v6.10 released

ffernandes · Tue Mar 11, 2014 1:19 am

dunno if anyone else is having the same problem but...
upgraded an rb435g to the 6.11rc1 with the fp wireless driver and its getting
"system rebooted because of kernel failure"
"router was rebooted without proper shutdown!!

file attached!!!

now at 11/03/14 was my rb800 that crashed :X
same error as the 435g

i'm using the 10/03/14 13:36 release....

sasskass · Tue Mar 11, 2014 5:30 am

hello
still ip/route table in snmp (dude) problem not solved!
sorry but we are currently not working on Dude problems, maybe later in the year

hello

does anyone working on igmp snooping feature - this year ,next year...?

Aleksander

Masyanich · Wed Mar 12, 2014 6:31 am

Auto-negotiation is not working properly

problem arose after updating to 6.10
tile

kozik · Wed Mar 12, 2014 10:23 am

There is a bug with dhcp-server on vlan. After update from 5.25 to 6.10 some devolo dlan pro modems cant get their IPs. Had to downgrade.

Same problem on x86 . Dhcp servers on Vlans not working correctly. Clients get Ip but after the lease expired cannot renew lease. Router shows "Server offering lease without success" After the reboot everything fine until next renew. Clear installation, no other configuration , no bridges . 3 vlans under the ethernet interface. The same issue on 6.0 and 6.10. On 5.26 everything works fine.

Jetrider · Wed Mar 12, 2014 11:25 am

There is a bug with dhcp-server on vlan. After update from 5.25 to 6.10 some devolo dlan pro modems cant get their IPs. Had to downgrade.
Same problem on x86 . Dhcp servers on Vlans not working correctly. Clients get Ip but after the lease expired cannot renew lease. Router shows "Server offering lease without success" After the reboot everything fine until next renew. Clear installation, no other configuration , no bridges . 3 vlans under the ethernet interface. The same issue on 6.0 and 6.10. On 5.26 everything works fine.

I can second this. This is OLD bug. v6.5 is fine, versions above it are broken. Talked to support - no avail. ... Achilles glared at him and answered, "Fool, prate not to me about covenants. There can be no covenants between men and lions."

michaelcarey · Wed Mar 12, 2014 11:26 am

Since upgrading to v6.10, my CRS125-24G-1S seems to be acting more like a hub than a switch! My normally quiet home network has become VERY busy!

I had a quick search through this thread and couldn't find anyone with my symptom... sorry if I missed it.

Port 1 is connected to an ADSL router. Port 2 is local master and all other ports are switch chip slaves to Port 2.

Any IP traffic received on any port is sent to ALL other running ports. I'm not sure if this is a bug introduced with v6.10 or something else is screwy with my network.

Here I am transferring a file from one computer to another... while the transfer is occurring, everything else sloooows right down.

Wed Mar 12, 2014 11:52 am

michaelcarey,
Enter this command to fix it on CRS125:

/interface ethernet switch port set [find] learn-restricted-unknown-sa=yes

michaelcarey · Wed Mar 12, 2014 11:59 am

michaelcarey,
Enter this command to fix it on CRS125:
Code: Select all
/interface ethernet switch port set [find] learn-restricted-unknown-sa=yes

Yay !!

Problem solved! Thank you so much!

Is this something that was introduced when I upgraded to v6.10? Will I need to worry about it again?

Michael.

saaremaa · Thu Mar 13, 2014 9:03 am

build (2014-03-12 14:58:04)
*) fixed 100% cpu usage on CCRs;

You can get more information on bug fixes. Under what conditions does it arise?

DrDeft · Thu Mar 13, 2014 3:15 pm

http://forum.mikrotik.com/viewtopic.php?f=1&t=79914

Hi
I have a customers who love your products, are subject to DDoS attacks (SYN flood), and it hurts that Mikrotik doesn't have "notrack" target, just SYN flood over his CCR will knock down CPU to 100%.
And if it had -j NOTRACK (or newer kernels: -j CT --notrack), it can be solved, he needed conntrack only for special case, and cannot turn it off completely.
Please consider adding this option, it should be very trivial to do, and will help a lot of people to solve their issues with conntrack overflow.
If possible take this matter seriously, because the only choice i have to show them how perfect are Mikrotik support, or to explain it is not, and to move them to alternative solution.
Thank you.

Raf · Thu Mar 13, 2014 3:36 pm

build (2014-03-12 14:58:04)
*) fixed 100% cpu usage on CCRs;
You can get more information on bug fixes. Under what conditions does it arise?

+1 to that.

Thu Mar 13, 2014 3:46 pm

build (2014-03-12 14:58:04)
*) fixed 100% cpu usage on CCRs;
You can get more information on bug fixes. Under what conditions does it arise?

Bug fixing doesn't work like that. We can fix the source of the problem, but it will be very hard to list all situations where this bug was showing itself.

spire2z · Fri Mar 14, 2014 12:51 am

Obviously you can't be 100% sure with these things but on two routers within a few hours of upgrade suffered ethernet port flapping. Fixing the port speed from auto-negotiate seems to have solved it so far for a few days.

Fri Mar 14, 2014 2:17 pm

dunno if anyone else is having the same problem but...
upgraded an rb435g to the 6.11rc1 with the fp wireless driver and its getting
"system rebooted because of kernel failure"
"router was rebooted without proper shutdown!!

file attached!!!

now at 11/03/14 was my rb800 that crashed :X
same error as the 435g

i'm using the 10/03/14 13:36 release....

Please upgrade to newest test release of v6.11 and check agin if you still see the kernel panics. If yes, then send that new support output files to support@mikrotik.com

rpingar · Fri Mar 14, 2014 6:48 pm

may we have more explanation about this changelog about 6.11:
*) fixed 100% cpu usage on CCRs;

regards
Ros

patrick7 · Fri Mar 14, 2014 9:18 pm

Hi,

Does anybody here also have the problem that IPv6 addresses sometimes aren't working until disable and re-enable (new added address or after a reboot)? Had that on 4 MikroTiks (RB750GL, RB2011UAS-RM, RB2011UiAS-2HnD-IN, RB951G-2HnD) but unfortunately MikroTik cannot confirm this bug. Very annoying if after a reboot nothing is working.
I'm mostly using IPv6 addresses on bridges.

Edit: Not working means, sometimes I can ping the IPv6 from the local MikroTik cli. But if I try to reach another host in that subnet, traffic goes through the default gateway.

Regards & have a nice weekend,
Patrick

dlj87 · Fri Mar 14, 2014 9:46 pm

Strange high ping fluctuations though OpenVPN tunnels between mikrotik routers (client-server) and mikrotik - centos 6 (client-server) appear in 6.10. This issue makes sip (voice) packets drop and clients are not able to hear each other for a second or two in the moment. Please do something...

Ответ от 192.168.192.10: число байт=32 время=5мс TTL=62
Ответ от 192.168.192.10: число байт=32 время=4мс TTL=62
Ответ от 192.168.192.10: число байт=32 время=4мс TTL=62
Ответ от 192.168.192.10: число байт=32 время=4мс TTL=62
Ответ от 192.168.192.10: число байт=32 время=5мс TTL=62
Ответ от 192.168.192.10: число байт=32 время=4мс TTL=62
Ответ от 192.168.192.10: число байт=32 время=5мс TTL=62
Ответ от 192.168.192.10: число байт=32 время=244мс TTL=62
Ответ от 192.168.192.10: число байт=32 время=4мс TTL=62
Ответ от 192.168.192.10: число байт=32 время=4мс TTL=62
Ответ от 192.168.192.10: число байт=32 время=4мс TTL=62
Ответ от 192.168.192.10: число байт=32 время=4мс TTL=62
Ответ от 192.168.192.10: число байт=32 время=4мс TTL=62
Ответ от 192.168.192.10: число байт=32 время=241мс TTL=62
Ответ от 192.168.192.10: число байт=32 время=4мс TTL=62
Ответ от 192.168.192.10: число байт=32 время=4мс TTL=62
Ответ от 192.168.192.10: число байт=32 время=242мс TTL=62

Shiro · Sun Mar 16, 2014 12:14 am

Hi,

I updated my RB493g from 6.11rc1 to 6.11 package from yesterday (14.03.).

I got the following issue, randomly ppp dies. I have to permanent pppoe dsl session, one permanent l2tp connection and two l2tp road warriors.

After login in with one of my l2tp roadwarriors, all ppp session crashed. I was able to reproduce this error two times. After that, ppp runs fine for around 2 hours.

The only way to fix that hanging ppp is to reboot, i see this message on the rb serial console.

Rebooting...
Stopping services...
failed to stop ppp: std failure: timeout (13)
could not move ram disk: Invalid argument
copying packages to flash...
Restarting system.

I found a autosupout.rif after reboot on the filesystem.

maarisl · Sun Mar 16, 2014 11:07 pm

On v5.x I had queue on subnet 10.1.1.0/24 which limited 3M up and 5M down per each IP and total max rate for subnet to 5M/15M but on v6.10 work only total queue but per IP does not.
Example:
/queue simple add max-limit=5M/15M name=10 priority=3/3 queue=up3/down5 target=10.1.1.0/24 \
total-priority=3 total-queue=default
/queue type add kind=pcq name=up3 pcq-classifier=src-address pcq-dst-address6-mask=64 \
pcq-rate=3M pcq-src-address6-mask=64
/queue type add kind=pcq name=down5 pcq-classifier=dst-address pcq-dst-address6-mask=64 \
pcq-rate=5M pcq-src-address6-mask=64

Please advise what shoud be done different to have limit 3M/5M per each IP of subnet ?

elmer · Mon Mar 17, 2014 12:16 am

RB951Ui-2HnD
First problem is poe firmware.
Second, installed dude 3.6, when i checked uninstall and reboot router dead.
Third - netinstall 6.10 won`t work, 5.26 unbricked router, but poe out don`t works...
After install 6.10 poe came to live, but firmware is still 0.0

PS. LED`s on ports don`t work!

Spaceath · Mon Mar 17, 2014 1:09 am

Im having issues with ntp client.
2 rb433 and 2rb951 doesnt sync the clock time with ntp servers.
anyone else having this problem?

Ansy · Mon Mar 17, 2014 10:00 am

elmer, I've got very same problem with my RB/750 upgrading it to v6.10.

How do your hardware "work" exactly?

My RB/750 turns all LEDs ON (despite of real state of Ethernet ports) and no reaction in WinBox, no traffic at all (after normal reboot) -- I've sniffed any packets with WireShark software.

If I reset device to defaults (short circuit contact hole with screwdriver or 1-2 sec RESET while powering on) hardware lights Ethernet LEDs right according to real port connections -- but no any Ethernet packets too.

Finally I've tried to netinstall some versions 5.xx and 6.xx in hope to reanimate the device. Pressing RESET button untill ACT LED goes off gives device to network (IP, bootp, visible to netinstall) and makes netinstall process possible... but loading any firmware doesn't repair my RB/750 after rebooting

[Ticket#2014031166000717] support says:

unfortunately no, this model does not support it [RS-232]. It may have serial port pads on
pcd, but they won't work without special software on router which we do not share
with public.
...
whole picture looks awflly like problem with nand flash on router, it could have
died of old age so to speek or just hardware failure of this specific chip. In
this case router hardware replacement will be required.

It's strange that two RB751U-2HnD devices have been upgraded to 6.10 successfully and work OK for now.

Please get us to know about progress with yours (and details, please).

P.S. My RB/750 worked not hard in good datacenter conditions for ~ 2 years. Sad

rextended · Mon Mar 17, 2014 10:28 am

Try this:
whit reset BUTTON pressed, (not the hole pin with screwdriver) plug the power, after 2 sec stop press the button.
Wait,
if the rb start, open winbox and update BIOS:
put BOTH this file in "files" folder
http://i.mt.lv/routerboard/files/ar7240_3.13.fwf
http://i.mt.lv/routerboard/files/ar7100_3.10.fwf
and launch

ros code

/system routerboard update

DO NOT REBOOT

After that prepare netistall 6.10 with 6.10 software
http://download2.mikrotik.com/routeros/ ... l-6.10.zip
http://download2.mikrotik.com/routeros/ ... e-6.10.npk
and launch this on console:

ros code

/system routerboard settings
set baud-rate=115200
set boot-delay=2s
set boot-device=try-ethernet-once-then-nand
set boot-protocol=bootp
set cpu-mode=regular
set enable-jumper-reset=yes
set enter-setup-on=delete-key
set force-backup-booter=no
set silent-boot=no

now reboot and use netinstall without keeping old configuration

Ansy · Mon Mar 17, 2014 10:44 am

if the rb start, open winbox and update BIOS:

I'm glad to do it, but:

My RB/750 turns all LEDs ON (despite of real state of Ethernet ports) and no reaction in WinBox, no traffic at all (after normal reboot) -- I've sniffed any packets with WireShark software.

If I reset device to defaults (short circuit contact hole with screwdriver or 1-2 sec RESET while powering on) hardware lights Ethernet LEDs right according to real port connections -- but no any Ethernet packets too.

Simply I can not see device with WinBox (MAC or IP) at all. Only Netinstall sees it somehow.

rextended · Mon Mar 17, 2014 11:00 am

Forget the hole with the screwdriver.

1) before power the rb, press reset button and leave pressed, insert power plug, and wait until devices are visible in netinstall (15~20sec) [assuming the pc is working], now release reset button.
2) If you can see correctly the device on netinstall 6.10, try to install the package 6.10.
3) If netinstall fail, repeat the steps 1-2 with netisntall 5.26 AND RouterOS 5.26.
4) When devices show "waiting reboot" on netisntall, fast remove the power,
5) before power the rb, press reset button and still with reset button pressed, insert power plug, and when any led display up, release reset button [load backup bios].
6) Wait if rb start and follow the previous instruction to upgrade the bios.

chm0d755 · Mon Mar 17, 2014 11:22 am

Delete this post. I found my problem.

Mon Mar 17, 2014 1:09 pm

I have just netinstalled 6.10 on RB2011UAS-2HnD and upgraded FW from 3.10 to 3.12. I was not able to run config script downloaded from v.6.9. so I did it by copy-paste thru terminal.

Reason: The Netwatch is missing in Tools.

How to get Netwatch back? I need it.

Mon Mar 17, 2014 1:12 pm

I have just netinstalled 6.10 on RB2011UAS-2HnD and upgraded FW from 3.10 to 3.12. I was not able to run config script downloaded from v.6.9. so I did it by copy-paste thru terminal.

Reason: The Netwatch is missing in Tools.

How to get Netwatch back? I need it.

I am a dumb. Advanced tools package is missing... sorry everyone for disturbing.

Ansy · Mon Mar 17, 2014 1:39 pm

Forget the hole with the screwdriver.
1) before power the rb, press reset button and leave pressed, insert power plug, and wait until devices are visible in netinstall (15~20sec) [assuming the pc is working], now release reset button.
2) If you can see correctly the device on netinstall 6.10, try to install the package 6.10.
3) If netinstall fail, repeat the steps 1-2 with netisntall 5.26 AND RouterOS 5.26.
4) When devices show "waiting reboot" on netisntall, fast remove the power,
5) before power the rb, press reset button and still with reset button pressed, insert power plug, and when any led display up, release reset button [load backup bios].
6) Wait if rb start and follow the previous instruction to upgrade the bios.

rextended, made all your checklist exactly (thanks for trick #4) twice with netinstall6.10+routeros6.10 & netinstall5.26+routeros5.26, but thats' doesn't help to repair my device.

After all, I can not see its MAC with WinBox in default boot mode (RESET button shortly pressed while power on) and free reboot.

rextended · Mon Mar 17, 2014 1:48 pm

Made one last try with [use SAME version of Netinstall and RouterOS!]
http://download.mikrotikindonesia.com/i ... l-5.11.zip
http://download.mikrotikindonesia.com/i ... e-5.11.npk

Yes, that version!...

I have the same problem, when I forget to upgrade BIOS [firmware] before upgrading some SXT to version 5.26 or 6.x
But after I successfully boot RB with old software [or backup BIOS] I can fix with the method described before.

If you reach to reboot properly the board, first do bios upgrade as I explain on previous post.

m3gaman · Mon Mar 17, 2014 6:25 pm

Hi, I have a problem using option DHCPv6-PD to send ipv6 prefixes via pppoe, the option send the prefix and create the route, works well and hangs dhcp when pppoe conection drop.
If I disconect normaly them dhcpv6 release the prefix but when hangs suddenly it simply stay conected and dhcpv4 stop too, cant get new prefixes or release the one with pppoe. When that happens one core of my cpu goes 100% and only a reboot normalize it.

Majklik · Mon Mar 17, 2014 10:55 pm

Does anybody here also have the problem that IPv6 addresses sometimes aren't working until disable and re-enable (new added address or after a reboot)? Had that on 4 MikroTiks (RB750GL, RB2011UAS-RM, RB2011UiAS-2HnD-IN, RB951G-2HnD) but unfortunately MikroTik cannot confirm this bug. Very annoying if after a reboot nothing is working.
I'm mostly using IPv6 addresses on bridges.

Edit: Not working means, sometimes I can ping the IPv6 from the local MikroTik cli. But if I try to reach another host in that subnet, traffic goes through the default gateway.

Yes, I have the same problem long time. Mostly on the bridge interface or on the VRRP interface. After rebooting the IPv6 do not works on the interface.
bridge - The bridge do not have a link local address after reboot sometimes. There is in the /ipv6 address listed correct link local address but interface is (unknown). After disable/enable bridge then is created new record with correct link local address and correct interface and IPv6 works. All bridges have admin MAC set.
VRRP - After rebooting the vrrp interface which starts as master sometimes do not operate with IPv6 (link local address is correctly assigned). After disable/enable or switch slave/master state IPv6 starts working.
These problems begin with ROS6.3 and ROS5.25 on RB1100AH/AHx2. Reported long time ago, never answered.

The bridge problem after resolving with disable/enable looks like this:

/ipv6 address print where address="fe80::ff:fe00:12/64"
Flags: X - disabled, I - invalid, D - dynamic, G - global, L - link-local 
 #    ADDRESS                                     FROM-POOL INTERFACE               ADVERTISE
20 DL fe80::ff:fe00:12/64                                   (unknown)               no       
21 DL fe80::ff:fe00:12/64                                   bridge-wan              no

Ansy · Tue Mar 18, 2014 6:37 am

Made one last try with [use SAME version of Netinstall and RouterOS!]
http://download.mikrotikindonesia.com/i ... l-5.11.zip
http://download.mikrotikindonesia.com/i ... e-5.11.npk
Yes, that version!...

Done.

RB750_netinstall5.11_ros5.11_before.png

But after reboot (fast power off when "Waiting...", then power on with RESET button pressed till LEDs on, then power off/on reboot) I see no my device RB/750 in WinBox

WinBox5.11.png

So I can't do next part of firmware/bios upgrade procedure. All LEDs on, no any network activity/visibility

I have the same problem, when I forget to upgrade BIOS [firmware] before upgrading some SXT to version 5.26 or 6.x
But after I successfully boot RB with old software [or backup BIOS] I can fix with the method described before.
If you reach to reboot properly the board, first do bios upgrade as I explain on previous post.

I'm very glad if your receipt helps you or somebody else in same trouble... but not me, sorry.

For now I've got LEDs box for Christmas Tree and spare 12V Power Supply

patrick7 · Tue Mar 18, 2014 8:38 pm

Thanks @Majklik, opened a ticket again

steen · Tue Mar 18, 2014 10:34 pm

Hello Folks!

Upgraded rb2011 and rb750 to RoS6.10 from RoS6.7, all l2tp links fails more or less, nothing in logs they simply stop working after some megabytes of data traversing them, average is 2-3 Mbyte, then dead.

This is VERY serious problem for our business, economic loss is result this time, many co-workers just sit doing nothing!!!

The same goes for routing marks, they are also ignored after upgrade.

Now trying to rollback.

Majklik · Wed Mar 19, 2014 12:07 am

Hi, I have a problem using option DHCPv6-PD to send ipv6 prefixes via pppoe, the option send the prefix and create the route, works well and hangs dhcp when pppoe conection drop.
If I disconect normaly them dhcpv6 release the prefix but when hangs suddenly it simply stay conected and dhcpv4 stop too, cant get new prefixes or release the one with pppoe. When that happens one core of my cpu goes 100% and only a reboot normalize it.

Yes, I agree. I see this from ROS6.7 days (at least on the RB800). A profile show that a dhcp task consumes my CPU.

3bs · Wed Mar 19, 2014 7:00 am

v6.10 every hour disconnect openvpn.

Chupaka · Wed Mar 19, 2014 10:03 pm

v6.10 every hour disconnect openvpn.

probably, it's already fixed in pre-release:

What's new in 6.11 (2014-Mar-18 11:14):

*) ovpn - fixed TLS renegotiation;

rextended · Wed Mar 19, 2014 10:38 pm

Actually are another bug, also present on 6.8, 6.9 and 6.10
and go worst on 6.11:

Winbox connection from pc directly connected on the gateway/pppoe server [really not matter where pc are connected]
to one CPE on IP obtained from pppoe with MRRU set,
is continually broken after receiving some data from CPE.

The problem regard exclusively Winbox service on RouterOS, bandwidth and other parameters working perfectly with very low latency.

On 6.7 is perfectly stable.

The problem is not present if one ip for wlan1 are used.

JanezFord · Thu Mar 20, 2014 9:54 am

I believe this issue is somehow v6.10 related .. I did not experience this kind of problems with older versions of routeros.

http://forum.mikrotik.com/viewtopic.php?f=3&t=83030

JF.

FlySt0nE · Thu Mar 20, 2014 9:55 am

951 and 751, after upgrade on 6.10 don't work ntp client.

Thu Mar 20, 2014 11:04 am

951 and 751, after upgrade on 6.10 don't work ntp client.

NTP client works fine after upgrade. Please check if the NTP server IP isn't changed.

FlySt0nE · Thu Mar 20, 2014 11:19 am

NTP client works fine after upgrade. Please check if the NTP server IP isn't changed.

IP of NTP server didn't changed, and after i found problem, i tried change IP to new. In 6.07 work fine.
We have 2 of 751 and 3 of 951, every have problem with NTP Client.

FlySt0nE · Thu Mar 20, 2014 11:23 am

Im having issues with ntp client.
2 rb433 and 2rb951 doesnt sync the clock time with ntp servers.
anyone else having this problem?

Another user has the same problems like me.

Majklik · Thu Mar 20, 2014 11:59 am

I believe this issue is somehow v6.10 related .. I did not experience this kind of problems with older versions of routeros.

http://forum.mikrotik.com/viewtopic.php?f=3&t=83030

This problem with full IPv4 route cache I see for whole ROS6 line. With the ROS 6.10 is only more fastelly cache filled.

951 and 751, after upgrade on 6.10 don't work ntp client.

On few RB912 I had this problem directly after upgrade too. But after second reboot SNTP client opetares OK.

skibi82 · Thu Mar 20, 2014 1:31 pm

Above version 6.7 I noticed a strange phenomenon.
Well, in the case when I have defined DSTNAT redirection inside the network.
And leaning forward Addres Lists After some time available services stop working.
It helps to reboot mikrotik and everything returns to normal.

It seems that this phenomenon is related to Ticket # 2014031066000782

My next observation is a problem with the DNS service
Submitted as Ticket # 2014031966000417

Well dns cache is not properly refresh, ignoring ttl times in the case of kiedi change the static dns entry on the aaaa bbbb is still visible in the cache address aaaa if it was used and can not do anything about it.

x1.png

Flusch d'ont help.
Wrong address is still in the cache given.
Computers using DNS cache get an incorrect address.

Temporary solution
before the change simply disable adrres.
Then change the content and turn it on.

How to hang an address in the cache. Please add it again under a name then disable and remove.

Please someone from the support looked at the problem.
Problems are critical to the sustainability of the action of ROS

FlySt0nE · Thu Mar 20, 2014 6:58 pm

Just downgrade to 6.07 two of devices, ntp client working fine.

rextended · Thu Mar 20, 2014 9:29 pm

Just downgrade to 6.07 two of devices, ntp client working fine.

Tomorrow exit 6.11

armandfumal · Fri Mar 21, 2014 8:45 am

NTP client works fine after upgrade. Please check if the NTP server IP isn't changed.
IP of NTP server didn't changed, and after i found problem, i tried change IP to new. In 6.07 work fine.
We have 2 of 751 and 3 of 951, every have problem with NTP Client.

I'm using 6.10 on RB951, no NTP problem...

ros code

ros code

Who is online