After testing with RB750 and RB3011, the upgrades are ok, but going from 6.39 to 6.40.5 reveals a problem on MLPPP. I’m using 2 DSL modems. If one of the modems loses its physical layer, such as loss of power or loss of ADSL sync, the MLPPPoE bundle gets confused and does not bring the link down to 1 line. If the second link is restored the bundle still stays degraded and doesn’t seem to recognize that the second link is back. Its just packet loss. Resetting both links at the same time seems to be ok as MLPPPoE will properly recover.
Generally I got 4 units: RB3011, RB2011, RB750Gr3 and CRS326 only RB3011 and CRS326 upgraded with no problem. RB2011 in loop (I already did Netinstall and unit works fine (downgraded to 6.38.7 just looking for stable version and on forum I found out that people recommend that version), RB3011 was upgraded with no problem but still somehow it was unstable so downgraded to 6.38.7. CRS326 running with no problem till now but not much happening there, its working as switch with no features. RB750Gr3 is down, well it boot to OS but not showing any OS version so tomorrow I will try first copy file with lower version and if this will not help Netinstall also…so to much trust on MikroTik from my side, next time I will wait at least week to see if anyone will have issues during upgrade.
Some of you can confirm that in NetWatch Interval time does not work properly.
Im using NetWatch to alarm me when of my server down and im using this code.
:local result [/ping 200.200.200.201 count=15]
:global status
:if ($result = 15 && $status != true) do={
tool sms send usb2 phone-number=8888888 message="SERVER UP"
:log info "UP SMS"
:set status true
} else={
:if ($result = 0 && $status != false) do={
tool sms send usb2 phone-number=8888888 message="SERVER DOWN"
:log info "DOWN SMS"
:set status false
}
}
21:58:23 system,info netwatch host modified by admin
21:58:41 script,info UP SMS
21:59:49 system,info netwatch host modified by admin
22:00:08 script,info DOWN SMS
admin@Fa1c0n] /tool netwatch> print
Flags: X - disabled
# HOST TIMEOUT INTERVAL STATUS SINCE
0 200.200.200.200 1s 1m up nov/07/2017 20:45:00
1 200.200.200.201 1s 1m30s up nov/09/2017 21:59:49
2 200.200.200.202 1s 3m up nov/07/2017 20:51:20
You can see into the log that intervel time did`t match from the time where i set in NetWatch
Any idea why that happend
Same here, SMS text garbled on RB411UAHR (factory reset done after upgrade 6.40.5), and also on router upgraded from 6.40.4 to 6.40.5 , Sierra Wireless MC8792V
“Fixed in next version”? Or should i upload supout?
UPDATE: never mind, works again in latest RC
Hello, just FYI, I have upgraded all of my RBs successfully on 1st try
List of RBs:
- RB433AH - previous version was 6.39.1 with factory default firmware
- RB951G-2HnD - previous version was 6.40.4 with 3.41 firmware
- RB2011UiAS-2HnD - previous version was 6.40.4 with 3.41 firmware
Also my report:
CCR1009-1C-7G-1S-1S+ including Dude on SD, 922UAGS-5HPacD (NetMetal5) + R11e-5Hnd, LHG5, 960PGS (hEX POE), OmniTIK UPA-5HnD, 921GS-5HPacD r2 (mANTBox), 951G-2HnD, 450G, 750GL and 911G-5HPnD (QRT5)
all upgraded from 6.40.4 without incidents.
Today I uploaded on all my units via ftp or Winbox and all of them upgraded and working with no problems 
anyway I love those units that there are no problems with recovering them
Units that I upgraded you can find in my sign
Can I report regression present now, but also traced back to earlier version?
Problem:
When using wireless interface in mode=station-pseudobridge-clone, bridged to ether1, infinite loop of "lost connection, found address to clone" is observed since v6.37 to v6.40.5. Naturally, the single device connected to ether1 does not have network connectivity, regardless of whether it tries to obtain address via DHCP, nor is it pingable with a static IP.
Before v6.37, pseudobridge-clone mode works as expected using exactly the same config.
Unfortunately, I cannot use station-wds or station-bridge mode as the other end is not a Mikrotik device, so I need pseudobridge-clone to work.
Hardware tested with:
hAP lite
Versions tested:
6.40.5 (current) - broken
6.40.4 (current) - broken
6.39.3 (bugfix) -broken
6.38.5 (current) - broken
6.37.3 (current) - broken
6.37.0 (current) - broken
6.36.4 (bugfix) - working
6.36.3 (current) - working
6.33.5 (current) - working
I have supout.rif file for 6.36.4, 6.37 and 6.40.5 if support would like to inspect them.
The device is fully Netinstall for each version, using this configure file:
jan/02/1970 00:35:16 by RouterOS 6.33.5
/interface bridge
add name=bridge1
/interface wireless security-profiles
add authentication-types=wpa2-psk eap-methods="" management-protection=allowed
mode=dynamic-keys name=LLP supplicant-identity=""
/interface wireless
set [ find default-name=wlan1 ] band=2ghz-b/g/n disabled=no frequency=auto
mode=station-pseudobridge-clone security-profile=LLP ssid=LLP
wireless-protocol=802.11 wps-mode=disabled
/interface bridge port
add bridge=bridge1 interface=wlan1
add bridge=bridge1 interface=ether1
/ip cloud
set update-time=no
/ip service
set telnet disabled=yes
set ftp disabled=yes
set api disabled=yes
set api-ssl disabled=yesOnly wpa2-pre-shared-key for security-profiles is removed.
deleted
deleted.
I have problem with two rb750GL both of them after upgrade to 6.40.5 are working fine for some time and then just freeze. I see them from other MT in the same bridge but cant log in , then go reboot and work fine for another 12 to 24 hours. Routerboard bios flashed to 3.41 , now I added WD to nearest device but i think that there is something wrong in this situation. Before there was no problem with any of them. Any ideas?
config is really simple one of them is just bridge other one has bridge and 3 vlan’s thats all.
Hello,
I have hAP ac router, after update from 6.38.x to 6.40.5 the signal of 5GHz not stable, I can connect only with 2GHz channel. When I can see the 5GHz signal it no possible connect to Wi-Fi. I didnt do any changes in setups after update, update was automaticaly from “Quick Set” page.
May be I have to check some setups to restore functionality?
I have upgraded to 6.40.5 on (2) DynaDisk 5 and RB2011 also upgraded the firmware. After the upgrade all the RB’s where cutting out. Using btest to any upgraded unit shows good speed, then no speed for several seconds intermittently. After downgrading all the speed test was constantly good. It looks like the firmware upgrade was the issue.
Hi,
I just upgrade my client router hap-lite2 and suddenly the connection lost. oh my god, their office is around 58.4KM away.
i wasn’t give up, tried to mac telnet and it do the trick, i was able to connect to the terminal.
then i could reach the terminal, ports was running (flag R), but the ip was not working. after some check, port name and physical port wasn’t match with the previous condition. why this could be happened?
upgraded from 6.29 to 6.40.5
My similar experience many years ago teached me not to rename the ports. I use comments to mark the ports by their meaning instead and keep the original port names.
Hi jarda,
Yes, agree. the interface name was remains default. ether1, ether2, ether3, ether4 and only commented on ether2 and ether4.
Ahhhh the first victim of the new firmware version notation keeping in touch with the RouterOS version notation. 
try putting a switch between your laptop nic and the mikrotik device, sometimes netinstall does not work directly to the nic, happened to me last week
connect the RouterBOARD device to power with the RESET button pushed, until you see the device in the Netinstall program interface. Then release the RESET button (around 15 seconds since power on).
Updated rb952ui-5ac2nd and https on wifi connection stopped working, even winbox is kicked-out after 2 seconds. But only on wifi (using capsman).
When connected by ether ports, it works just fine, both https and winbox, from wan side is everything accesible also.
Hello, I have the typical set of rules of drop ssh stages that is shown in the Mikrotik wiki, but I notice that in this version 6.40.5 does not detect when the connection is new, therefore it does not work. I had to make the new connection markup from the “mangle” table because in filter I stop working. Is it just my problem? Is it some new feature that I do not know?
Thank you.