v6.20 released!

Thu Oct 02, 2014 12:33 pm

What's new in 6.20 (2014-Oct-01 10:06):

*) cert scep - use fingerprints for transaction ids;
*) ipsec - support fqdn as my id;
*) fetch - allow fetching files larger than 4G;
*) fetch - fixed problem where files fetched over https were trimmed in size;
*) fixed problem - it was not possible to see %26 uninstall dude package;
*) stores are replaced with folders and disks are now managed under /disk menu;
*) added support for SMSC750x USB Gigabit Ethernet on x86;
*) ups - support selftest for smart and hid UPS;
*) pppoe client - increase connection timeout to make connection establishment
possible on busy pppoe server;
*) dhcp server - change default lease time from 3 days to 10 minutes
to avoid running out of IPs;
*) ipsec - allow binding modeconf address to username;
*) eoip/eoipv6/gre/gre6/ipip/ipipv6/6to4 tunnels have new features:
auto mtu (enabled by default for new tunnels);
dscp (inherit/specific value, inherit by default for new tunnels);
clamp-tcp-mss (yes by default for new tunnels);
*) eoip/gre/ipip/6to4 tunnels have dont-fragment option (inherit/no, no by default for new tunnels);
*) bridge has auto mtu feature (enabled by default for new bridges);
*) pppoe-server has auto mtu feature (enabled by default for new pppoe servers);

If you already run some RouterOS v6.x version, simply click “Check for updates” in QuickSet, Webfig or Winbox packages menu.

Others: http://www.mikrotik.com/download

zyzelis · Thu Oct 02, 2014 12:58 pm

What's new in 6.20 (2014-Oct-01 10:06):

*) cert scep - use fingerprints for transaction ids;
*) ipsec - support fqdn as my id;
*) fetch - allow fetching files larger than 4G;
*) fetch - fixed problem where files fetched over https were trimmed in size;
*) fixed problem - it was not possible to see %26 uninstall dude package;
*) stores are replaced with folders and disks are now managed under /disk menu;
*) added support for SMSC750x USB Gigabit Ethernet on x86;
*) ups - support selftest for smart and hid UPS;
*) pppoe client - increase connection timeout to make connection establishment
possible on busy pppoe server;
*) dhcp server - change default lease time from 3 days to 10 minutes
to avoid running out of IPs;
*) ipsec - allow binding modeconf address to username;
*) eoip/eoipv6/gre/gre6/ipip/ipipv6/6to4 tunnels have new features:
auto mtu (enabled by default for new tunnels);
dscp (inherit/specific value, inherit by default for new tunnels);
clamp-tcp-mss (yes by default for new tunnels);
*) eoip/gre/ipip/6to4 tunnels have dont-fragment option (inherit/no, no by default for new tunnels);
*) bridge has auto mtu feature (enabled by default for new bridges);
*) pppoe-server has auto mtu feature (enabled by default for new pppoe servers);

If you already run some RouterOS v6.x version, simply click “Check for updates” in QuickSet, Webfig or Winbox packages menu.

Others: http://www.mikrotik.com/download

No changes with wireless??

Thu Oct 02, 2014 1:13 pm

zyzelis, what exact changes do you want to see for wireless?

infused · Thu Oct 02, 2014 1:18 pm

/me waits for people to upgrade their access point 100km away and have it fail then complain on the forums

mishaM · Thu Oct 02, 2014 1:21 pm

hi sergejs

well be good if implement custom channel widths on C standart and MCS if be checkable manually

Basdno · Thu Oct 02, 2014 1:22 pm

/me waits for people to upgrade their access point 100km away and have it fail then complain on the forums

LOL

Maybe this is the time people learn to test in a Lab-setup first, then deploy to Production devices afterwards............ u never know!

However, that said, I do like Mikrotiks new policy of offering the Release Candidates publicly on downloadsite for some time prior to main releases for the last couple of releases. I think this is a very good way forward, giving MT a lot more feedback PRIOR to release and thus probably making the stable versions even better and STABLE then before!

Hope they continue this way of working!

Keep up the good work Mikrotik!

meetriks2 · Thu Oct 02, 2014 1:23 pm

ros code

[admin@xxxx] /ppp active> print
Flags: R - radius 
 #   NAME         SERVICE CALLER-ID         ADDRESS         UPTIME   ENCODING  
 0   probe        sstp   x     0.0.0.0         57s      AES256-CBC
 1   probe        sstp   x     0.0.0.0         43s      AES256-CBC
 2   probe        sstp   x     0.0.0.0         42s      AES256-CBC
 3   probe        sstp   x     0.0.0.0         39s      AES256-CBC
 4   probe        sstp   x     0.0.0.0         38s      AES256-CBC

The address is not 0.0.0.0 but a valid adres. It all works, this is a cosmetic bug only.
On v6.19 forum i posted the configuration already.

Hope its getting fixed in v6.21

Cheers!

napismizpravu · Thu Oct 02, 2014 1:32 pm

RB433UAH

6.20 (6.21rc3) no micro-sd disk ( microSDHC Kingston 4GB) > downgrade 6.19 disk view

I already wrote here, but it was deleted. (20 days ago)

Thu Oct 02, 2014 1:37 pm

meetriks2, thank you very much for the report.

napismizpravu, please provide us with support output file from your router (send to support@mikrotik.com).

zyzelis · Thu Oct 02, 2014 1:49 pm

zyzelis, what exact changes do you want to see for wireless?

Sergejs,
in changelog no word about wireless section, meanwile we know, that at least 6.20rc15 you have changed wireless to good side.

I thought, that changelog is for registering what was done with firmware(OS)
and for customers, who has updated to latest version and the if things are going not in expected way, to which version downgrade.

napismizpravu · Thu Oct 02, 2014 2:04 pm

napismizpravu, please provide us with support output file from your router (send to support@mikrotik.com).[/quote]

is it needed? (file repair folder - picture)

Another problem rename disk1 > usb1 (+reboot) = problem - picture

prawira · Thu Oct 02, 2014 2:28 pm

any explanation and or example for this ?

*) cert scep - use fingerprints for transaction ids;

Tq

Paul

Thu Oct 02, 2014 3:20 pm

Why Wireless-fp package is still outside the the all-package bundle? The wireless package inside is still legacy wireless package?

k0st3k · Thu Oct 02, 2014 4:03 pm

Uploaded to rb711, reboot and.... no wireless config and wlan1 is disabled --like default

Thu Oct 02, 2014 4:12 pm

jarda, wireless-fp package is included in the bundle. At 6.20 (not on rc releases) fp is included to bundle.
Please provide us with more information about your router, where wireless-fp is not included.

k0st3k,
make sure you have used the same wireless package before and after upgrade (for example wireless-fp was installed and now wireless is installed).

prawira · Thu Oct 02, 2014 4:25 pm

i am using CRS...
when i click check for updates, it say System is already up to date

check 4 update.png

when i upgrade manually with upload the *npk file and reboot, the funny things happend.

after manual upgrade.png

what are wrong ?

P

prawira · Thu Oct 02, 2014 4:30 pm

where is wireless-fp ??

all-packages-mipsbe-6.20.png

all the above files are from all-packages-mipsbe-6.20.zip

P

Thalid · Thu Oct 02, 2014 5:02 pm

where is wireless-fp ??
all-packages-mipsbe-6.20.png
all the above files are from all-packages-mipsbe-6.20.zip

P

happens when i check the package aswll

Torrent version for mipsbe contains it

Thu Oct 02, 2014 5:03 pm

prawira
wireless-fp is included to all-in-one routerboard package and it is available for download as separate package,
http://download2.mikrotik.com/routeros/ ... mipsbe.npk

There might be some issues with check-for-latest version, as it was reported 6.19 version. Now it should work fine.

Thu Oct 02, 2014 5:47 pm

zyzelis, what exact changes do you want to see for wireless?

More stability for ac Nv2

More tuning of the ac protocol

Thu Oct 02, 2014 5:50 pm

can Mikrotik make a clear and well illustrated presentation of fp?
This would be for many welcome and a greatfull help.

avenn · Thu Oct 02, 2014 5:51 pm

zyzelis, what exact changes do you want to see for wireless?
More stability for ac Nv2

More tuning of the ac protocol

plus 1 from me

mxmxmxmxmx · Thu Oct 02, 2014 5:59 pm

What's new in 6.20 (2014-Oct-01 10:06):

*) fixed problem - it was not possible to see %26 uninstall dude package;

Maybe now I can see dude on package list, but it is not possible to login to dude.

down to 6.18 - OK. Up to 6.20 Login failed.

usx · Thu Oct 02, 2014 6:05 pm

Had no issues with the upgrade on a 2011UiAS-2HnD.

But on a RB450G all but one of the Queue Graphs lost their data (the graph curve started as if it just got set up). The one which didn't lose the data is now showing an erroneous date (Nov/08/2150).

The 2011UiAS-2HnD had no Queue Graphs, so I can't compare.

The RB450G (and the 2011UiAS-2HnD) have their primary NTP servers set to al local (inside the LAN) GPS-based NTP server. The clocks on both devices show the correct time.

Shiro · Thu Oct 02, 2014 7:16 pm

CCR1009-8G-1S-1S+ no longer detecting SD Cards. Tried two sd cards that worked before. Tried restart without success.

lucky79 · Thu Oct 02, 2014 7:34 pm

RB922UAGS-5HPacD hanged after upgrade, needed power cycle to restore access.

minimum tx-power 6dBm, seriously guys? Current Tx power still shows 0dBm. Automatic Tx power based on antenna gain still not working. No manual noise floor threshold... No spectral scan... I guess I am holding the AC products until the wireless-fp will be ready for production. Its a shame I have already upgraded one PTP link with SXTacs which at the current state has lower performance and shitty CCQ (with R52n-M was perfect).

littlebill · Thu Oct 02, 2014 8:48 pm

has all the sstp and pptp issues been resolved from the 6.8 debacle?

littlecake · Thu Oct 02, 2014 9:10 pm

Please how it works auto MTU function by pppoe?

k0st3k · Thu Oct 02, 2014 11:07 pm

k0st3k,
make sure you have used the same wireless package before and after upgrade (for example wireless-fp was installed and now wireless is installed).

5.26 > 6.20

tolkotov · Thu Oct 02, 2014 11:19 pm

when upgrading from 6.1 to 6.20 tik ccr1036 went in down

.

eltopia · Fri Oct 03, 2014 4:29 am

I'm finding that going from 6.19 -> 6.20 breaks 6to4 tunnels on x86. Not sure on mips, though I could test if needed. (Running 6to4 between RB750GL and x86 VM)

Fri Oct 03, 2014 7:56 am

I have 2 EOIP-tunnels and working good after the upgrade.
This with omnitik and crs125-24G-1S-RM no problem

finalcutroot · Fri Oct 03, 2014 9:00 am

the latest update on hot spot was in version 6.0 the (login by mac cookies)
since then no updates

any way as i use MT for hotspot / update to 6.20 no problem

Thanks

msaxl · Fri Oct 03, 2014 9:29 am

After I upgrade to v6.20 I have the problem that with v6.19 it was possible to set a my-user-fqdn also in psk main mode.
I know that this does not make much sense, but if you must connect to a juniper from behind a firewall, the my-id does not match the public ip address and the juniper ipsec does reject the connection when it first sees the "wrong" id. The workaround is setting the user-fqdn on both sides and binding the psk key on the public ip address on the juniper side.

But with version 6.20 setting user-fqdn or fqdn is rejected, so this is a showstopper for me

arturw · Fri Oct 03, 2014 12:42 pm

OVPN server "require-client-certificate" set and CRL option set is still not working.

my case number: [Ticket#2014082766000625]

I found that other people have the same problem.
http://forum.mikrotik.com/viewtopic.php ... 60#p449760

a.devecerski · Fri Oct 03, 2014 12:50 pm

Winbox v2.x still works?

skillful · Fri Oct 03, 2014 1:25 pm

Winbox v2.x still works?

Yes, Winbox 2.2.18 still works.

fabiandr · Fri Oct 03, 2014 2:19 pm

Please how it works auto MTU function by pppoe?

steen · Fri Oct 03, 2014 3:13 pm

Hello Folks!

On each device we have upgraded (except RB411) from RoS6.19 to RoS6.20 we got alarms on DISK !?

There is no problem with disk, it is not full or damaged on any device.

What is this ?

rzirzi · Fri Oct 03, 2014 3:27 pm

changelog: pppoe-server has auto mtu feature (enabled by default for new pppoe servers)

Where is that feature? When I can enable/disable/set that feature? (Auto MTU for PPPoE server)

abis · Fri Oct 03, 2014 3:40 pm

Hello,

after 6.19 -> 6.20 the monitor service disk for the dude doesn't work anymore...

p.s. *) stores are replaced with folders and disks are now managed under /disk menu;
I can create folders from winbox or i need to create via console?

JanezFord · Fri Oct 03, 2014 5:12 pm

After upgrade all of my static DHCP leases on CRS125 rack mount were lost ... on other devices upgrade went well and on dhcp leases were lost. Already contacted support.

JF.

Fri Oct 03, 2014 5:20 pm

I have update to with all-packages on crs125-24G-1S-RM
I don't lost my static ip's

SiB · Fri Oct 03, 2014 5:36 pm

topic about v6.19 was lock then I write in this place in my situation.

Upgrade from v6.4 to 6.19 at 27.09.2014. On calendar:

all services are monitor by Nagios and TheDude - all stuff works by 3days.

My problem is on voip communication on small CallCenter (20 users 24/7) and rest voip phones (300users in 40branches in UE). NAT helper (ip firewall service-port) a SIP is always disable on all MikroTiks - voip was work perfect.

Problem:
On 1 at 10 calls have a missing voice in one direction - this was start after 3days. I try connect voip SIPTrunks to other ISP but it was even worst (90% RTP was as UDP unreply), remove connection from VoIP server to SIPTrunks is not work on long time.

Solution:
Today I must do downgrade to 6.4 and now >90% RTP of UDP are Assured status and all voip works perfect on v6.4.
It was my 3th to do upgrade to higer version then 6.4

.

SupOut files (before and after downgrade) and (day before) was send to MikroTik Support Team, I still wait for answers.
The same configuration was on v6.4 and on v6.19, v6.4 works "perfect" (without bugs like auto reboots by L2TP

).

For me, v6.19 isn't stable and v6.20 and v6.21RCx haven't any fix on this.

voxframe · Fri Oct 03, 2014 6:39 pm

I have also noticed problems starting with VOIP traversing NAT.

It worked fine, then as things were upgraded, it started to break. Give one way audio, etc.

Moving back down to very early firmware.

docmarius · Fri Oct 03, 2014 6:58 pm

Reposted in Forwarding Protocols since it is not 9.20 specific.
http://forum.mikrotik.com/viewtopic.php?f=14&t=89893

Fri Oct 03, 2014 6:59 pm

when upgrading from 6.1 to 6.20 tik ccr1036 went in down .

what exactly happened? Can you see in the Winbox by Layer2 connection?
Are you able to connect to the CCR via serial console?

Fri Oct 03, 2014 7:04 pm

RB922UAGS-5HPacD hanged after upgrade, needed power cycle to restore access.

minimum tx-power 6dBm, seriously guys? Current Tx power still shows 0dBm. Automatic Tx power based on antenna gain still not working. No manual noise floor threshold... No spectral scan... I guess I am holding the AC products until the wireless-fp will be ready for production. Its a shame I have already upgraded one PTP link with SXTacs which at the current state has lower performance and shitty CCQ (with R52n-M was perfect).

Current-tx-power reading isn't made yet for the AC radios.
TX-power changing is working on the AC.
What signal you have on your SXT AC link?
If you set the band to the A/N on both sides do you get the same CCQ and performance like with the R52nM cards?
Try to reset the wireless interface configuration and check if the situation gets bettwe with default values.

rmmccann · Fri Oct 03, 2014 7:28 pm

has all the sstp and pptp issues been resolved from the 6.8 debacle?

Would like to know as well. Have held at 6.7 because I need to maintain SSTP compatibility with Windows clients.

steen · Fri Oct 03, 2014 11:15 pm

Hello,

after 6.19 -> 6.20 the monitor service disk for the dude doesn't work anymore...

p.s. *) stores are replaced with folders and disks are now managed under /disk menu;
I can create folders from winbox or i need to create via console?

Does this mean we soon now will se a new version of The Dude to ?
It is really needed a new The Dude version, a stable release, we have been using the same Dude for years now..

avenn · Sat Oct 04, 2014 4:14 pm

Random SXTac kernel failures still evident

Netinstall 6.20
Nv2
PtP SXTacSA to SXTac
100 metres
-49/50dBm
Very simple config
Very clean freg scan using airMAX

Works great at first then stops responding then kernel errors.

Switched to 802.11 rather than Nv2 equalled lost comma to SXTacSA.

Switching back to airMAX for external wireless for now. Customer is becoming unhappy.

Will try again in 6 to 12 months.

poizzon · Sat Oct 04, 2014 7:14 pm

18:31:36 system,info verified ntp-6.20-mipsbe.npk
18:31:36 system,info verified ppp-6.20-mipsbe.npk
18:31:36 system,info verified security-6.20-mipsbe.npk
18:31:36 system,info verified advanced-tools-6.20-mipsbe.npk
18:31:36 system,info verified system-6.20-mipsbe.npk
18:31:36 system,info installed system-6.20
18:31:36 system,info installed advanced-tools-6.20
18:31:36 system,info installed security-6.20
18:31:36 system,info installed ppp-6.20
18:31:36 system,info installed ntp-6.20
18:31:36 system,info installed dhcp-6.20
18:31:55 interface,info ether1-gateway link up (speed 100M, full duplex)
18:31:55 interface,info ether2-master-local link up (speed 100M, full duplex)
18:32:05 l2tp,info first L2TP UDP packet received from AAA.BBB.CCC.DDD
18:32:06 l2tp,ppp,info,account NewYork logged in, 172.16.2.2
18:32:07 l2tp,ppp,info l2tp-NewYork: authenticated
18:32:07 l2tp,ppp,info l2tp-NewYork: connected
18:33:29 l2tp,info first L2TP UDP packet received from EEE.FFF.GGG.HHH
18:33:30 l2tp,ppp,info,account AmsTerDam logged in, 172.16.0.2
18:33:30 l2tp,ppp,info l2tp-AmsTerDam: authenticated
18:33:30 l2tp,ppp,info l2tp-AmsTerDam: connected
18:34:24 ipsec,error phase1 negotiation failed due to time up XXX.YYY.ZZZ.XYZ[500]<=>172.16.0.2[500] d3315d1866b8c332:0000000000000000
18:34:25 ipsec,error phase1 negotiation failed due to time up XXX.YYY.ZZZ.XYZ[500]<=>172.16.3.2[500] 16586dd9987fdd61:0000000000000000
18:34:25 ipsec,error phase1 negotiation failed due to time up XXX.YYY.ZZZ.XYZ[500]<=>172.16.2.2[500] eb7dcfc791c0bee6:0000000000000000
18:35:20 system,info,account user poi logged in from xx.xx.xx.xx via winbox
18:35:25 ipsec,error phase1 negotiation failed due to time up XXX.YYY.ZZZ.XYZ[500]<=>172.16.3.2[500] 093864a77ece5743:0000000000000000
18:36:25 ipsec,error phase1 negotiation failed due to time up XXX.YYY.ZZZ.XYZ[500]<=>172.16.3.2[500] 4c4becbcae12daaf:0000000000000000
18:37:25 ipsec,error phase1 negotiation failed due to time up XXX.YYY.ZZZ.XYZ[500]<=>172.16.3.2[500] 4f9ce96e1d5c59fc:0000000000000000
18:38:35 ipsec,error phase1 negotiation failed due to time up XXX.YYY.ZZZ.XYZ[500]<=>172.16.3.2[500] deb801af0ba9d6c9:0000000000000000
18:39:45 ipsec,error phase1 negotiation failed due to time up XXX.YYY.ZZZ.XYZ[500]<=>172.16.3.2[500] 1dcfedef4e117603:0000000000000000
18:40:45 ipsec,error phase1 negotiation failed due to time up XXX.YYY.ZZZ.XYZ[500]<=>172.16.3.2[500] cd73a72de5a4bac4:0000000000000000
18:41:55 ipsec,error phase1 negotiation failed due to time up XXX.YYY.ZZZ.XYZ[500]<=>172.16.3.2[500] 283e9d748f979c57:0000000000000000
18:43:05 ipsec,error phase1 negotiation failed due to time up XXX.YYY.ZZZ.XYZ[500]<=>172.16.3.2[500] 1ec10de96c1e35d1:0000000000000000
18:44:05 ipsec,error phase1 negotiation failed due to time up XXX.YYY.ZZZ.XYZ[500]<=>172.16.3.2[500] 5242da5b92f02b49:0000000000000000
18:45:15 ipsec,error phase1 negotiation failed due to time up XXX.YYY.ZZZ.XYZ[500]<=>172.16.3.2[500] af1788a953e26d08:0000000000000000
18:46:15 ipsec,error phase1 negotiation failed due to time up XXX.YYY.ZZZ.XYZ[500]<=>172.16.3.2[500] ff0ccd07cae83c2e:0000000000000000
18:47:15 ipsec,error phase1 negotiation failed due to time up XXX.YYY.ZZZ.XYZ[500]<=>172.16.3.2[500] 808e73705a1348d8:0000000000000000
18:48:25 ipsec,error phase1 negotiation failed due to time up XXX.YYY.ZZZ.XYZ[500]<=>172.16.3.2[500] 6035bc208336ee44:0000000000000000
18:49:35 ipsec,error phase1 negotiation failed due to time up XXX.YYY.ZZZ.XYZ[500]<=>172.16.3.2[500] 01b1c837e087c94c:0000000000000000
18:50:35 ipsec,error phase1 negotiation failed due to time up XXX.YYY.ZZZ.XYZ[500]<=>172.16.3.2[500] 981177407414de2b:0000000000000000
18:51:45 ipsec,error phase1 negotiation failed due to time up XXX.YYY.ZZZ.XYZ[500]<=>172.16.3.2[500] 0dc357f7303910f6:0000000000000000
18:52:55 ipsec,error phase1 negotiation failed due to time up XXX.YYY.ZZZ.XYZ[500]<=>172.16.3.2[500] 5f75d3969bcac9a1:0000000000000000
18:53:55 ipsec,error phase1 negotiation failed due to time up XXX.YYY.ZZZ.XYZ[500]<=>172.16.3.2[500] f788b880bcbb35af:0000000000000000
18:55:05 ipsec,error phase1 negotiation failed due to time up XXX.YYY.ZZZ.XYZ[500]<=>172.16.3.2[500] a70224a893e2dd19:0000000000000000
18:56:15 ipsec,error phase1 negotiation failed due to time up XXX.YYY.ZZZ.XYZ[500]<=>172.16.3.2[500] a18784aafea7ce3e:0000000000000000
18:57:15 ipsec,error phase1 negotiation failed due to time up XXX.YYY.ZZZ.XYZ[500]<=>172.16.3.2[500] 600e0d1fcba18246:0000000000000000
18:58:25 ipsec,error phase1 negotiation failed due to time up XXX.YYY.ZZZ.XYZ[500]<=>172.16.3.2[500] aee578115f534004:0000000000000000
18:59:35 ipsec,error phase1 negotiation failed due to time up XXX.YYY.ZZZ.XYZ[500]<=>172.16.3.2[500] 89f6b5acc9731f3a:0000000000000000
19:00:35 ipsec,error phase1 negotiation failed due to time up XXX.YYY.ZZZ.XYZ[500]<=>172.16.3.2[500] 83cea2c786936e7c:0000000000000000
19:01:45 ipsec,error phase1 negotiation failed due to time up XXX.YYY.ZZZ.XYZ[500]<=>172.16.3.2[500] 58f4bf218c569a07:0000000000000000
19:02:55 ipsec,error phase1 negotiation failed due to time up XXX.YYY.ZZZ.XYZ[500]<=>172.16.3.2[500] acd4b68e28f0a7ff:0000000000000000

poizzon · Sat Oct 04, 2014 7:31 pm

p.s.: after disabling inactive IPsec Peer - errors gone.

Sat Oct 04, 2014 9:59 pm

I'm trying to install x86 on a fresh VirtualBox VM, and on the very first run of the ISO (i.e. before getting the installation dialog), I get this:

2014-10-04_21-56-27.png

after which the VM reboots, and the same thing happens again.

I have a suspicion some of the other things reported so far are related to this.

(6.19 installs fine on the same VM on the same host machine)

pitr · Sat Oct 04, 2014 10:20 pm

My 6to4 tunnel to tunnelbroker.net broke with 6.20. Can't even get them working again if I redo the configuration completely.

ponko · Sun Oct 05, 2014 3:05 am

RB433UAH

6.20 (6.21rc3) no micro-sd disk ( microSDHC Kingston 4GB) > downgrade 6.19 disk view

I already wrote here, but it was deleted. (20 days ago)

same case here, and already sent support output file to support@mikrotik.com).

Belgarion186 · Sun Oct 05, 2014 4:44 am

My 6to4 tunnel to tunnelbroker.net broke with 6.20. Can't even get them working again if I redo the configuration completely.

Thanks for the heads up.

I was too busy the past few days to upgrade mine to 6.20 (they're currently on 6.19). Guess I'll leave them at 6.19 till the problem is fixed.

lucky79 · Sun Oct 05, 2014 11:48 am

RB922UAGS-5HPacD hanged after upgrade, needed power cycle to restore access.

minimum tx-power 6dBm, seriously guys? Current Tx power still shows 0dBm. Automatic Tx power based on antenna gain still not working. No manual noise floor threshold... No spectral scan... I guess I am holding the AC products until the wireless-fp will be ready for production. Its a shame I have already upgraded one PTP link with SXTacs which at the current state has lower performance and shitty CCQ (with R52n-M was perfect).
Current-tx-power reading isn't made yet for the AC radios.
TX-power changing is working on the AC.
What signal you have on your SXT AC link?
If you set the band to the A/N on both sides do you get the same CCQ and performance like with the R52nM cards?
Try to reset the wireless interface configuration and check if the situation gets bettwe with default values.

In 6.19 I could set manual tx-power to 0 but starting 6.20 the driver wont let me go below 6dBm (on Netmetal 5).... So apparently there is a change in wireless but its not in the changelog. Yes, manual tx-power works but not the automatic based on the antenna gain... at least it did not work the way I am used to from previous versions.

I have -57/-54 signal on the link with 0 tx-power (its 50m link across the road), I cannot get close to my previous setup with R52n-M and standard wireless package with 5.26... Even if I switch the SXTs to A/N its similar to AC.. lower CCQ, lower throughput (40MHz jumping between 100 and 170Mbps).

When I tried wireless-fp with R52n-M on another link some time ago (I guess it was 6.15) I have seen similar issues - lower CCQ and performance, so switched back to default wireless package = stable 99-100% CCQ and throughput

ditonet · Sun Oct 05, 2014 6:33 pm

pitr wrote:
My 6to4 tunnel to tunnelbroker.net broke with 6.20. Can't even get them working again if I redo the configuration completely.

Thanks for the heads up.

I was too busy the past few days to upgrade mine to 6.20 (they're currently on 6.19). Guess I'll leave them at 6.19 till the problem is fixed.

My 6to4 tunnel to tunnelbroker.net works as expected after upgrade to 6.20.

Regards,

newranman · Mon Oct 06, 2014 1:21 am

I have had multiple kernel failures then reboots on my RB450G, submitted report to my account support. Ticket#2014100466000271, just sent another autosupout. Had to downgrade to 6.19

I don't recommend 6.20 at least for my setup.

Ran

bjh21 · Mon Oct 06, 2014 1:37 am

My 6to4 connection failed too, when I upgraded from 6.13 to 6.20. I removed it all and followed the instructions at http://wiki.mikrotik.com/wiki/Setting_u ... l_via_6to4 with appropriate changes for the addressing on my network:

[admin@kattegat] > /interface 6to4 add mtu=1280 name=ipng-tunnel local-address=86.26.1.97 remote-address=192.88.99.1 disabled=no
[admin@kattegat] > /ipv6 address add address=2002:561a:0161::1/3 interface=ipng-tunnel
[admin@kattegat] > /ipv6 route add dst-address=2000::/3 gateway=::192.88.99.1%ipng-tunnel

But this leaves the routing table claiming that the interface is unreachable:

[admin@kattegat] > /ipv6 route print detail
Flags: X - disabled, A - active, D - dynamic, 
C - connect, S - static, r - rip, o - ospf, b - bgp, U - unreachable 
 0 ADC  dst-address=2000::/3 gateway=ipng-tunnel 
        gateway-status=ipng-tunnel unreachable distance=0 scope=10 

 1   S  dst-address=2000::/3 gateway=::192.88.99.1%ipng-tunnel 
        gateway-status=::192.88.99.1%ipng-tunnel unreachable distance=1 
        scope=30 target-scope=10 

 2 ADC  dst-address=2002:561a:161:1::/64 gateway=ether2-master-local 
        gateway-status=ether2-master-local reachable distance=0 scope=10 

 3 ADC  dst-address=2002:561a:161:2::/64 gateway=wlan1 
        gateway-status=wlan1 reachable distance=0 scope=10

Majklik · Mon Oct 06, 2014 9:53 am

[admin@kattegat] > /ipv6 address add address=2002:561a:0161::1/3 interface=ipng-tunnel
[admin@kattegat] > /ipv6 route add dst-address=2000::/3 gateway=::192.88.99.1%ipng-tunnel

This is not a probably OK. A maybe better:
/ipv6 address add address=2002:561a:0161::1/16 interface=ipng-tunnel
/ipv6 route add dst-address=::/0 gateway=::192.88.99.1%ipng-tunnel

My 6to4 tunnel to tunnelbroker.net broke with 6.20. Can't even get them working again if I redo the configuration completely.

Check how you have defined IPv6 address and default gateway on the tunnel. It looks that 6.20 is more strictly that previous version. The definition like this works with 6.20 OK:
/ipv6 address add address=2001:db8

5678::2/64 advertise=no interface=wan6he
/ipv6 route add check-gateway=ping distance=1 gateway=2001:db8

5678::1
The default gateway in this way do not (older ROS version probably accepts this):
/ipv6 route add distance=1 gateway=wan6he

MetUys · Mon Oct 06, 2014 10:34 am

Hi,

Upgraded RB751U-2HnD (v6.19 -> v6.20) = no problems as yet.
Upgraded RB2011UAS-2HnD (v6.19 -> v6.20) = IP Routes got lost, Queues also went missing and obviously the Graphs on the Queues were not linking correctly (no queue to link to).
I manually added back the routes and queues and fixed the graph links from an export done just before hand, and all is working again.
(no other noticeable issues as yet).

PS: Regarding the PPTP issues on v6.8 = I suffered from these but in the last few versions things have been good, no issues on our setups.

Conclusion: be careful on remote installs in case similar issues come through for you, as it did on my RB2011UAS-2HnD, in our case we had someone on the other end before we did the upgrade. (upgrade rollout planning is key

)

Mon Oct 06, 2014 12:15 pm

Conclusion: be careful on remote installs in case similar issues come through for you, as it did on my RB2011UAS-2HnD, in our case we had someone on the other end before we did the upgrade. (upgrade rollout planning is key )

Conclusion: Wait for a stable release.

pitr · Mon Oct 06, 2014 1:28 pm

[admin@kattegat] > /ipv6 address add address=2002:561a:0161::1/3 interface=ipng-tunnel
[admin@kattegat] > /ipv6 route add dst-address=2000::/3 gateway=::192.88.99.1%ipng-tunnel
This is not a probably OK. A maybe better:
/ipv6 address add address=2002:561a:0161::1/16 interface=ipng-tunnel
/ipv6 route add dst-address=::/0 gateway=::192.88.99.1%ipng-tunnel

My 6to4 tunnel to tunnelbroker.net broke with 6.20. Can't even get them working again if I redo the configuration completely.
Check how you have defined IPv6 address and default gateway on the tunnel. It looks that 6.20 is more strictly that previous version. The definition like this works with 6.20 OK:
/ipv6 address add address=2001:db85678::2/64 advertise=no interface=wan6he
/ipv6 route add check-gateway=ping distance=1 gateway=2001:db85678::1
The default gateway in this way do not (older ROS version probably accepts this):
/ipv6 route add distance=1 gateway=wan6he

Ok, that fixed it. Thanks

vadimbn · Mon Oct 06, 2014 3:15 pm

Dear developers, please see the attached screenshots. Can You explain, where are my disks? Before update to 6.20 i have two disks - internal and sd-card. After - nothing.

Mon Oct 06, 2014 3:18 pm

Sorry about that, this is a confirmed bug, and we are already working on it. You can currently downgrade to v6.19 to see microSD cards

Mon Oct 06, 2014 3:34 pm

[admin@kattegat] > /ipv6 address add address=2002:561a:0161::1/3 interface=ipng-tunnel
[admin@kattegat] > /ipv6 route add dst-address=2000::/3 gateway=::192.88.99.1%ipng-tunnel
This is not a probably OK. A maybe better:
/ipv6 address add address=2002:561a:0161::1/16 interface=ipng-tunnel
/ipv6 route add dst-address=::/0 gateway=::192.88.99.1%ipng-tunnel

My 6to4 tunnel to tunnelbroker.net broke with 6.20. Can't even get them working again if I redo the configuration completely.
Check how you have defined IPv6 address and default gateway on the tunnel. It looks that 6.20 is more strictly that previous version. The definition like this works with 6.20 OK:
/ipv6 address add address=2001:db85678::2/64 advertise=no interface=wan6he
/ipv6 route add check-gateway=ping distance=1 gateway=2001:db85678::1
The default gateway in this way do not (older ROS version probably accepts this):
/ipv6 route add distance=1 gateway=wan6he
Ok, that fixed it. Thanks

We also managed to reproduce this issue which appears after upgrade. Please follow our changelog on download page and wait for version which will include fix for it.
http://www.mikrotik.com/download

lcrhea25 · Mon Oct 06, 2014 4:38 pm

Normis anything on my Ticket#2014092966000638. Haven't heard anything back. Colocation of 3 120 degree 2.4 sectors. Problems when all three run nv2. Put 1 on channel 1 with nv2, 1 on channel 11 nv2 and 1 on channel 6 nv2 and problem goes away. Basically I am interfereing with myself when running nv2 5 channels away.

remojames · Mon Oct 06, 2014 7:50 pm

Somebody make firmware update from 3.18 to 3.19 ?
where can i find change log to this firmware?

ffernandes · Mon Oct 06, 2014 8:56 pm

it should be here: http://wiki.mikrotik.com/wiki/RouterBOOT_changelog

but its a little *outdated

Somebody make firmware update from 3.18 to 3.19 ?
where can i find change log to this firmware?

bjh21 · Mon Oct 06, 2014 10:42 pm

[admin@kattegat] > /ipv6 address add address=2002:561a:0161::1/3 interface=ipng-tunnel
[admin@kattegat] > /ipv6 route add dst-address=2000::/3 gateway=::192.88.99.1%ipng-tunnel
This is not a probably OK. A maybe better:
/ipv6 address add address=2002:561a:0161::1/16 interface=ipng-tunnel
/ipv6 route add dst-address=::/0 gateway=::192.88.99.1%ipng-tunnel

Thanks for the suggestion. That's actually the configuration I was using before the upgrade, and it has precisely the same problem:

[admin@kattegat] /ipv6 route> print detail
Flags: X - disabled, A - active, D - dynamic, 
C - connect, S - static, r - rip, o - ospf, b - bgp, U - unreachable 
 0   S  dst-address=::/0 gateway=::192.88.99.1%ipng-tunnel 
        gateway-status=::192.88.99.1%ipng-tunnel unreachable distance=1 
        scope=30 target-scope=10 

 1 ADC  dst-address=2002::/16 gateway=ipng-tunnel 
        gateway-status=ipng-tunnel unreachable distance=0 scope=10 

 2 ADC  dst-address=2002:561a:161:1::/64 gateway=ether2-master-local 
        gateway-status=ether2-master-local reachable distance=0 scope=10 

 3 ADC  dst-address=2002:561a:161:2::/64 gateway=wlan1 
        gateway-status=wlan1 reachable distance=0 scope=10

Ansy · Tue Oct 07, 2014 9:33 am

0 items in Tools-->Profile on some upgraded devices

I can't realize what this issue depends of...

Tue Oct 07, 2014 10:14 am

0 items in Tools-->Profile on some upgraded devices
I can't realize what this issue depends of...

Thank you about this information. This issue is already reproduced and will be fixed.

cdiedrich · Tue Oct 07, 2014 11:39 am

Upgraded one of our 2011 yesterday from 6.12 to 6.20.
All IPsec installed SAs now show up with state=dying but are fully functional.
Expiration time is always more than 5min...

Since it's basically working, I consider this a cosmetic bug

-Chris

24tammam · Tue Oct 07, 2014 8:58 pm

Dear developers, please see the attached screenshots. Can You explain, where are my disks? Before update to 6.20 i have two disks - internal and sd-card. After - nothing.

i have this

what i can to do ?
I do not want to return 6.19!
I want to repair the problem ؟

finnter · Tue Oct 07, 2014 10:29 pm

| fetch - fixed problem where files fetched over https were trimmed in size;

This doesn't appear to be fixed. Before files were being trimmed down to 1024 B, they are now being trimmed down to 4096 B.

levifig · Wed Oct 08, 2014 12:02 am

Upgraded a couple of 2011 and have had all sorts of trouble with the DHCP client, particularly on ETH1. No IP, or no default route, or no DNS, or nothing, even though other devices get DHCP just fine on the same physical cable. (RB2011UiAS)

Waiting for 6.21 now…

strokers · Wed Oct 08, 2014 2:22 pm

Hello

I found 2 bugs:
1) i use ppp tunel - when is one person connected by ppp to mikrotik all it's ok when the second person connecting after few minutes system is 100% cpu load and i must reboot (unplug 230v)

2) when i conneting by ppp from other network and i have in this network i have one address ip for ex. 192.168.1.200 now i diconnecting and connecting from other computer with other ip address for ex. 192.168.1.201 i cant connect - i must wait about 10-15 minutes and then i can connect.

the first problem is on 6.20 the second problem is now on 6.2 and on old software when i bought the 751U-2HnD (4.17)

Strokers

strokers · Wed Oct 08, 2014 2:24 pm

Hello

I found 2 bugs:
1) i use ppp tunel - when i use it - when i connect by ppp after few minutes system is 100% cpu load and i must reboot (unplug 230v)

2) when i conneting by ppp from other network and i have in this network i have one address ip for ex. 192.168.1.200 now i diconnecting and connecting from other computer with other ip address for ex. 192.168.1.201 i cant connect - i must wait about 10-15 minutes and then i can connect.

the first problem is on 6.20 the second problem is now on 6.2 and on old software when i bought the 751U-2HnD (4.17)

Strokers

zdiv · Wed Oct 08, 2014 2:35 pm

RB922UAGS-5HPacD hanged after upgrade, needed power cycle to restore access.

minimum tx-power 6dBm, seriously guys? Current Tx power still shows 0dBm. Automatic Tx power based on antenna gain still not working. No manual noise floor threshold... No spectral scan... I guess I am holding the AC products until the wireless-fp will be ready for production. Its a shame I have already upgraded one PTP link with SXTacs which at the current state has lower performance and shitty CCQ (with R52n-M was perfect).
Current-tx-power reading isn't made yet for the AC radios.
TX-power changing is working on the AC.
What signal you have on your SXT AC link?
If you set the band to the A/N on both sides do you get the same CCQ and performance like with the R52nM cards?
Try to reset the wireless interface configuration and check if the situation gets bettwe with default values.
In 6.19 I could set manual tx-power to 0 but starting 6.20 the driver wont let me go below 6dBm (on Netmetal 5).... So apparently there is a change in wireless but its not in the changelog. Yes, manual tx-power works but not the automatic based on the antenna gain... at least it did not work the way I am used to from previous versions.

I have -57/-54 signal on the link with 0 tx-power (its 50m link across the road), I cannot get close to my previous setup with R52n-M and standard wireless package with 5.26... Even if I switch the SXTs to A/N its similar to AC.. lower CCQ, lower throughput (40MHz jumping between 100 and 170Mbps).

When I tried wireless-fp with R52n-M on another link some time ago (I guess it was 6.15) I have seen similar issues - lower CCQ and performance, so switched back to default wireless package = stable 99-100% CCQ and throughput

Same here plus:
911G-5HPacD hanged after upgrade , needed power cycle to revive
SXT G-5HPacD upgrade process went ok
I am staying on 6.19 ... 6.20 seams to be unreliable with to much problems...

Kickoleg · Wed Oct 08, 2014 5:02 pm

What is this "SIP Direct Media" ?
Where information about this new feature ?

ditonet · Wed Oct 08, 2014 6:00 pm

This 'new' option exists even in ROS 5.x and is described here:
http://wiki.mikrotik.com/wiki/Manual:IP ... vice_Ports

Regards,

zervan · Wed Oct 08, 2014 6:56 pm

has all the sstp and pptp issues been resolved from the 6.8 debacle?
Would like to know as well. Have held at 6.7 because I need to maintain SSTP compatibility with Windows clients.

I am using SSTP with 6.20 (and 6.19) with Windows clients and I didn't experienced troubles. What kind of issues should I check?

rmmccann · Wed Oct 08, 2014 7:12 pm

has all the sstp and pptp issues been resolved from the 6.8 debacle?
Would like to know as well. Have held at 6.7 because I need to maintain SSTP compatibility with Windows clients.
I am using SSTP with 6.20 (and 6.19) with Windows clients and I didn't experienced troubles. What kind of issues should I check?

For a while there people were complaining about the inability to establish SSTP tunnels from Windows as well as stability issues with them from MT to MT.

I never could get a concrete yes or no from anyone using SSTP whether or not they had success with Windows clients since 6.7. Your response is encouraging.

jondavy · Wed Oct 08, 2014 10:09 pm

In CloudCore in /system resources not appear the Status of:
Sector Writes Sice Reboot
Total Sector Wrires
Bad Blocks

avenn · Wed Oct 08, 2014 11:18 pm

has all the sstp and pptp issues been resolved from the 6.8 debacle?
Would like to know as well. Have held at 6.7 because I need to maintain SSTP compatibility with Windows clients.

SSTP to win7 clients is fine for me on 6.19

MTeeker · Thu Oct 09, 2014 2:14 am

V6.20 minor bug?

Time, Date, CPU, Memory and Uptime indicators disappear when Winbox is next opened.

It's minor thing but if it can be fixed, it would be great.

I am running the latest version of Winbox. I don't think it works on the previous version of Winbox either.

Many thanks.

scampbell · Thu Oct 09, 2014 9:37 am

We have upgraded several RB951 series routers in a class situation. 2 or 3 of these have all exhibited odd firewall behavior when rules are disabled.

It shows a rule disabled in Winbox yet the rule continues to work, e.g a rule that logs traffic keeps logging even when disabled. We did not test this in CLI as time was tight.

A reboot fixes the issue.

Not all devices have shown this so perhaps user error - has anyone else seen this ?

Kickoleg · Thu Oct 09, 2014 11:54 am

This 'new' option exists even in ROS 5.x and is described here:
http://wiki.mikrotik.com/wiki/Manual:IP ... vice_Ports

Regards,

SIP Direct Media appeared only in 6.20, You need more accurate read my question ...
And I interesting in how it work? What changes it made and etc...

ditonet · Thu Oct 09, 2014 3:46 pm

ros code

[grzegorz@Some_Gateway] /ip firewall service-port> export 
# oct/09/2014 13:38:06 by RouterOS 5.26
#
/ip firewall service-port
set ftp disabled=no ports=21
set tftp disabled=no ports=69
set irc disabled=no ports=6667
set h323 disabled=no
set sip disabled=no ports=5060,5061 sip-direct-media=no
set pptp disabled=no

All I wanted to say, is that 'sip-direct-media' option is not new feature introduced in ROS 6.20.

And I interesting in how it work

Just Google it...

Regards,

mgob1985 · Thu Oct 09, 2014 6:32 pm

We had an issue going from 6.15 to 6.20 on a CCR1009:

ros code

> /export compact  
# oct/09/2014 08:18:04 by RouterOS 6.20
#error exporting /ip ipsec mode-config
#error exporting /ip ipsec policy group
#error exporting /ip ipsec proposal
#error exporting /snmp community
#error exporting /certificate scep-server
#error exporting /certificate scep-server ra
#error exporting /ip ipsec peer
#error exporting /ip ipsec policy
#error exporting /ip ssh
#error exporting /routing bfd interface
#error exporting /snmp
#error exporting /tool e-mail
#error exporting /tool mac-server mac-winbox

> /snmp print 

error - contact MikroTik support and send a supout file (2)

I downgraded to 6.19 and everything worked again.... anyone else experience something like this?

Shiro · Thu Oct 09, 2014 9:48 pm

RB433UAH

6.20 (6.21rc3) no micro-sd disk ( microSDHC Kingston 4GB) > downgrade 6.19 disk view

I already wrote here, but it was deleted. (20 days ago)
same case here, and already sent support output file to support@mikrotik.com).

Same issue with CCR1009. I created a ticket to mt.

Shiro · Thu Oct 09, 2014 10:00 pm

What is this "SIP Direct Media" ?
Where information about this new feature ?

You can find more about this in iptables documentation. It's a helper that rewrite SIP Header if you using NAT.

robdeep · Fri Oct 10, 2014 4:22 am

Only issue I've noticed with 6.20 is with a bandwidth Queue I previously had defined in RouterOS but not enabled or applied to an interface. When I upgraded to 6.20, the queue became enabled and was being applied to the entire router, even though there was no "target" specified in the queue definition. So weird... I just disabled, it and all is good again. This was on a RB2011 btw.

Not a big deal for me, but I would be worried if I was a service provider and had a ton of customer queues setup, which I've used a lot in the past in my WISP days.

Fri Oct 10, 2014 6:19 pm

We had an issue going from 6.15 to 6.20 on a CCR1009:

ros code

> /export compact  
# oct/09/2014 08:18:04 by RouterOS 6.20
#error exporting /ip ipsec mode-config
#error exporting /ip ipsec policy group
#error exporting /ip ipsec proposal
#error exporting /snmp community
#error exporting /certificate scep-server
#error exporting /certificate scep-server ra
#error exporting /ip ipsec peer
#error exporting /ip ipsec policy
#error exporting /ip ssh
#error exporting /routing bfd interface
#error exporting /snmp
#error exporting /tool e-mail
#error exporting /tool mac-server mac-winbox

> /snmp print 

error - contact MikroTik support and send a supout file (2)

I downgraded to 6.19 and everything worked again.... anyone else experience something like this?

try to do a /system check-installation

Could you make a support output file from 6.19 where it works ok and then upgrade to v6.20 where it doesn't work and make another file. Send both files to support@mikrotik.com

2dfx · Sat Oct 11, 2014 12:12 am

Hi!

I have error with lcd max-speed setup via winbox

But i can set max-speed via terminal

After set speed via terminal - winbox show error

sisal · Sat Oct 11, 2014 4:26 pm

I upgraded an RB2011 from 6.12 to 6.20. Now the configuration of networks for DHCP Server in webfig does not work anymore.

I create a network:

ros-620-dhcp-net-1.png

After saving it shows this:

ros-620-dhcp-net-2.png

[admin@MikroTik] /ip dhcp-server network> export
# oct/12/2014 00:20:22 by RouterOS 6.20
# software id = 6GM9-Z8QY
#
/ip dhcp-server network
add address=192.168.33.0/24

I can change the settings on the console, it works. Just missing the working webfig.
sisal

Belgarion186 · Sun Oct 12, 2014 5:04 am

I upgraded an RB2011 from 6.12 to 6.20. Now the configuration of networks for DHCP Server in webfig does not work anymore.

Facing the same problem with Webfig on the RB850Gx2 too (ROS 6.20).

Had to use Winbox to configure DHCP Server - Networks.

napismizpravu · Sun Oct 12, 2014 2:23 pm

... other error RouterOS 6.21rc5 uptime 3d crash! RB433UAH

Dispose of version 6.20 and all 6.21rcX, again from 6.19

When you fix errors with USB Flash (drive) and microSDHC?

<Google Translator>

add error

router crash> power off/on > graf statistic all interfaces wireless lost, not in the menu (connection work
)....over time appeared

rextended · Mon Oct 13, 2014 10:49 am

Again problems with user-manager.

6.18 on x86 with user manager, upgraded to 6.20 and user-manager stop working, not see the files of the database.

Revert back to 6.18 solve the problem.

Mon Oct 13, 2014 11:17 am

Upgrading an SXT G-5HPacd from 6.19 worked but after upgrading
firmware to 3.18.1 device was no longer accessible.
Power Cycle helped.

aliereza · Mon Oct 13, 2014 6:38 pm

Upgraded NetMetal firmware to 3.18.1 and the device crashes every time that I use wireless scanner. Totally inaccessible, and it needs power cycle.

sirdir · Wed Oct 15, 2014 1:54 am

FYI, something seems to be broken with IPSEC. Sometimes the configuration just disappears. If you try to change anything in the submenu then that command just times out.
Rebooting fixes all issues for some time... suddenly the problems come back.

Regards
Patrick

alfonzz · Wed Oct 15, 2014 12:26 pm

Upgraded NetMetal firmware to 3.18.1 and the device crashes every time that I use wireless scanner. Totally inaccessible, and it needs power cycle.

I agree, this hapens from v6.19, not only scan wireless, but snooper too.
in log is system reboot because watchdog timer, if you turned off timer, then system hangs and no response to anything, powercycle helped.

napismizpravu · Wed Oct 15, 2014 1:28 pm

RB433UAH

Proč se objevila ve verzích 6.21rc7 a 6.21rc8 chyba v NTP klient?
Why has appeared in versions 6.21rc7 and 6.21rc8 error in NTP client?

still gain errors ...USB FLASH (drive); microSDHC; NTP, crash

Uvažuji, že se vrátím na verzi 6.19 a nový, ať si zkouší testeři.
I'll go back to version 6.19

Verze 6.19 mi běžela 19 dní a pak jsem jí aktualizoval na 6.20.
Version 6.19 uptime 19 days then update to 6.20.

Za překlad do angličtiny děkuji Google překladači.
The translation into English thanks to Google translator.

scampbell · Wed Oct 15, 2014 11:07 pm

Again problems with user-manager.

6.18 on x86 with user manager, upgraded to 6.20 and user-manager stop working, not see the files of the database.

Revert back to 6.18 solve the problem.

We are also seeing an issue with UM and WG500MP printers interfaced via API. Ticket for 1 Hour or 4 Hour print as 1 or 4 seconds. Printing was correct on RoS 6.17. Avccount still works for 1 or 4 hours though so only a "cosmetic" bug we hope.......

noyo · Thu Oct 16, 2014 12:07 pm

Ticket # 2014091366000293
if you have taken any attempts to remove my problem?
Because the WATCHDOG-REBOOT becoming unbearable.
Unfortunately it has the network RB2011iL which I can not give you downgrade to 5.26

Already three people with support wrote to me and no person has not solved it.

Thu Oct 16, 2014 3:41 pm

RB SXT G-5HPacd still rebooting due to kernel failure.
Same as with 6.19. supout.rif sent long ago and repeated.
No solution from Mikrotik.
FW: 3.18.1

napismizpravu · Fri Oct 17, 2014 1:06 pm

What's new in 6.21rc11 (2014-Oct-17 10:55)
*) disks - fixed support for MMC/SD cards;

Zkoušel jsem a nefunguje, chyba nebyla opravena.
I tried and does not work, the error was not corrected.

upgrade 6.19 > 6.21rc11
rename disk1 >usb1, rename disk2 >usb2 ...reboot
images USB1 alias drive1 error 0 B free capacity, no name
power off/on
unchanged
downgrade 6.19

Došlo ve verzi 6.21rc11 ke zvýšení kapacity disků?
Increased capacity drives in 6.21.rc11? images usb2, microSDHC

<Google Translator>

pampi · Fri Oct 17, 2014 4:41 pm

this "router was rebooted without proper shutdown by watchdog timer" problem solved?

I still have this issue from 6.7 to 6.19 mostly on all RB2011/ RB750G / RB911 / SXT Lite5 devices approx 70 pieces of our devices affected by this error.

Mostly all devices reboot between 1am to 3am. Devices that are affacted are not on the same location, (20-40km) all of them are running through uninterruptible power supply, so it could not be a power problem. At the same time, and on the same tower other devices which are not running Mikrotik 6.x version are still running.

noyo · Fri Oct 17, 2014 4:58 pm

this "router was rebooted without proper shutdown by watchdog timer" problem solved?

I still have this issue from 6.7 to 6.19 mostly on all RB2011/ RB750G / RB911 / SXT Lite5 devices approx 70 pieces of our devices affected by this error.

Mostly all devices reboot between 1am to 3am. Devices that are affacted are not on the same location, (20-40km) all of them are running through uninterruptible power supply, so it could not be a power problem. At the same time, and on the same tower other devices which are not running Mikrotik 6.x version are still running.

I have the same problem
even sometimes when you try to generate or export supout.rif configure itself can perform a watchdog-reboot

This is simply fucking hate.

honzam · Fri Oct 17, 2014 5:52 pm

this "router was rebooted without proper shutdown by watchdog timer" problem solved?

I still have this issue from 6.7 to 6.19 mostly on all RB2011/ RB750G / RB911 / SXT Lite5 devices approx 70 pieces of our devices affected by this error.

Mostly all devices reboot between 1am to 3am. Devices that are affacted are not on the same location, (20-40km) all of them are running through uninterruptible power supply, so it could not be a power problem. At the same time, and on the same tower other devices which are not running Mikrotik 6.x version are still running.
I have the same problem
even sometimes when you try to generate or export supout.rif configure itself can perform a watchdog-reboot

This is simply fucking hate.

+1

Fri Oct 17, 2014 6:11 pm

It happens on 6.x when running on 32mb devices for me. Have you seen it also on other devices?

noyo · Fri Oct 17, 2014 6:22 pm

It happens on 6.x when running on 32mb devices for me. Have you seen it also on other devices?

RB2011 has 60M
RB2011iL has 64M
RB SXT G-5HPnD r2 has 64M
and the same thing happens

Fri Oct 17, 2014 6:38 pm

Have you tried netinstall? I wasn't able to solve this problem only when there was not enough free memory.
I would like to see some effective memory management optimisation in the next ros release as 32mb devices really need it.

noyo · Fri Oct 17, 2014 6:48 pm

Have you tried netinstall? I wasn't able to solve this problem only when there was not enough free memory.
I would like to see some effective memory management optimisation in the next ros release as 32mb devices really need it.

yes, mikrotik suggested execution netinsall, so I made for test on RB2011iL.
Nothing helped. Excluded api, snmp, ssh.
I included rstp across the network. did not help.

Already I wrote with support from S. Martins next Uldis next Janis B. None of these people has not helped.

by the underdeveloped software equivalent of good reviews about the company.

I do not understand how you can spend the AC equipment is not working properly, because the software is not worked out since the release of the first v6

mandrade · Sat Oct 18, 2014 4:07 pm

Any NEWS about Ticket#2014051266001345???

Any news on supporting LACP/802.3ad for trunking on CRS (without taxing the CPU)? We have just activated our 10GE link from our Data Center and the point where our radio trunk arrives in a 6+0 configuration. We're now obligated to keep a V1910 on site just for the aggregation of the radio links. On the remote site we have a similar situation, since our V1910 doesn't support 10GE.

byteman · Sat Oct 18, 2014 4:43 pm

We have issues with EoIPs after upgrade intermediate routers to 6.20 wich are doing routing. The tunnel is not broken, yo can make pings to internet, but no pings more than 1458 size and the clients can not surf to web. Example:

SXT Client (5.26)----wireless link----912G AP(5.26)__ether bridge__|RB935G(6.20)|__ether bridge__Sextant P2P (5.26)----wireless link---Sextant P2P (5.26)__ether bridge__ |CCR Core router 5.26|

with 6.13 version and same configuration in the RB935G router, the problem doesn't exist...

is this problem related with the news features like?:

*) bridge has auto mtu feature
*)...all non-ppp tunnel interfaces got optimized and common set of features..

what can we check?

Thks

steen · Sat Oct 18, 2014 10:32 pm

Hello Folks!

Regarding the routing marks for policy based routing, it seems like that problem is solved now.
I just upgraded from RoS6.7 to RoS6.20 on one of our routers that make use of it.

I am still in doubt, because one of the routes did not work directly, it took about 3 hours, then it started to work by itself by unknown reason. It means in my world it can stop woring at any time to.

I will go on and update one more router and see what happens.
.
.
.
And it was successful to, same effect one route took some time till it started working.

Now remains l2tp vpn tunnels, they were severely broken after 6.7 to, whole our vpn network (which is large) collapsed, a few megabyte could be sent through a link before it went down and took the whole router with it down the drain to.
Eventually it will be tested tomorrow, depends on the business.

thanxes · Sun Oct 19, 2014 12:54 am

Hello, I also have 6to4 tunnel problem. In my case it seems obvious that it is due to missing autogenerated link-local ipv6 address on tunnel interface. In version 6.19 and lower it was there as Dynamic&Link-local fe80::<last 4 bytes of ipv6 net>/64
My ISP (netbox.cz) provides me with only one /64 IPv6 space, therefore I have no address to spare in the tunnel itself when I run ipv6 in autoconfiguration. They suggest this kind of configuration.

noyo · Sun Oct 19, 2014 6:55 pm

Ticket#2014091366000293

In September RB2011iL MTv6.19 was hooked like this:
x86 MTv5.26 (server-pppoe,nat) ---> RB2011LS MTv5.26 ---> UBNT NanoBridge ---> UBNT NanoBridge ---> RB2011iL MTv6.19 ---> ...
and at some time had a watchdog-reboot.
RB2011iL MTv6.19 swapped for RB450GL MTv5.26 because I could not wait until I flee customers by defective v.6.xx

Now the same RB2011iL MTv6.19 strapped so
x86 MTv5.26 (server-pppoe,nat) ---> RB2011LS MTv5.26 ---> RB711G-5HnD MTv5.26 --->
---> RB Lite5 MTv5.26 ---> RB2011L MTv5.26 ---> RB711GA-5HnD MTv5.26 ---> RB SXT G-5HPacD MTv6.20 ---> RB2011iL MTv6.20 ---> ... my house
and the problem disappeared. RB2011iL no longer watchdog-reboot. Uptime 16 day

I also hooked up two more RB2011iL MTv6.20 and they watchdog-reboot
x86 MTv5.26 (server-pppoe,nat) ---> RB2011LS MTv5.26 ---> RB711G-5HnD MTv5.26 --->
---> RB Lite5 MTv5.26 ---> RB2011L MTv5.26 ---> RB711G-5HnD MTv5.26 ---> RB Lite5 MTv5.26 ---> RB2011iL MTv6.20 --->
---> RB Lite5 MTv5.26 ---> RB Lite5 MTv5.26 ---> RB450GL MTv5.26 ---> UBNT NanoBeam ---> UBNT NanoBeam ---> RB2011iL MTv6.20 ---> ...
and all the RB2011iL MTv6.20 is watchdog-reboot.

is all network is bridge

NO RB v5.26 no such problems

What's going on?

Mon Oct 20, 2014 8:07 am

Upgraded a link to 6.20 with both sides doing OSPF/MPLS. After 2 days a IP subnet on one side of the link was no longer reachable.
Routes are there but packets do not pass. Reboot resolved the issue.

6.20 is a release to avoid. Things do not get better they get worse.

skyfx · Mon Oct 20, 2014 3:41 pm

I found this:
in 6.20 does not show the correct values for the data rates and current transmit power

Mon Oct 20, 2014 11:32 pm

What's new in 6.20 (2014-Oct-01 10:06):

*) cert scep - use fingerprints for transaction ids;
*) ipsec - support fqdn as my id;
*) fetch - allow fetching files larger than 4G;
*) fetch - fixed problem where files fetched over https were trimmed in size;
*) fixed problem - it was not possible to see %26 uninstall dude package;
*) stores are replaced with folders and disks are now managed under /disk menu;
*) added support for SMSC750x USB Gigabit Ethernet on x86;
*) ups - support selftest for smart and hid UPS;
*) pppoe client - increase connection timeout to make connection establishment
possible on busy pppoe server;
*) dhcp server - change default lease time from 3 days to 10 minutes
to avoid running out of IPs;
*) ipsec - allow binding modeconf address to username;
*) eoip/eoipv6/gre/gre6/ipip/ipipv6/6to4 tunnels have new features:
auto mtu (enabled by default for new tunnels);
dscp (inherit/specific value, inherit by default for new tunnels);
clamp-tcp-mss (yes by default for new tunnels);
*) eoip/gre/ipip/6to4 tunnels have dont-fragment option (inherit/no, no by default for new tunnels);
*) bridge has auto mtu feature (enabled by default for new bridges);
*) pppoe-server has auto mtu feature (enabled by default for new pppoe servers);

If you already run some RouterOS v6.x version, simply click “Check for updates” in QuickSet, Webfig or Winbox packages menu.

Others: http://www.mikrotik.com/download

Why not wireless AC updates for a better performance?
I would appreciate that very much

zxdpxl · Tue Oct 21, 2014 12:53 pm

Hi
My family has a Mikrotik Router ,The Model is RB951G-2hnd, The current version of the OS is 6.20, My family is connected to internet via China-Telecom ADSL. At my router has a IPv6 tunnel, Hurricane Electric's provided this IPv6 tunnel. China-Telecom will reset the ADSL connection in 48 hours. The WAN IP will changed. I have to change WAN IP the tunnel to on Router setting and web settings in Hurricane Electric's. Then the router will automatically connect to IPv6 tunnel. 6.19 version of the RouterOS can automatically connect to the server. I upgrade the OS version to 6.20. Must manually disable this tunnel, and then enabled this tunnel. IPV6 server to connect. In previous versions have no such problem.

PS:I use this tunnel to access Youtube, Google, Facebook, China Network can't be used to directly access Google, Youtube, facebook.

leonix · Thu Oct 23, 2014 9:39 am

Upgraded a link to 6.20 with both sides doing OSPF/MPLS. After 2 days a IP subnet on one side of the link was no longer reachable.
Routes are there but packets do not pass. Reboot resolved the issue.

We have a similar OSPF problem. E.g. one system 6.19 the other one 6.20; OSPF does not work! (Doesnt find the OSPF router).
After going back to 6.19 - OSPF is working.

Leo.

onnoossendrijver · Thu Oct 23, 2014 10:15 am

I have no problems with OSPF between 6.19 and 6.20 machines. Can you check all MTU settings on both devices? OSPF is very picky about wrong MTU sestings.

nicolasmartinez · Thu Oct 23, 2014 6:24 pm

I can't seems to be able to configure anything on "/ip dhcp-server network" fields like default gateway after applying the new config gets reseted back to the previous values!!

Router model: RB2011iLS-IN

BenCo · Thu Oct 23, 2014 11:51 pm

*) stores are replaced with folders and disks are now managed under /disk menu;

Where can I add a folder instead of a store? Can't seem to find how to do it.
The manual still references the stores.

Stillhard · Sun Oct 26, 2014 1:22 pm

Again problems with user-manager.

6.18 on x86 with user manager, upgraded to 6.20 and user-manager stop working, not see the files of the database.

Revert back to 6.18 solve the problem.

Check my reply here will solve it: http://forum.mikrotik.com/viewtopic.php ... 21#p452421

rextended · Mon Oct 27, 2014 1:06 pm

Again problems with user-manager.

6.18 on x86 with user manager, upgraded to 6.20 and user-manager stop working, not see the files of the database.

Revert back to 6.18 solve the problem.
Check my reply here will solve it: http://forum.mikrotik.com/viewtopic.php ... 21#p452421

Thanks

gustavomam · Tue Oct 28, 2014 4:27 pm

Scampbell it's happen to me too on 6.20.
I was teaching MTCNA and when a student disabled all firewall filter rules it didn't take changes until reboot.

apteixeira · Tue Oct 28, 2014 10:08 pm

We have upgraded several RB951 series routers in a class situation. 2 or 3 of these have all exhibited odd firewall behavior when rules are disabled.

It shows a rule disabled in Winbox yet the rule continues to work, e.g a rule that logs traffic keeps logging even when disabled. We did not test this in CLI as time was tight.

A reboot fixes the issue.

Not all devices have shown this so perhaps user error - has anyone else seen this ?

Hello,

I am experience the same problem. I made a video that shows the bug: http://www.youtube.com/watch?v=2umo6Kvd ... e=youtu.be

Best regards.

yozz · Tue Oct 28, 2014 11:09 pm

please update in RUSSIA NTP client!! our time in moscow http://www.vniiftri.ru/
GMT +3

scampbell · Wed Oct 29, 2014 7:13 am

We have upgraded several RB951 series routers in a class situation. 2 or 3 of these have all exhibited odd firewall behavior when rules are disabled.

It shows a rule disabled in Winbox yet the rule continues to work, e.g a rule that logs traffic keeps logging even when disabled. We did not test this in CLI as time was tight.

A reboot fixes the issue.

Not all devices have shown this so perhaps user error - has anyone else seen this ?
Hello,

I am experience the same problem. I made a video that shows the bug: http://www.youtube.com/watch?v=2umo6Kvd ... e=youtu.be

Best regards.

That confirms that one

coylh · Wed Oct 29, 2014 9:42 pm

I found this:
in 6.20 does not show the correct values for the data rates and current transmit power

I've run into something similar.

honzam · Thu Oct 30, 2014 7:29 pm

6.21 released
http://www.mikrotik.com/download

spacemind · Thu Oct 30, 2014 9:21 pm

Attention:

Mikrotik 6.21 bricked 3 rb951-G after upgrade.

Beep and lights looping.

Mikrotik upgrade failed rb951-G 6.21: http://youtu.be/O0kJacEIdh0

How to fix? RMA?!

Ticket#2014103066001015

Thanks

staslabs · Thu Oct 30, 2014 10:08 pm

Mikrotik 6.21 on CCR1036-12G-4S - so terrible!!!!! Restarts after some minutes of work. Again and again.

heislerb · Thu Oct 30, 2014 10:36 pm

RB750G 6.20 upgraded to 6.21 by copying the npk file to the file list and rebooted.

Could not access RB750G. Downloaded and run netinstall. Was able to access the unit and downgrade to 6.20. All OK on 6.20.

I have upgraded many times form 6.5 and this was a real surprise.

edited to change 450 to 750

heislerb · Fri Oct 31, 2014 1:13 am

RB750G ROS 6.20 upgraded to 6.21 (second try) and it looks good.

This is my test router. I lost a bit of code I was working on before the first try.

I would like to know why things went wrong the first time.

ffernandes · Fri Oct 31, 2014 1:15 am

tenta com netinstall.... se nao der... deita ao lixo

Attention:

Mikrotik 6.21 bricked 3 rb951-G after upgrade.

Beep and lights looping.

Mikrotik upgrade failed rb951-G 6.21: http://youtu.be/O0kJacEIdh0

How to fix? RMA?!

Ticket#2014103066001015

Thanks

spacemind · Fri Oct 31, 2014 2:17 am

tenta com netinstall.... se nao der... deita ao lixo

Attention:

Mikrotik 6.21 bricked 3 rb951-G after upgrade.

Beep and lights looping.

Mikrotik upgrade failed rb951-G 6.21: http://youtu.be/O0kJacEIdh0

How to fix? RMA?!

Ticket#2014103066001015

Thanks

I manage after a few tens of tries with netinstall to roll back to 6.20, but only in 1 951G.

napismizpravu · Fri Oct 31, 2014 4:19 am

RB433UAH RouterOS 6.21

still a problem USB FLASH disk Kingston DataTraveler 2.0 (4GB) from version 6.20rc1 drop compatibility in proxy? FTP transfers on disk function! ( in <6.19 works)

USB2 disk Kingston DataTraveler G3 works

edit:
after several format (ext3 and fat32) and eject the USB flash drive, reboot the routerboard now it works,stores data on disk function > do not rename the disk and use fat32

with a second disc such problems are not!

if a disk and the proxy will make mistakes, so I go back to version 6.19 where it works

<Google Translate>

drusha · Fri Oct 31, 2014 4:44 am

playing with 6.21 on my test x86
noticed a minor winbox bug, while trying to edit Hotspot Server Profile: window default minimal width is bigger than expected and cannot be made smaller

update: few hours later - window appears normally

is it because I've enabled the hotspot?

Fri Oct 31, 2014 11:14 am

MikroTik RouterOS 6.21 is released.
http://forum.mikrotik.com/viewtopic.php?f=2&t=90792