX option under Channel width?
I am wondering too

Was thinking that but would like to be sure

Posted: **Fri Apr 20, 2018 1:19 am**

I just realized earlier that after I updated to v6.42 that all my ip>firewall>raw rules were missing, not sure if that was the only config missing, I just restored a backup and everything was back.

Posted: **Fri Apr 20, 2018 3:04 am**

Rb2011 lost wireless option after the upgrade. I see the wireless 6.42 package is still in the files. I added the wireless package from the extras package and it won't load either. Many reboots and still no install.

Just upgraded my rb2011 this afternoon and wireless just works..... try to downgrade and see if it comes back. If not, radio is dead. If yes, upgrade again and see what happens...

Posted: **Fri Apr 20, 2018 3:07 am**

What is the wireless XX option under Channel width?
Maybe set your wifi channel to combined ce and ec!

?? How is that? It can only be Ce or eC. The eCe would make a 60Mhz wide channel which is not supported by 802.11ac

Posted: **Fri Apr 20, 2018 3:10 am**

Ce - eC automatic selection? Just a guess.

For AP? Because for station any 20/40/80 setting wil work with AP. No matter what AP is using. As long as the frequency of the scan range can 'see' the AP it will adapt.

Posted: **Fri Apr 20, 2018 5:33 am**

I'm pretty happy with this version so far.

Speedtest history shows the difference - upgrade at 4:10am. No configuration changes.
Sadly this will not protect my beloved RB from replacement - I have available 500Mbps downlink since 2 weeks. ;(

Posted: **Fri Apr 20, 2018 9:01 am**

I'm having an issue upgrading a CHR instance from 6.42rc20 to 6.42 current/stable using

/system upgrade install

. I have it licensed at a P1 level and it shows it's current when I run

/system license print

prior to the upgrade but when it reboots after the upgrade, it says starting as it normally does then it says my demo license has expired on the console and powers off immediately after that. It's running on an ESXI 6.5 host.

Do I need to do a fresh install instead of attempting to upgrade?

Posted: **Fri Apr 20, 2018 9:06 am**

first time in my life i had to downgrade back to 6.41.4 coz on this 6.42 my wifi signal was dropping every 15mins or so and in log was constantly seeing unicast key exchange time. Im using metal 52 with level 4 license.
Back on 6.41.4 and everything back to stable again

Posted: **Fri Apr 20, 2018 9:07 am**

Jamesits, csi -

Please write to support@mikrotik.com
We will need to see your network topology and your supout.rif from both devices.
There is something specific in your configuration or your setup that might be causing these issues.

I have send out both supout files. Did you need more information? Btw. it would be helpful if you support area on your website would allow to upload more then one supout.rif file or allow zip files. I added the the secound supout as picture

Posted: **Fri Apr 20, 2018 9:13 am**

any changes in the group-ciphers settings in wifi?

I've had to change from

group-ciphers=aes-ccm
to
group-ciphers=tkip,aes-ccm

in order for my MT clients to connect.

... disconnected, group key exchange timeout

Posted: **Fri Apr 20, 2018 9:45 am**

PCC (Per Connection Classifier) load balancer totally broke after upgrade 6.40.5 -> 6.42.
Downgrade to 6.41.4 made it work again. Tested on two RB2011

Posted: **Fri Apr 20, 2018 10:07 am**

What is the wireless XX option under Channel width?
I am wondering too
Was thinking that but would like to be sure

Still no any answer

Posted: **Fri Apr 20, 2018 2:44 pm**

Rb2011 lost wireless option after the upgrade. I see the wireless 6.42 package is still in the files. I added the wireless package from the extras package and it won't load either. Many reboots and still no install.

I had same problem year or two ago. Reseting configuration to default/empty solved me the problem.

Posted: **Fri Apr 20, 2018 3:49 pm**

netwatch - limit to read, write, test and reboot policies for Netwatch script execution;

Maybe i don't understand something, but why?
Now we can not use /tool fetch (for example, telegram bots), can not set global variables!. Please add at least "ftp" and "policy"!
It's very annoying.

Posted: **Fri Apr 20, 2018 3:54 pm**

netwatch - limit to read, write, test and reboot policies for Netwatch script execution;
Maybe i don't understand something, but why?
Now we can not use /tool fetch (for example, telegram bots), can not set global variables!. Please add at least "ftp" and "policy"!
It's very annoying.

If you had a "write" only account, you could make a netwatch rule, that executes a script, that creates a "full" user, or worse. Permission escalation bug.

Posted: **Fri Apr 20, 2018 4:16 pm**

I upgraded 2 CCR and it went okay, when i upgraded 1 of my ospf routers, it went into loop, and would not boot up, tried another one and same, It looks something is wrong with CCR upgrade, other routerboards are working fine after upgrade but i just did couple of them.

Posted: **Fri Apr 20, 2018 4:39 pm**

If you had a "write" only account, you could make a netwatch rule, that executes a script, that creates a "full" user, or worse. Permission escalation bug.

Okay... understandable. Wrote a script using /file set contents. It works for now.
thanks for answer!

Posted: **Fri Apr 20, 2018 4:50 pm**

installed now 6.42,works fine on rb3011.

Posted: **Fri Apr 20, 2018 6:52 pm**

Hello

How to use this new feature
*) fetch - added "output" option for all modes in order to return result to file, variable or ignore it;
?

Posted: **Fri Apr 20, 2018 6:54 pm**

I can't connect to my cAP ac on 5GHz from my Nokia 6 and SGS3. Phones see 5GHz network but nothing happens when I tap connect. Router don't show antyhing on log. Anyone have the same problem?

Posted: **Fri Apr 20, 2018 8:18 pm**

I've upgraded a handfull of CCR units running OSPF and VPLS endpoints on MPLS and everything seems fine.

Also done a few of our CRS units and it seems to work fine.

Though on one CCR I noticed a few of the Ether interfaces on a bridge that were not connected were 'flapping' at Ethernet connected, then not connected, then connected, kind of randomly.
That was very odd, and the logs showed it, but nothing physical was actually connected to those ports, lol!
I don't see that anymore, so maybe it was a fluke.

Also, I'm still having an issue where I can't torch a lot of the interfaces.
It just shows a blank Torch screen, like it won't start or something.
But that goes back a few versions too, same thing on a lot of my CCRs on previous versions 6.41.x as well.

Posted: **Fri Apr 20, 2018 9:25 pm**

Upgraded a CCR1009, 2xCRS226 and 4xwAP AC without any issues. Was running a previous build of v6.42RCxx --> v6.42 final.

Posted: **Sat Apr 21, 2018 12:59 am**

Hello Folks!

I tried two devices, rb411, one worked and the other disappeared from network will visit it physically tomorrow trying to figure out what happened.
The first device took unusual long time come back on network.

Physical visit, unplugging and plugging back power worked, the device came back and we could continue with /system routerboard upgrade followed by the required reboot and the device come back on network afterwards.

However, the high number of rb411, ospf everywhere, tons of vpn tunnels of all kinds criss crossing our network islands, tons of nv2 p2mp setups with critical customers and so on, made us think once more before we roll it out everywhere. I do as learnt at IBM, waiting to ".2 version" of current came out.

Posted: **Sat Apr 21, 2018 3:22 am**

!) w60g - increased distance for wAP 60G to 200+ meters;

I can confirm and I am super-impressed. Just set up a 190m shot on a festival site with not perfectly clear LoS and got signal 90 @ MCS8.
Constantly pushing 850MBit.

Wow.
-Chris

Posted: **Sat Apr 21, 2018 9:11 am**

mszru, zilf, sterling, sterling - Traffic within switch (between HW bridge ports) does not reach CPU and will not show up on Torch.
hknet, ErfanDL, irghost, JohnTRIVOLTA, artemk, dravnieks, Jamesits, filzek, Guentha, xbipin, XaTTa6bl4 - Please send supout files to support@mikrotik.com.
bbs2web, JanezFord, Chupaka, wenasong - You should try you 6.43rc where additional fixes are included.
Spirch, jarda, yhfung - We are constantly working on improvements regarding this matter. You will see changes in changelog when we will manage to fix these problems.
alibalalo - Few posts ago: "Currently there is no RouterBOARD with such built-in wireless functionality".
mattjhaigh, Abdock - Netinstall should resolve the problem. However, serial console output might be useful for support in order to see the cause of the problem.
Simono - You can now choose "interface-list" instead of "interface" under these settings. Use multiple interfaces at the same time.
Chupakabra303, eworm, diablonet, miencek - Unfortunately we still have a work to do regarding this feature. It is not 100% complete yet.
Wave, glat, rajo, raymondr15, timoid, Simono - Please report this to support@mikrotik.com.
stefanca - Is mAP lite powered by RouterBOARD device which is also upgraded at the same time?
raymondr15, amt, WirelessRudy, ErfanDL, raymondr15 - This feature allows to select channel automatically by selecting the best one from available.
koolkid1935 - Problem is with rc not this release. You will have to create a new CHR instance.

Posted: **Sat Apr 21, 2018 6:04 pm**

There is no NTP package in the all_packages-smips-6.42.zip (hAP mini, hAP lite)

Posted: **Sat Apr 21, 2018 6:18 pm**

Redmor - In order to see improvements you must upgrade AP devices. Please note that in most cases you can see the difference only, if there are more than 10 stations connected (number can vary, it can be more and it can be less than 10).
What do you mean with "upgrade", isn't 6.42 enough? I have to upgrade Routerboard's firmware?
I also have to upgrade clients to 6.42 and also RB firmware?

You just have to upgrade AP and all connected clients ROS.

Further reading :
viewtopic.php?f=21&t=132181&sid=30d50ab ... 36ce6531f4

Posted: **Sat Apr 21, 2018 6:58 pm**

6.42 on a CRS125-24G-1S with Fastpath enabled blackholes traffic. Toggling Fastpath off restores connectivity.

Posted: **Sat Apr 21, 2018 9:08 pm**

Netinstall will put the router backup but I am afraid i cannot upgrade these over the air, I tried to downgrade rb1100x2 and all went into same reboot cycle.

Posted: **Sat Apr 21, 2018 11:10 pm**

There is no NTP package in the all_packages-smips-6.42.zip (hAP mini, hAP lite)

That is correct observed.

Posted: **Sun Apr 22, 2018 4:01 am**

SXT bricked after 5.26 upgrade to 6.42
Please withdraw this release or add some warning on top.

Posted: **Sun Apr 22, 2018 7:25 am**

Does 5.26 to 6.41 upgrade go without problems? Jumping 43 versions ahead with MAJOR change of bridge and switch implementations? Brave move.

Posted: **Sun Apr 22, 2018 8:58 am**

koolkid1935 - Problem is with rc not this release. You will have to create a new CHR instance.

That seems to have worked. Created a new CHR instance using the 6.42 VMDK, restored config, transferred the P1 license to the new instance, all is well again.

Posted: **Sun Apr 22, 2018 10:40 am**

I did test upgrade from 5.26 to this version. Just for testing. Everything went well. Upgrade failed for you due to configuration or something like thta. In general it works just fine.

Posted: **Sun Apr 22, 2018 11:16 am**

After update some AP's i have cpu load 100% .... (Omnitik)
Disable wlan not solve problem....
Reset config ang manual set all settings not solve problem...
Latest 42rc have some bug... but disable SNMP solve problem ...!!??
Ping to affected router 100-1000ms, and router randomly rebooted, reboot reason in log is : router rebooted becouse critical program crashed
Downgrade to 6.40.7 solve problem

Posted: **Sun Apr 22, 2018 12:58 pm**

Does 5.26 to 6.41 upgrade go without problems? Jumping 43 versions ahead with MAJOR change of bridge and switch implementations? Brave move.

5.26 to 6.40 - no problems, Mr Smartie

Posted: **Sun Apr 22, 2018 1:20 pm**

My experience of new NV2 implementation:

1) Overall throughput increased, that's true. Previously, as I look on graphs on 2 test APs it didn't reached 40MBits. Now can reach 85mbits easily.
2) However there is some problem with TDMA timing.
For example, let run btest between AP with 6.42 and some 2 stations:

2) Do test with 1 tcp count to this 2 stations. Speeds will be almost equally splitted (as pre-6.42)
3) Do test with 20tcp count to 1st station, and 1tcp count to 2nd station. 1st station will get full speed, and 2nd station just 0.5-1.5mbits

So there is something wrong with new NV2 and tcp connections count.

Posted: **Sun Apr 22, 2018 1:38 pm**

Updated RB2011 and also installed Winbox 3.13
Now when I open bridge menu, ports tab the winbox hangs.
Don't know if it was the 6.42 or the Winbox update that causes it.
Bridge is configured with 4 VLAN's (not changed from 6.41)

Posted: **Sun Apr 22, 2018 3:59 pm**

perhaps not as important as some other bugs being posted, but i noticed that kid control in webfig seems to be broken
after adding a schedule in winbox, the times in webfig are showing as incorrect. it looks as though it is showing the end times as the same as the start times (?)
this is on an RB750Gr2
(am I correct in assuming that Mikrotik staff monitor this forum and these things don't need to be reported somewhere else?)

Posted: **Sun Apr 22, 2018 4:13 pm**

(am I correct in assuming that Mikrotik staff monitor this forum and these things don't need to be reported somewhere else?)

They do but don't rely on that, this is not an official feedback channel. An e-mail report to support@mikrotik.com will be properly tracked by a trouble ticket tracking system.

Posted: **Sun Apr 22, 2018 4:26 pm**

I'm pretty happy with this version so far.
Speedtest history shows the difference - upgrade at 4:10am. No configuration changes.
Sadly this will not protect my beloved RB from replacement - I have available 500Mbps downlink since 2 weeks. ;(

I am really surprised you don't get more out of this device. Could you identify what is the limiting factor ?

Posted: **Sun Apr 22, 2018 8:41 pm**

My rb850gx2 (early - non ipsec version) went into reboot loop for an hour ... then managed to stabilise itself and seems to be running stable now for more than an hour. Log reports: kernel failure in previous boot. Upgraded from 6.41.3.

Same here. I have the RB850Gx2 with hardware IPsec capabilities. Software update to 6.42 was fine. Firmware update to 6.42 resulted in a loop that eventually resolved itself (no more than 5 minutes later). After it came back up, Winbox reported the firmware was 6.42. I tried the upgrade again just to see what would happen, and it failed in the same manner, a couple reboot cycles. Kernel failure on boot is reported each time. Winbox UI still says firmware is at 6.42. Perhaps this is a bug with one of the post-upgrade checks that fails while the rest of it succeeds?

Posted: **Sun Apr 22, 2018 9:15 pm**

Updated RB2011 and also installed Winbox 3.13
Now when I open bridge menu, ports tab the winbox hangs.
Don't know if it was the 6.42 or the Winbox update that causes it.
Bridge is configured with 4 VLAN's (not changed from 6.41)

Just tossing in a +1 here for this happening to me, too. Winbox crashes in my case. Both Winbox 3.12 and 3.13. It only happens on my RB850Gx2. My RB750gl and NetMetal5 are fine.

These devices aren't critical to any business, so I haven't reached out to support. If the crash is unusual among folks here and anyone on the support team wants more details, feel free to reach out to me.

Posted: **Mon Apr 23, 2018 3:07 am**

Hi all,

Slow performance even uploading a package file.

Regards,

Posted: **Mon Apr 23, 2018 7:35 am**

Copper, Jotne - There never have been one. SMIPS platform devices does hot have a NTP package since devices are designed for simple home usage and there is not enough space on router to store such packages.
bongani, Neovr, karwos, en1gm4, abcwarbot - Please send supout files to support@mikrotik.com and report this problem.
Abdock - Please provide serial console output to support@mikrotik.com from the router which is rebooting itself and can not boot up properly.
karwos - Please write to support@mikrotik.com and provide your configuration which was used on 5.26 version. We will try to reproduce the same problem locally.
pe1chl, bcall - We already recieved such problem report which is most likely caused by Winbox. We will fix it as soon as possible.
bcall - Problem should be resolved in 6.43rc version.

Posted: **Mon Apr 23, 2018 10:49 am**

There is bug finding a disabled ppp-client interface.

[admin@PinceGW-teszt] > /interface ppp-client 
[admin@PinceGW-teszt] /interface ppp-client> print
Flags: X - disabled, R - running 
 0 X  name="modem" max-mtu=1500 max-mru=1500 mrru=disabled port=usb1 data-channel=0 info-channel=1 
      apn="lant" pin="0000" user="" password="" profile=default phone="" dial-command="ATDT" 
      modem-init="" null-modem=no dial-on-demand=no add-default-route=yes default-route-distance=1 
      use-peer-dns=yes keepalive-timeout=30 allow=pap,chap,mschap1,mschap2 
[admin@PinceGW-teszt] /interface ppp-client> /delay 20; set modem disabled=no
no such item (4)

Notice the /delay 20 command before trying to enable the interface. The problem occurs only if
there is a pause after the last reference to the interface.

I observed this bug in v 6.41, reported and sent supout.rif on 2018-apr-14.
(I'm using RB951G 2HnD)

Posted: **Mon Apr 23, 2018 11:10 am**

Does anyone have tried to run CHR 6.42 with more than 7 connected NICs on vsphere? After upgrade to 6.42 I am getting reboot loop... before upgrade reboot loop was only if 10NICs are connected...
So I can't understand if that is related with my vsphere/VM configuration or RouterOS... More or less no response from support...
The error what I can see in shell before reboot is "No irq handler for vector...."

Posted: **Mon Apr 23, 2018 12:57 pm**

RB2011UiAS-2HnD after upgrade works fine for 4 days, but found some strange thing after 22 hours uptime I've got >3000 Sector writes.
Can anyone check is it for everyone or only my config. The same situation was about a year or two ago in latest v5 or earlier v6 versions of ROS and was fixed, maybe those problem returned.

Posted: **Mon Apr 23, 2018 1:10 pm**

Notice the /delay 20 command before trying to enable the interface.

Should that not be :delay 20;?

Posted: **Mon Apr 23, 2018 1:33 pm**

Notice the /delay 20 command before trying to enable the interface.
Should that not be :delay 20;?

..hmm. Interesting. I didn't knew that commands with : can be used with / instead of : ... or something like that.

[admin@pincegw] > :global a 10; /set a 5; /put $a
5

Is there any difference between for example :put and /put ?

Posted: **Mon Apr 23, 2018 3:05 pm**

i managed to fix my wifi drop issue on 6.42, had to reduce transmit power by 10 dB by inserting this value in antenna gain, day by day its getting impossible to use a mikrotik AP in a noisy environment, i was better off using my previous UBNT AP which was much more immune to noise.

the second complaint being since i bought this metal 52ac device more than a year back, i was told in future updates it would show the current transmit power table which till date hasnt been implemented at all.

Posted: **Mon Apr 23, 2018 3:23 pm**

Version 6.42.1 has been released in current channel:
viewtopic.php?f=21&t=133535

Posted: **Mon Apr 23, 2018 6:05 pm**

I have hap lite /smips/ with 6.41.2 . With 6.42 in station mode and WEP security /40bit/not work - can not connect. I downgrade to bugfix 6.40.7 and everything is fine - the routrboard is connected !

What is your AP - also RouterOS or some other equipment?

Posted: **Thu May 31, 2018 9:49 am**

Everyone who complained about the Netwatch issue - Please see this topic viewtopic.php?f=2&t=134538

MikroTik

v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]

Re: v6.42 [current]