*) firewall - added "ein-snat" and "ein-dnat" connection NAT state matchers for filter and mangle rules
ovpn - added "tls-auth" option support for imported .ovpn profiles;
mpls - added option to match and set MPLS EXP with bridge and mangle rules;
Any info/docs for this? We want to test as soon as possible.*) mpls - added option to match and set MPLS EXP with bridge and mangle rules;
ovpn - added "tls-auth" option support for imported .ovpn profiles;
@Larsa: Endpoint-Independent NAT: https://help.mikrotik.com/docs/pages/vi ... pendentNAT
Thanks. Been waiting for long.ssh - added support for user ed25519 public keys;
Ask and ye shall receive...*) console - added "terminal/ask" command;
:put [/terminal/ask "Is there going to be BTH for xMIPSx?"]
A protocol from 1992. Why not change to OSPFNo IS-IS, no 6VPE. The wait continues...
In the build it's :tonsec NOT ":tosec", but seems to work:*) console - improved ":totime" and ":tonum" commands and added ":tosec" command for time value manipulation;
:put [:tonsec [:timestamp]]
1692285621621769359
I have no clue about.*) console - added ":jobname" command;
ROS v7.11, upon which this beta is based, came with "improved fan control on CRS3xx" ... so do verify fan speed. With RJ45 SFP modules it's critical to maintain adequate air flow to ensure cooling, so in your case no fan slowdowns should be accepted. If needed, set minimum fan speed to (fairly) high value to maintain SFP temperature below critical thresholds.after the update from 7.11 stable to 7.12Beta1 my SFP+ module S+RJ10 reports high temperature and is temporary disabled.
Thanks for the tip , but the CRS326-24G-2S+ doesn't have any fans, it's fanless.Due to excessive heat, produced by RJ45 SFP modules, Mikrotik issued S+RJ10 general guidance, which specifies recomended placement of such modules.
*) ssh - added support for user ed25519 public keys
Does this meaningless change line refer to the problem introduced late in the 7.11 release (VLAN-filtering bridge problems with 2 switch chips)??*) bridge - improved system stability;
I prefer mature protocols :)
A protocol from 1992. Why not change to OSPF
They already answered this question higher in the thread. This isn't really a new feature. They've just restored a missing feature that was there in RouterOS v6 but was missing from v7. The behaviour is now the same as it was in v6. The description made me excited thinking it included some new features we didn't have before, instead it is the bug fix that I already knew was coming because MikroTik sent me an automated email letting me know the bug was fixed.*) mpls - added option to match and set MPLS EXP with bridge and mangle rules; - need more info pls
If you have 2 Mikrotiks on each end, it is possible using EOIP on top of the link.How long are my 10+ months old hAP ax2s going to be stored in a drawer, because ROS 7 still does not support repeater mode? Tonnes of new features, but still not fully on parity with ROS 6.
What I have found was not EOIP solution, but creating a VXLAN interface on both sides, adding those to local bridges, doing few twists to filtering, etc. Will experiment with that - viewtopic.php?t=180369#p990815If you have 2 Mikrotiks on each end, it is possible using EOIP on top of the link.How long are my 10+ months old hAP ax2s going to be stored in a drawer, because ROS 7 still does not support repeater mode? Tonnes of new features, but still not fully on parity with ROS 6.
Workaround ? Yes.
But a solution.
Search forum for post from, I think, Amm0 who describes the steps.
Why they did not name it just :toepoch then if it is sec or nsec would be covered.In the build it's :tonsec NOT ":tosec", but seems to work:Code: Select all:put [:tonsec [:timestamp]] 1692285621621
I believe there is much more in "full-cone" definition then just Endpoint-Independent NAT.Endpoint-Independent NAT
all world call it Full Cone NAT
Well, VXLAN and EOIP are different ways to achieve the same thing: packaging a raw ethernet frame into an IP packet, so you can transfer it over the nontransparent WiFi link and still maintain full functionality like broadcast and VLANs at ethernet level.What I have found was not EOIP solution, but creating a VXLAN interface on both sides, adding those to local bridges, doing few twists to filtering, etc.
According to original authoer, EOIP required change to MTU, which transport over wifiwave2 did not allow. I am not an expert on that, in fact EOIP solution came to my mind earlier too, just was lazy to try it out. Might experiment with both, but still wondering, how difficult it might be to add 4 address frame support, especially if it was part of ROS already (although a different wireless stack).Well, VXLAN and EOIP are different ways to achieve the same thing: packaging a raw ethernet frame into an IP packet, so you can transfer it over the nontransparent WiFi link and still maintain full functionality like broadcast and VLANs at ethernet level.What I have found was not EOIP solution, but creating a VXLAN interface on both sides, adding those to local bridges, doing few twists to filtering, etc.
Only VXLAN is even more inefficient than EOIP (in terms of header overhead).
I guess not so easy ... it requires tinkering with (almost) raw wireless frames ... which normally does wireless driver (and MT is using "stock" wireless driver in wifiwave2). If they went with chipset vendor's reference driver because they don't want to tinker with wireless drivers any more, then implementing 4-address mode would effectively revert their decision.... still wondering, how difficult it might be to add 4 address frame support.
The current name ":tonsec" is "number of seconds since epoch" – so they could have pick any part of that ;)Why they did not name it just :toepoch then if it is sec or nsec would be covered.
:put [:totime 60]
00:01:00
:put ([:totime "2023-08-18"] - [:timestamp])
-13:46:04.114613691
:put [:totime "2023-08-18T00:00:00"]
# blank / nothing is output with a T, but a space works...
:put [:totime "2023-08-18 00:00:00"]
2798w1d00:00:00
That why I'm not sure using a tunnel isn't such a bad option. It does let you treat wireless same as wired, vs using Wi-Fi specific WDS-like things.4-address mode is not standard in 801.11. each manufacturer that offers it has implemented their own hacks to negotiate and support it,
*) firewall - added "ein-snat" and "ein-dnat" connection NAT state matchers for filter and mangle rules;
No, you can set a higher MTU on a wireless interface (limit in old wireless is 1600, don't know what it is in wifiwave2) to have some headroom to allow 1500 byte MTU for the tunneled traffic...Interesting. But you have 1500 MTU, exactly, today...so even 16 bytes is too many ;)
We do not use that. We have use-bfd=yes on BGP peers (that are over wireless). That would still work with a tunnel over wireless or wifiwave2.I'm surprised you didn't mention the lack of BFD on a static /ip/route via check-gateway=bfd on a wifiwave2/etc interface. ;)
Oh you can set MTU no problem on wifiwave2, that ain't the problem. It does nothing if you look in sniffer, still 1500. Regardless if you set tunnel and interface higher. The wifiwave2 maxes at 1500.No, you can set a higher MTU on a wireless interface (limit in old wireless is 1600, don't know what it is in wifiwave2) to have some headroom to allow 1500 byte MTU for the tunneled traffic...
Me too.. All the wifiwave2 things are in the field right now. I got a RB1100AHx4, wAPacRs and CHRs right now, and all work with the v7.12beta ;)Ok I cannot test that here because I have no device that realistically can be used with wifiwave2...
My pet peeve is that my RB4011iGS+5HacQ2HnD, the flagship home router, which has a supported 5GHz radio and all the storage and RAM required for wifiwave2, cannot use wifiwave2 because installing that driver stupidly disables the wireless driver so the 2GHz radio isn't recognized anymore...Me too.. All the wifiwave2 things are in the field right now. I got a RB1100AHx4, wAPacRs and CHRs right now, and all work with the v7.12beta ;)Ok I cannot test that here because I have no device that realistically can be used with wifiwave2...
Actually Cisco calls it Restricted Cone NAT, but for example Juniper, Fortinet and many BSD variants call it Endpoint Independent mapping NAT...Endpoint-Independent NAT
all world call it Full Cone NAT
ovpn - added "tls-auth" option support for imported .ovpn profiles
Are you sure?! My doubtsHow long are my 10+ months old hAP ax2s going to be stored in a drawer, because ROS 7 still does not support repeater mode? Tonnes of new features, but still not fully on parity with ROS 6.
You have posted this to support@mikrotik.com and got a sup number, if not, nothing will happen.viewtopic.php?t=172400 please fix this.
Yes, VXLAN is the way I work around this at the moment, and it is working very well for me. I would recommend that approach until such a time as they have real four-address-mode support.What I have found was not EOIP solution, but creating a VXLAN interface on both sides, adding those to local bridges, doing few twists to filtering, etc. Will experiment with that - viewtopic.php?t=180369#p990815
*) console - improved ":totime" and ":tonum" commands and added ":tosec" command for time value manipulation;
For remarks that broad you should at least reference a SUP number.MT, are you going to fix WiFi on x86 platform ? It was broken since ROS 7.7
No problem. That's it: SUP-104565For remarks that broad you should at least reference a SUP number.MT, are you going to fix WiFi on x86 platform ? It was broken since ROS 7.7
Interface - added "macvlan" interface support;
Probably did not pass internal quality testing...The new release shouldn't be 7.12beta2?
Please push a liitle life into these dry facts*) switch - improved switch chip stability for CCR2004-16g-2s+ devices; what caused problem with ..... and there one or two sentences of the crucial problem
Like the approach.*) console - added "transform" property for ":convert" command;
:put [:convert [:convert "long live the emperor" transform=rot13 to=hex] transform=rot13 from=hex]
# long live the emperor
Is that on a physical console? (serial port and terminal program)console died so far i can reproduce this on a spare CCR1036 and CRS317 so this is not architecture specific
Terminal inside winbox, prior to upgrade both my 1036 and 317 devices is from v7.11Is that on a physical console? (serial port and terminal program)console died so far i can reproduce this on a spare CCR1036 and CRS317 so this is not architecture specific
As I cannot reproduce that on a terminal window...
I believe there is a mistake "port - add support for Huawei MS237h-517" it should be MS2372h-517RouterOS version 7.12beta has been released on the "v7 testing" channel!
Before an upgrade:
1) Remember to make backup/export files before an upgrade and save them on another storage device;
2) Make sure the device will not lose power during upgrade process;
3) Device has enough free storage space for all RouterOS packages to be downloaded.
What's new in 7.12beta3 (2023-Aug-24 12:15):
!) ethernet - changed "advertise" and "speed" arguments, and removed "half-duplex" setting under "/interface ethernet" menu;
!) sfp - convert configuration to support new link modes for SFP and QSFP type of interfaces;
*) bgp - fixed "atomic-aggregate" always set in output;
*) bgp - fixed local and remote port settings for BGP connections;
*) bgp - increase "hold-time" limit to 65000;
*) bridge - fixed fast-path forwarding with HW offloaded vlan-filtering (introduced in v7.11);
*) bridge - fixed untagged VLAN entry disable;
*) bridge - fixed vlan-filtering stability with HW and non-HW offloaded ports (introduced in v7.10);
*) bridge - improved vlan-filtering bridge stability with CAPsMAN (introduced in v7.11);
*) bth - added "Back To Home" VPN service for ARM, ARM64, and TILE devices;
*) calea - improved system stability when trying to add rules without the CALEA package;
*) console - added "transform" property for ":convert" command;
*) console - fixed scheduler "on-event" script highlighting when editing;
*) console - improved multi-argument property parsing into array;
*) console - improved stability when editing long scripts;
*) console - show full date and time in scheduler "next-run" property;
*) dhcp - fixed DHCP server and relay related response delays;
*) ethernet - added "supported" and "sfp-supported" values for "monitor" command;
*) interface - added "macvlan" interface support;
*) ipsec - fixed IPSec policy when using modp3072;
*) ipv6 - fixed IPv6 RA delay time from 5s to 500ms according to RFC;
*) ipv6 - send RA and RA deprecate messages out three times instead of just once;
*) log - improved logging for user actions;
*) lte - added at-chat support and increased wait time on modem at-chat for Dell DW5821e, DW5821e-eSIM, DW5829e and DW5829e-eSIM;
*) lte - added SINR reporting for FG621-EA modem;
*) lte - fixed Sierra modem detection for modems with vendor-specific USB descriptors;
*) lte - fixed startup race condition when SIM card is in non-default slot for LtAP mini;
*) netinstall-cli - prioritise interface option over address option;
*) ospf - fixed adding ECMP routes;
*) ospf - fixed OSPFv3 not working with NSSA areas;
*) ospf - fixed parsing of opaque LSAs used by TE;
*) ospf - fixed translated NSSA routes not showing in backbone;
*) port - add support for Huawei MS237h-517;
*) port - expose NMEA/DIAG ports for Dell DW5821e and DW5821e-eSIM;
*) quickset - fixed "LAN" interface list members if configuration does not contain bridge;
*) rip - added BFD support;
*) rip - fixed session not working in VRF;
*) route - fixed gateway after link restart;
*) route - removed deprecated "received-from" property;
*) sfp - improved interface stability for SFP and QSFP types of interfaces;
*) switch - improved switch chip stability for CCR2004-16g-2s+ devices;
*) tile - improved system stability when using queues;
*) traffic-generator - added "priority" property for "inject" command;
*) wifiwave2 - added comment property for registration-table;
*) wifiwave2 - enable changing interface MTU and L2MTU;
*) wifiwave2 - fixed malformed Interworking packet elements;
*) winbox - allow to set multiple addresses and added IPv6 support under "Interface/VETH" menu;
*) wireguard - added "wg-add-client" configuration wizard (CLI only);
*) wireguard - added "wg-export" and "wg-import" functionality (CLI only);
*) wireless - fixed malformed Interworking packet elements;
*) x86 - added support for Mellanox ConnectX-6 Dx NIC;
What's new in 7.12beta1 (2023-Aug-15 16:14):
*) bgp - fixed typos and missing spaces in log messages;
*) bridge - improved system stability;
*) bth - added "Back To Home" VPN service for ARM, ARM64, and TILE devices;
*) certificate - allow to get and maintain Let's Encrypt certificate in IPv6 environment;
*) certificate - fixed "subject-alt-name" duplicating itself when SCEP is used;
*) certificate - improved certificate validation logging error messages;
*) certificate - log CRL HTTP errors under the "error" logging topic;
*) chr - increased OVA default RAM amount from 160MB to 256MB;
*) console - added ":jobname" command;
*) console - added "as-string" and "as-string-value" properties for "get" command;
*) console - added "terminal/ask" command;
*) console - improved ":totime" and ":tonum" commands and added ":tonsec" command for time value manipulation;
*) console - improved stability and responsiveness;
*) console - improved stability when using "special-login";
*) firewall - added "ein-snat" and "ein-dnat" connection NAT state matchers for filter and mangle rules;
*) ike1 - log an error when non-RSA keys are being used;
*) iot - fixed an issue where applying a script to GPIO pin caused GPIO to stop working;
*) iot - fixed behavior where GPIO output state would change on boot;
*) lte - fixed Sierra modem initialization;
*) lte - use more compact logging messages;
*) modbus - added additional security settings for Modbus TCP;
*) mpls - added option to match and set MPLS EXP with bridge and mangle rules;
*) mpls - fixed "propagate-ttl=no" setting;
*) netinstall - added option to discard branding package;
*) ospf - fixed BFD on virtual-link with configured VRF;
*) ovpn - added "tls-auth" option support for imported .ovpn profiles;
*) sfp - fixed missing "rx-power" monitor with certain modules (introduced in v7.10);
*) ssh - added support for user ed25519 public keys;
*) ssh - allow to specify key owner on import;
*) ssh - fixed SSH tunnel performance (introduced in v7.10);
*) supout - added LLDP power to supout.rif;
*) supout - fixed BFD section;
*) system - improved system stability when MD5 checksums are used;
*) tile - improved system stability when using IPv6 queues;
*) wifiwave2 - list APs with a higher maximum data rate as more preferable roaming candidates;
*) winbox - allow to change port numbers for SCTP, DCCP, and UDP-LITE protocols under "IP/Firewall" menus;
To upgrade, click "Check for updates" at /system package in your RouterOS configuration interface, or head to our download page: http://www.mikrotik.com/download
If you experience version related issues, then please send supout file from your router to support@mikrotik.com. File must be generated while a router is not working as suspected or after some problem has appeared on the device
Please keep this forum topic strictly related to this particular RouterOS release.
Changelog edit:
- changed from ":tosec" to ":tonsec".
That's handy. Tiny/minor inconsistencies:*) wireguard - added "wg-add-client" configuration wizard (CLI only);
*) wireguard - added "wg-export" and "wg-import" functionality (CLI only);
maybe because is-is performes a bit betterA protocol from 1992. Why not change to OSPFNo IS-IS, no 6VPE. The wait continues...
Betamax was better than VHS ;)maybe because is-is performans is a bit better
No it wasn't, see https://www.youtube.com/watch?v=hGVVAQVdEOs ;)Betamax was better than VHS ;)maybe because is-is performans is a bit better
*) console - added "transform" property for ":convert" command;
It would be desirable if mikrotik first restored the same range of functions as in ros 6 and only then implemented new features. I'm mainly waiting for MPLS in HW at least at the same level as in ROS6.I think We can expect IS-IS in Mikrotik Very Soon!!!
ISIS.png
Happy to Fast Developments from Mikrotik Team............
I'm looking for MD5 to simplify some scripts.
:put [:convert "text to hash to MD5" transform=md5 to=hex]
What MPLS Hardware functionality was in RouterOS v6 that is not in v7 ?
It would be desirable if mikrotik first restored the same range of functions as in ros 6 and only then implemented new features. I'm mainly waiting for MPLS in HW at least at the same level as in ROS6.
yes. that is a welcome addition - hopefully this will come to winbox (and maybe webgui) too*) wireguard - added "wg-add-client" configuration wizard (CLI only);
*) wireguard - added "wg-export" and "wg-import" functionality (CLI only);
Thank you very much.
..with support for QR-codes like BTH has.yes. that is a welcome addition - hopefully this will come to winbox (and maybe webgui) too
If I recall correctly the CRS317 supported MPLS hardware offload when acting as a P router in RouterOS v6. I'm not sure if this has been brought to v7 yet.What MPLS Hardware functionality was in RouterOS v6 that is not in v7 ?
that's exactly what I meant. We have decided to exclusively use CRS504 and CRS518 as P-Routers in all new networks (we have also already started to convert old ones). Then the first 10 PoPs were ready and should go into operation and the feature was removed. Currently we have 23 PoPs equipped with the systems (CCR2004 working as PE routers in front of the OLTs) and we are waiting to get started with MPLS. At the moment we switch the whole thing in HW. Against this background, it is a pity that now new features like IS-IS are being implemented although the full range of functions of ros 6 is not yet available again.If I recall correctly the CRS317 supported MPLS hardware offload when acting as a P router in RouterOS v6. I'm not sure if this has been brought to v7 yet.What MPLS Hardware functionality was in RouterOS v6 that is not in v7 ?
This apparently is rocket science! MikroTik simply cannot get it working right.Watch out with 7.12beta3 if you have 100G ports on a CCR2216 and you use QSFP28 which are attached to a cable (DAC or AOC or so).
Warning:
Watch out with 7.12beta3 if you have 100G ports on a CCR2216 and you use QSFP28 which are attached to a cable (DAC or AOC or so).
We had several links going out of service after upgrading to 7.12beta3. Going back to 7.11 restored the ports. Other nodes who had a optical LR4 transceiver in them did not show this effect.
*) wifiwave2 - added comment property for registration-table;
A way to do it via ACL works in 7.11(wifiwave2) and worked in 6.x.Edit: OK, managed to do it via an access list, setting a simple rule for the particular MAC address. But then - I miss the context menu action "Copy to access list", while at the registration table tab.
the weird thing is that also the mikrotik qsfp28-sr4 module works on 7.11 and doesn't on 7.12beta..............Miktotik seems to have a special "Testing department" for this and they update the lists on a regular basis as you can see in the Docs history:
https://help.mikrotik.com/docs/pages/vi ... d=13500447
Sadly the SFP sections are not recently being tested or updated.
In the Wiki you can see QSFP...why that is not transferred to the Docs is not know
https://wiki.mikrotik.com/wiki/MikroTik ... patibility
The registration table did display the comment, but but you could not filter (print where ...) or access via scripting (find, get, ...).A way to do it via ACL works in 7.11(wifiwave2) and worked in 6.x.Edit: OK, managed to do it via an access list, setting a simple rule for the particular MAC address. But then - I miss the context menu action "Copy to access list", while at the registration table tab.
netinstall solved the terminal issueThis is how it look like, same for 1036,RB4011 i'll try to netinstall them later if i can reproduce the issue
2.PNG
> :local addr [/terminal/ask value-name="Enter IP address: "]; :put "Your IP address is $addr."
Enter IP address: 10.25.0.13
Your IP address is 10.25.0.13.
alpha137 doesn't fix the qsfp issue.@rpingar, afink - thanks for the feedback. Can you also share the output from "/interface ethernet monitor" for thouse QSFP modules that do not work?
I have found the root cause of this behavior on 7.12beta or 7.12alpha.@rpingar, afink - thanks for the feedback. Can you also share the output from "/interface ethernet monitor" for thouse QSFP modules that do not work?
check if fec are disabled on both side.they are enabled on my case
try to disable auto negotiation and fix the proper speed/mux at least on ccr2216fec is configured on switch and mikrotik using fec92 / rs . this is mandatory for link to come up in 7.11. Auto negotiation failed otherwise.
welcome to the 100g qsfp world!! :DDDDDDDDDDDDDDDDDSwitching off auto negotiation and configuring it for 100G-baseLR4-ER4 brought the interface up.
Interesting...
It may let you set that... But still limited by 802.11 specs, so 2290 is max L2MTU.*) wifiwave2 - enable changing interface MTU and L2MTU;
Now if you can only adjust the wireless MTU to 9000+ Bytes for bridging l2 networks for jumbo frame support(e.g MEF 3 carrier grade connections ) in ptp wireless setups :)
It's in nanoseconds. So the "n" is nano, not number, of seconds.Why they did not name it just :toepoch then if it is sec or nsec would be covered.In the build it's :tonsec NOT ":tosec", but seems to work:Code: Select all:put [:tonsec [:timestamp]] 1692285621621
:put ([:tonsec 1s]/1000000000)
1
:put [:tonum 1s]
1
:put [:tonum [:timestamp]]
1693515184
I know it is old, very old, but the BCP allows for full 1500byte MTU in a tunneled connection which runs over links with smaller MTU. It is fragmented in transit, but the connection is delivered unfragmented between the bridges. Even the NetFlix algoritms didn't see that the traffic did pass a VPN tunnel, and their content can be forwarded, what could not be done over the normal VPN connection.. BCP is explained in the wiki for PPTP only, but it works with SSTP,L2TP and PPTP. Set MRRU to a higher value (eg 1600) to have the full unfragmented 1500bytes MTU. BCP is Intended for wireless, but I used it also over ethernet as tunnel, to forward NetFlix between different networks. (Netflix tries to detect a tunnel, and then says "something went wrong".) PPTP was faster than SSTP. It was internal in the local network, so security is not facing the public internet.That why I'm not sure using a tunnel isn't such a bad option. It does let you treat wireless same as wired, vs using Wi-Fi specific WDS-like things.4-address mode is not standard in 801.11. each manufacturer that offers it has implemented their own hacks to negotiate and support it,
The bigger issue today with doing tunneling over wifiwave2 is the interface won't sent packets larger than 1500*, even if the MTU is set higher – so tunnels get fragmented.
* I don't have any wifiwave2 devices here to check specific in 7.12, so maybe it's fix, but been open 2 years.
This was mainly about transparent bridging across Wi-Fi point-to-point links. Issues with 3rd parties refusing to transport or fragment them do not apply in that case. You can see the point-to-point link as a /30 IP network with your own routers at the endpoints and a limited-MTU link in between. It is your own decision to use your routers to do fragment/reassemble. (or not to have the functionality)eoip and other tunneling protocols would only work if transporting routers along the path fragment packets on the way instead of rejecting and dropping it.
That's been showing up that way for a while and isn't related specifically to this beta.Why in IPv6 DHCP server POOL option do I get a double static-only entry's listed:
It happens here too, so it will probably be easy to reproduce (my config has a pool from DHCPv6 client but in addition it lists static-only twice).If no one make a support case out of it, it will possible stay like this.
iso.3.6.1.2.1.4.24.4.1.1.185.X.Y.Z.255.255.255.255.0.10.26.35.2 = IpAddress: 185.X.Y.Z
iso.3.6.1.2.1.4.24.4.1.1.185.X.Y.Z.255.255.255.255.0.0.0.0.0 = IpAddress: 185.X.Y.Z
Error: OID not increasing: iso.3.6.1.2.1.4.24.4.1.1.185.X.Y.Z.255.255.255.255.0.10.26.35.2
>= iso.3.6.1.2.1.4.24.4.1.1.185.X.Y.Z.255.255.255.255.0.0.0.0.0
iso.3.6.1.2.1.4.24.4.1.1.185.X.Y.Z.255.255.255.255.0.0.0.0.0 = IpAddress: 185.X.Y.Z
iso.3.6.1.2.1.4.24.4.1.1.185.X.Y.Z.255.255.255.255.0.0.0.0.0 = IpAddress: 185.X.Y.Z
iso.3.6.1.2.1.4.24.4.1.1.185.X.Y.Z.255.255.255.255.0.0.0.0.0 = IpAddress: 185.X.Y.Z
iso.3.6.1.2.1.4.24.4.1.1.185.X.Y.Z.255.255.255.255.0.0.0.0.0 = IpAddress: 185.X.Y.Z
terminal is not a TTY
died with signal 11 on Tue Sep 12 09:33:36 2023
We dont have insight into the testers lab but you may have provided a clue to what they dont use for testing ;-)Additional info:
no such problem on RB5009, AX2, AX3.
mAP and cAP Lite do have this problem, both are mipsbe. Coincidence ?
Feedback from support:mAP - 7.12beta3
when trying to open terminal via winbox connected via ROMONConnecting via Winbox / WG does allow terminal to be opened.Code: Select allterminal is not a TTY died with signal 11 on Tue Sep 12 09:33:36 2023
It worked before on 7.11.2.
Supout created, support contacted (SUP-127788).
OMG!!!What's new in 7.12beta7 (2023-Sep-13 09:58):
*) wifiwave2 - added station-bridge interface mode (CLI only);
Indeed. That was also the first thing which catched my eye ...OMG!!!What's new in 7.12beta7 (2023-Sep-13 09:58):
*) wifiwave2 - added station-bridge interface mode (CLI only);
I have been waiting for this too, so that I can use the Audience with wifiwave2 and wireless mesh. It looks like it has been added finally with this beta!How long are my 10+ months old hAP ax2s going to be stored in a drawer, because ROS 7 still does not support repeater mode? Tonnes of new features, but still not fully on parity with ROS 6.
I am not at the location where I have the Audience devices in use, but I will try it out when I get there in a couple weeks.* wifiwave2 - added station-bridge interface mode (CLI only);
Can we say, that Station bridge, having two hAP ax2, can serve a repeater purpose?
This looks like something that will affect me. Currently I've only enabled the -1 interface for each QSFP28 port and explicitly disabled -2, -3 and -4. Sounds like my config would no longer link up at 100G, is that right? The other subinterfaces must just be left enabled, with no other configuration necessary?qsfp - use sub-interface configuration for establishing link (for 40Gbps and 100Gbps links, all sub-interfaces must be enabled);
Is this the preferred setup for Capsman? Or do we still connect as a station?The station-bridge mode works together with the AP mode, there is no need for a new type of AP mode. The WifiWave2 station-bridge is only compatible with WifiWave2 APs.
Audience configuration won't be automatically converted in this case.
That's a bummer on two fronts.The station-bridge mode works together with the AP mode, there is no need for a new type of AP mode. The WifiWave2 station-bridge is only compatible with WifiWave2 APs.
Audience configuration won't be automatically converted in this case.
Well, your hAPac and cAPac owners with the hardware, got screwed by the wifiwave2 package size long ago ;)So I guess I'm screwed and have to use legacy wireless package on Audience forever?Audience configuration won't be automatically converted in this case.
bridge mode is already incompatible between manufacturers. basically you have to see wifiwave2 devices as a different manufacturer.Please, any chance to make the station-bridge mode compatible over the air between different generations of devices?
... it simply works*) wifiwave2 - added station-bridge interface mode (CLI only);
Agree.Complain will not help, appreciate the improvements, it is what it is...
There is some compatibility "over the air" between MT and UBNT, and MT should certainly be able to make their own products compatible with itself. Tunnels have their own issues, such as MTU overhead. In my small WISP network (mix of UBNT, MT and Cambium radios) I can easily bridge even slightly larger "mini jumbo" frames - RFC4638 PPPoE over VLAN just works, no fragmentation. It's a pity that WDS/bridge has not been made an official WiFi standard even after all these years, it's an useful feature in widespread use.bridge mode is already incompatible between manufacturers. basically you have to see wifiwave2 devices as a different manufacturer.
when you require transparent bridging over different models/manufacturers devices, look at the possibility of adding a tunnel layer on top of a normal wifi connection, as explained elsewhere in this and other topics.
You can configure an L2 tunnel (EoIP, for example) to run on the wireless link, then bridge that tunnel on the AP and the station.I also use old netmetal 5 and hap ac in station-bridge mode to connect to the Audience, but those don't support wifiwave2. So I guess I'm screwed and have to use legacy wireless package on Audience forever?
Will you install completly different generations of Ruckus, to be on pair with your MT complaints? If you are ready to buy new Ruckus-everything, go scrap your old MT stuff and go wifiwave ax full-force too?I'm getting very close to selling all my MikroTik APs and going to Ruckus for wireless. I'll still use MikroTik for routing, though.
MikroTik wireless is flexible, but just not performant. And this wifiwave2 business is looking to be the last nail in the coffin for me. I've been waiting for years, and now we finally have station-bridge, only for me to find out here that we can't interop wireless station-bridge to wifiwave2 ap mode? So now my capsman wireless bridge setup between 2x Audience, 1x hap ac, and 1x netmetal 5 will be hindered to use the legacy wireless package on the Audience forever. Correct me if I'm wrong, but otherwise I'm in the market to dump MikroTik wireless for something else known to be much more performant since I would have to replace two of my MikroTik APs just to get wifiwave2.
My point was that if I'm going to have to buy new MikroTik APs to get better performance, I may as well buy Ruckus instead since they dramatically outperform MikroTik in speed and coverage.Will you install completly different generations of Ruckus, to be on pair with your MT complaints? If you are ready to buy new Ruckus-everything, go scrap your old MT stuff and go wifiwave ax full-force too?
Once again - you are using crapload of mixed generation of devices and complain. Throw it out and stick to wifivave2, as it will improve over time. Noone's going to fix the old stuff.My point was that if I'm going to have to buy new MikroTik APs to get better performance, I may as well buy Ruckus instead since they dramatically outperform MikroTik in speed and coverage.Will you install completly different generations of Ruckus, to be on pair with your MT complaints? If you are ready to buy new Ruckus-everything, go scrap your old MT stuff and go wifiwave ax full-force too?
In one location I already replaced 3 MikroTik full-power/gain APs with a single Ruckus and I have the same coverage and faster speeds. It's well known in the industry that Ruckus has the best coverage of all vendors, and nearly the fastest (I think Aruba was faster at short range, iirc).
Ruckus supports wireless bridge, which is what I really need at the other location where I'm currently using 4 MikroTik APs. And they allow setting DTIM to help with mobile power saving, which almost all vendors except MikroTik have supported for ages. This particular feature has been requested from MikroTik as far back as I can remember, and I started using MikroTik around 15 years ago. I've installed hundreds of MikroTik routers and APs in homes and businesses, but I think it's time to move on from the wireless side unless cost is the most important factor. But I digress.
Stupid comment without telling why.version 7 was an anti-climax for Mikrotik, not as expected.
You say forwarding but then you are testing from router to router?I have some TCP performance issue on a CCR2004 (Router B) running 7.12beta3 and beta7, but just on forwarding plane.
No, how far away is that router 10ms?I have some TCP performance issue on a CCR2004 (Router B) running 7.12beta3 and beta7, but just on forwarding plane.
Anyone having similar issues?
The router that I'm having issues with is router B, the one in the middle, which is running 7.12beta7. The other two are running 7.10.2. I was testing from router to router to actually see if there are issues on links. The last measurement is end to end, and the CCR (Router B) is the only router in the middle. These are two segments of our core network.You say forwarding but then you are testing from router to router?I have some TCP performance issue on a CCR2004 (Router B) running 7.12beta3 and beta7, but just on forwarding plane.
To test forwarding performance, put an end-system (e.g. a PC) at each end behind the routers, and test between the PCs.
E.g. running iperf3 on each of them.
Forgotten line:What's new in 7.12beta7 (2023-Sep-13 09:58):
!) sfp - convert configuration to support new link modes for SFP and QSFP type of interfaces;
*) qsfp - added 50Gbps rate support for QSFP28 interfaces;
*) qsfp - fixed sub-interface EEPROM monitor data output (introduced in v7.12beta3);
*) qsfp - improved auto link detection for 100G CWDM4 modules and AOC cables (introduced in v7.12beta3);
*) qsfp - use sub-interface configuration for establishing link (for 40Gbps and 100Gbps links, all sub-interfaces must be enabled);
https://download.mikrotik.com/routeros/ ... a3-arm.npkDowngrading to 7.12beta3 would also have worked but you can't download that release anymore.
Just change the URL to whatever version/arch you need.doesnt help. I need arm64 and the other packages as well. but thanks for sharing... Ill try again with 7.12 release when mikrotik finally figured out all its (Q)SFP negotiation mysteries.
For stable release, you can see the "directory listing"i tied that before but couldnt fiure out what was wrong. directory listing shows "forbidden". not helpful neither
:global dlros do={
:local lver "7.11.2"
:local larch "arm64"
:if ([:typeof $1]="str") do={
:set lver $1
:if ([:typeof $arch]="str") do={
:set larch $arch
}
}
:local curl "https://download.mikrotik.com/routeros/$lver/routeros-$lver-$larch.npk"
:put $curl
/tool fetch url=$curl
}
$dlros 7.12beta1 arch=arm
:global askbug [/terminal/ask "Should the preinput= be included in return?" preinput="I think no. Bug? "]
Should the preinput= be included in return?
I think no. Bug? YES
:put $askbug
Here just the "YES" should be returned, but I get the entire preinput= too...I think no. Bug? YES
:put [/terminal/ask preinput="> "]
> something
> something
Well "value-name" does what I want: keep question on same line. I swear this was changed someplace...@amm0 do you remember my implementation of this. I added a help/info line that was displayed underneath and the next input field was shifted down and displayed directly under that help/info.
So if you want to use pre-input as help/info then better ask for that extra line being displayed.
/ip/address/print
:put [/ip/address/get 0 address]
111.211.111.61/29
:put [/ip/address/get 0 value-name=address]
111.211.111.61/29
I have it hooked to a laptop I use for work so at night it goes to sleep and when I come back the next day the connection will not establish without disabling interface and enabling it or other intrusive of stuff like that.Also have S-RJ01 and RB5009, it's even my WAN link so I would know pretty soon if something goes wrong there.
How do you define brief itf down periods ? How does that happen?
1- what are the implications of this removal in scenarios where we have to match a device with fixed speed on the other side?!) ethernet - changed "advertise" and "speed" arguments, and removed "half-duplex" setting under "/interface ethernet" menu;
!) sfp - convert configuration to support new link modes for SFP and QSFP type of interfaces;
/interface ethernet set ether1 speed=10M-baseT-half auto-negotiation=no
Thanks for fixing this so quickly! it's working! (viewtopic.php?p=1027265#p1027265)*) route - reverse community "delete" and "filter" command behavior;
Oh! 😳 So I can not access sensitive data, use*) console - restrict permissions to "read,write,reboot,ftp,romon,test" for scripts executed by DHCP, Hotspot, PPP and Traffic-Monitor services;
/tool/fetch
/tool/fetch
This issue is fixed in all RouterOS releases available on our download page (v7.7 and v6.49.7 and newer).
Perhaps not exactly, but same family of "policy escalation".I think CVE-2023-30799 is unrelated, as stated in the linked blog post:
Is this present in 7.11.2 and previous or is it isolated to 7.12 versions?*) wifiwave2 - limit L2MTU to 1560 until a fix is available for a bug causing interfaces to fail transmitting larger frames than that;
It seems that SFP support is the "rocket science" of today. Unlike in their early days, rockets today often work on the first try and failures are quite rare. SFP changes usually fail every time (something is fixed, another problem is introduced).@MT: please check all of YOUR SFP(+) models on compatibility with a new ROS version!
For how many years now has this not been working? It was years before the end of v6 that passwords were no longer retrievable (only stored encrypted)..."/log [/user/get XXX password]"
Mayor/Importante changes vs normal/minor changesWhat would be the difference between itens beginning with "!)" and beginning "*)" on the release notes?
No, probably is for this....It's the "fix"* for https://blog.mikrotik.com/security/cve-2023-30799.html — a user with no "sensitive" could write one of these "on-action" scripts to do "/log [/user/get XXX password]".
So /tool/fetch should work, but it too be blocked from retrieving the various sensitive/"password attributes".
But this begs the question, does a user with only "write,read" need to have "reboot,ftp,romon,test" to be able to save one of the "on-action" scripts?
*) mqtt - added on-message feature for subscribed topics;
*) mqtt - added wildcard topic subscription support;
/iot mqtt {
:local mytopic "mikrotik/mqtt"
:local myhost "test.mosquitto.org"
# clear existing message
# /iot mqtt subscriptions recv clear
# setup new broker
brokers add address=$myhost name=mosquitto-test port=1883
# add subscription with "on-message"
subscriptions add broker=mosquitto-test topic=$mytopic on-message=":log info \"test \$msgTopic \$msgData\""
:delay 1s
# create 10 message with different "seq:" with UTF-8 encoded emoji.
:for i from=1 to=10 do={
publish topic=$mytopic broker=mosquitto-test message="{ \"seq\": $i; \"msg\": \"\F0\9F\A7\90\"}"
:delay 1s
}
# show them on screen
subscriptions recv print
# remove out broker (to force a disconnect since test server has timeouts)
broker remove [find name="mosquitto-test"]
}
** never notice but the non-printable ASCII gets removed in normal "print" but "print detail" has the \F0\9F\A7\90 still...09-27 19:24:54 script,info test mikrotik/mqtt { "seq": 10; "msg": "F09FA790"}
09-27 19:31:16 script,info test mikrotik/mqtt { "seq": 2; "msg": "F09FA790"}
09-27 19:31:17 script,info test mikrotik/mqtt { "seq": 3; "msg": "F09FA790"}
09-27 19:31:18 script,info test mikrotik/mqtt { "seq": 4; "msg": "F09FA790"}
09-27 19:31:19 script,info test mikrotik/mqtt { "seq": 5; "msg": "F09FA790"}
09-27 19:31:20 script,info test mikrotik/mqtt { "seq": 6; "msg": "F09FA790"}
09-27 19:31:21 script,info test mikrotik/mqtt { "seq": 7; "msg": "F09FA790"}
09-27 19:31:22 script,info test mikrotik/mqtt { "seq": 8; "msg": "F09FA790"}
09-27 19:31:23 script,info test mikrotik/mqtt { "seq": 9; "msg": "F09FA790"}
09-27 19:31:24 script,info test mikrotik/mqtt { "seq": 10; "msg": "F09FA790"}
> mosquitto_sub -t "#" -v
mikrotik/mqtt { "seq": 1; "msg": "🧐"}
mikrotik/mqtt { "seq": 2; "msg": "🧐"}
...
mikrotik/mqtt { "seq": 10; "msg": "🧐"}
The "first" script initiation failure happens because of the "different" MQTT packet sequence. You can inspect it using the WireShark.*) mqtt - added on-message feature for subscribed topics;
*) mqtt - added wildcard topic subscription support;
Great work! It works* and there are docs for it, too! "on-message" script works, even UTF-8 encoding passes through okay. "#" wildcard seem to work in quick test.
*minor issues:
- In some quick test (script below), it always seems to lose the first received message. /iot/mqtt/publish side works, since I get those on another host, just the /iot/mqtt/subscribe seem lost the very first one – then fine. Adding delays etc doesn't seem to change the "first lost" issue.
- Winbox shows the new settings, but seem if you delete at CLI, it doesn't always get reflected in winbox. Closing windows/reopen shows any removes.
- Perhaps reasonable. But the variable "msgTopic" and "msgData" in the "on-message="are a departure in y'all naming conventions — but the docs has example, and showed camelCase – just example $topic .
- Related, MQTT subscription message's broker=/$msgBroker and time=/$msgTime aren't included in "on-message" While those can be inferred... time= might be different if MQTT was queue'd someplace...
- Log topic is "script" which is normal for the "on-" scripts, but be nice it was under "mqtt" topic for filtering from /system/script etc...
- There is no "disable", only remove – so no way to for a disconnect/reconnect.
Here was some test code that use the Eclipse's test server, see https://test.mosquitto.org:
which shows the logged messages, except the /subscribe hook doesn't find the first one.Code: Select all/iot mqtt { :local mytopic "mikrotik/mqtt" :local myhost "test.mosquitto.org" # clear existing message # /iot mqtt subscriptions recv clear # setup new broker brokers add address=$myhost name=mosquitto-test port=1883 # add subscription with "on-message" subscriptions add broker=mosquitto-test topic=$mytopic on-message=":log info \"test \$msgTopic \$msgData\"" :delay 1s # create 10 message with different "seq:" with UTF-8 encoded emoji. :for i from=1 to=10 do={ publish topic=$mytopic broker=mosquitto-test message="{ \"seq\": $i; \"msg\": \"\F0\9F\A7\90\"}" :delay 1s } # show them on screen subscriptions recv print # remove out broker (to force a disconnect since test server has timeouts) broker remove [find name="mosquitto-test"] }
** never notice but the non-printable ASCII gets removed in normal "print" but "print detail" has the \F0\9F\A7\90 still...09-27 19:24:54 script,info test mikrotik/mqtt { "seq": 10; "msg": "F09FA790"}
09-27 19:31:16 script,info test mikrotik/mqtt { "seq": 2; "msg": "F09FA790"}
09-27 19:31:17 script,info test mikrotik/mqtt { "seq": 3; "msg": "F09FA790"}
09-27 19:31:18 script,info test mikrotik/mqtt { "seq": 4; "msg": "F09FA790"}
09-27 19:31:19 script,info test mikrotik/mqtt { "seq": 5; "msg": "F09FA790"}
09-27 19:31:20 script,info test mikrotik/mqtt { "seq": 6; "msg": "F09FA790"}
09-27 19:31:21 script,info test mikrotik/mqtt { "seq": 7; "msg": "F09FA790"}
09-27 19:31:22 script,info test mikrotik/mqtt { "seq": 8; "msg": "F09FA790"}
09-27 19:31:23 script,info test mikrotik/mqtt { "seq": 9; "msg": "F09FA790"}
09-27 19:31:24 script,info test mikrotik/mqtt { "seq": 10; "msg": "F09FA790"}
But RouterOS did send it:
Code: Select all> mosquitto_sub -t "#" -v mikrotik/mqtt { "seq": 1; "msg": "🧐"} mikrotik/mqtt { "seq": 2; "msg": "🧐"} ... mikrotik/mqtt { "seq": 10; "msg": "🧐"}
Well I didn't read the WHOLE page. ;). Adding a "/iot/mqtt/connect broker=" after subscription before publish fixed the "lost first message" problem. Thanks!Can you please elaborate on the "There is no "disable", only remove – so no way to for a disconnect/reconnect."?
Glad to hear it worked!Well I didn't read the WHOLE page. . Adding a "/iot/mqtt/connect broker=" after subscription before publish fixed the "lost first message" problem. Thanks!Can you please elaborate on the "There is no "disable", only remove – so no way to for a disconnect/reconnect."?
But, in my defense, using root /iot/mqtt with a broker= for CLI is kinda odd... I only looked in /iot/mqtt/broker, which is where "connected=yes" is shown – but was documented .
It also be nice if "connected" status showed as a field in winbox in IoT>MQTT>Broker list. I guess winbox could have connect/disconnect button & show the /iot/mqtt/subscription/recv queue in a tab — but just showing if a broker is connected in winbox that be handy.
I can confirm it is the same on CCR2216, although the CLI command still works, so it looks like it's some sort of web ui specific issue.I opened a support ticket [SUP-129558], but I'm going to post it here, too.
On both builds 7 and 9, when attempting to add any kind of dynamic interface (bridge, VLAN, bonds, IPIP/EOIP tunnels, etc.) from within Webfig, I get an error that the interface type is not supported.
Screenshot 2023-09-29 at 2.23.07 AM.png
This is happening on hAP AX3 and CCR2116, with completely different base configurations.
Yeah, support acknowledged it as well. I did what I needed to with the CLI, but wanted to report it here.I can confirm it is the same on CCR2216, although the CLI command still works, so it looks like it's some sort of web ui specific issue.
this should be put into the wiki/help pages under "system > packages" :)For stable release, you can see the "directory listing"i tied that before but couldnt fiure out what was wrong. directory listing shows "forbidden". not helpful neither
https://mikrotik.com/download/archive
If you want to download it directly using a version name from RouterOS, this may help:Code: Select all:global dlros do={ :local lver "7.11.2" :local larch "arm64" :if ([:typeof $1]="str") do={ :set lver $1 :if ([:typeof $arch]="str") do={ :set larch $arch } } :local curl "https://download.mikrotik.com/routeros/$lver/routeros-$lver-$larch.npk" :put $curl /tool fetch url=$curl } $dlros 7.12beta1 arch=arm
name: sfp-sfpplus1
status: link-ok
rate: 2.5Gbps
full-duplex: yes
tx-flow-control: no
rx-flow-control: no
sfp-module-present: yes
sfp-rx-loss: no
sfp-tx-fault: no
sfp-type: SFP/SFP+/SFP28
sfp-connector-type: SC
sfp-link-length-sm: 20km
sfp-vendor-name: Hisense-Leox
sfp-vendor-part-number: LXT-010S-H
sfp-vendor-revision: 1.0
sfp-manufacturing-date: 22-08-27
sfp-wavelength: 1310nm
sfp-temperature: 54C
sfp-supply-voltage: 3.357V
sfp-tx-bias-current: 16mA
*) interface - added "macvlan" interface support;
The output comes from the 7.11 firmware and I posted it to show what SFP module I am using. My device is RB5009Upr+S+In. Using 7.12b9 there is no obvious reaction when inserting SFP module. No light, no indication in the winbox that something has been plugged in - there is no checkbox next to "Module Present" in SFP interface status.@forteller, thanks for the feedback!
But we need more details to reproduce this issue in our labs. The sfp-sfpplus monitor actually shows that "status: link-ok". Can you share supout.rif file when the issue is active and send it to support@mikrotik.com? If not, can you specify what device are you using, show the full output of the monitor command and printout from "/interface ethernet export"?
Interesting that you found an actual problem resulting from that behavior. But did you really confirm it to be the reason?For example: Cloudflare tunnel cannot be started because the returned (argotunnel.com) domain record does not contain an SOA record.