May 2016 - Public - talk.mikrotik.com.au

VDSL modems
by Mike Everest 07 Oct '16

07 Oct '16

Hi Folks, A few months back, I asked for opinions on a good adsl modem to use a bridge for MikroTik PPPoE. With increase in prevalence of VDSL services (especially with #nbwhen ;) does anyone have some opinion of readily available and 'good' (not necessarily cheap ;) vdsl modems? Cheers! Mike.

10 14

queues I cant get my head around
by Patrick Sayer 26 May '16

26 May '16

I want to stop any given hotspot user from maxing out the upload bandwidth on a connection (ADSL2+ in most cases). I dont want to have to add queues for each user as this obviously changes. I don't really want to rely on the burst settings as this is time based and if the user is the only one using bandwidth they will get limited after the burst time is used up regardless if noone else wants bandwidth at that time still. What I would like to see happens is each user is allocated a minimum amount of bandwidth (like PCQ can do), but allow them to burst indefinitely rather than on a timed basis. And if multiple users try to use bandwidth then the burst they receive gets shared in the same way that PCQ is doing in the first place. Having 10 users in a PCQ list with a total of 1Mbit bandwidth will limit each user to 100Kbps even if only one of them is using the system at the time from what I can see, and this is not a desired outcome. Does this explanation make sense and has anyone come up with a solution along those lines? Thanks in Advance, Patrick

1 0

bgp peering with cisco
by Alex Samad - Yieldbroker 26 May '16

26 May '16

Hi Wonder if any one has come across any issue with MED attribute not been see from a cisco router. I have a client saying they are setting the MED attribute on a prefix and its not showing up on my end. Alex

3 7

Re: [MT-AU Public] VDSL modems
by Stephen Schwetz 20 May '16

20 May '16

Most NBN fftn providers will give you an VDSL modem when you sign up. These have been fully tested with the NBN, and most support bridging out of the box, excluding OPTUS, but it can be bridged with some reading and trial and error. My sagemcomm from OPTUS is rock solid and has not needed a reboot in bridged mode since it was installed in Jan On 20 May 2016 16:34, "Mike Everest" <mike(a)duxtel.com> wrote: Hi Folks, A few months back, I asked for opinions on a good adsl modem to use a bridge for MikroTik PPPoE. With increase in prevalence of VDSL services (especially with #nbwhen ;) does anyone have some opinion of readily available and 'good' (not necessarily cheap ;) vdsl modems? Cheers! Mike. _______________________________________________ Public mailing list Public(a)talk.mikrotik.com.au http://talk.mikrotik.com.au/mailman/listinfo/public_talk.mikrotik.com.au

1 0

EOIP Epic Fail
by Paul Julian 17 May '16

17 May '16

Hi guys, thought I would post this information to see if anybody had experienced the same issue. We have an RB1100AHx2 in a data centre that connects to a CCR in another data centre, unfortunately we can't do QINQ between the DC's so we run an EOIP tunnel to connect two vlans for a customer through to there, all has been working fine for about 2 years. The other day, just out of the blue, the customer said their link had gone down, after a number of hours troubleshooting, convinced that it was a problem on site, we discovered that the EOIP tunnel just stopped passing traffic. Nothing had changed on those two routers for months but all of a sudden they just stopped working. The CCR has some other EOIP tunnels on it which were still working fine, but the 1100 was refusing to work. We checked firewall rules, vlan ID's, tunnel ID's, everything, and there were no issues, the tunnel just refused to pass data. I tried changing the tunnel ID and it made no difference, tried re-adding the vlans to it, nothing, disabled and enabled it, nothing made a difference. Out of desperation we created a new EOIP tunnel on that router with a new ID and pointed it to the existing EOIP interface at the other end and it came up straight away. So apart from a big WTF, has anybody experienced this before ? Regards Paul

6 7

Re: [MT-AU Public] simple queue
by Alex Samad - Yieldbroker 17 May '16

17 May '16

Okay another attempt /queue simple add comment="guest to yb network is max" dst=10.0.0.0/8 name=guestYB target=10.172.202.0/24 add comment="YBGuest to internet" max-limit=10M/10M name=guestInternet target=10.172.202.0/24 no other rules, remove the mangle rules the bottom works, everything gets limited. The top one is meant to capture guest to internal ip addresses but it doesn't seem to work. A -----Original Message----- From: Alex Samad - Yieldbroker Sent: Tuesday, 17 May 2016 11:47 AM To: public(a)talk.mikrotik.com.au Subject: simple queue Hi Trying to limit up / down from 10.172.202.0/24 to the internet http://wiki.mikrotik.com/wiki/Manual:Queue#Queue_Types http://wiki.mikrotik.com/wiki/Manual:IP/Firewall/Mangle /queue simple remove [ find where ! dynamic ] add comment="ratelimit for yboGuest network and only for marked packets" max-limit=10M/10M name=guestToInternet packet-marks=guestInternet target=10.172.202.0/24 /ip firewall mangle remove [ find where ! dynamic ] add action=mark-packet chain=forward new-packet-mark=guestInternet dst-address=10.172.202.0/24 src-address=!10.0.0.0 place-before=0 comment="Internet to guest" add action=mark-packet chain=forward dst-address=!10.0.0.0 src-address=10.172.202.0/24 new-packet-mark=guestInternet place-before=0 comment="Guest to internet" I can see packets matching the mangle rules but not in the queue. Also I am using fastconnect which is why I have the pattern matching mangle rules above the fastconnect rules ... if that matters ! Do I have to turn of fast connect or what am I doing wrong !!! A

3 7

simple queue
by Alex Samad - Yieldbroker 17 May '16

17 May '16

Hi Trying to limit up / down from 10.172.202.0/24 to the internet http://wiki.mikrotik.com/wiki/Manual:Queue#Queue_Types http://wiki.mikrotik.com/wiki/Manual:IP/Firewall/Mangle /queue simple remove [ find where ! dynamic ] add comment="ratelimit for yboGuest network and only for marked packets" max-limit=10M/10M name=guestToInternet packet-marks=guestInternet target=10.172.202.0/24 /ip firewall mangle remove [ find where ! dynamic ] add action=mark-packet chain=forward new-packet-mark=guestInternet dst-address=10.172.202.0/24 src-address=!10.0.0.0 place-before=0 comment="Internet to guest" add action=mark-packet chain=forward dst-address=!10.0.0.0 src-address=10.172.202.0/24 new-packet-mark=guestInternet place-before=0 comment="Guest to internet" I can see packets matching the mangle rules but not in the queue. Also I am using fastconnect which is why I have the pattern matching mangle rules above the fastconnect rules ... if that matters ! Do I have to turn of fast connect or what am I doing wrong !!! A

3 3

interpreting sniffer logs
by Mike O'Connor 10 May '16

10 May '16

Hi All I'm looking at a sniffer log and trying to understand what a ':' in vlan id might mean, as seen below. I've not been able to find a reference to this else where. 5 time=436.901 num=461 direction=rx src-mac=00:1B:D5:8A:63:9C dst-mac=01:00:0C:CC:CC:CC vlan=2001:5 interface=AAPT protocol=802.2 size=503 cpu=1 fp=no Cheers Mike

3 2

Router Recomendation for 400/400 link
by Ben Jackson 09 May '16

09 May '16

Hi all, Does anyone have a recommendation for the cheapest possible MT router that will support a symmetrical 400mbps fibre link without compromising the available bandwidth. As far as I know there should be no complex firewall rules involved - just basic port forwarding / NAT masquerade. Thanks, Ben Jackson eLogik m:0404 924745 e: ben(a)elogik.net w: www.elogik.com.au [image: http://www.elogik.com.au] <http://www.elogik.com.au>

4 4

Multicast
by Alex Samad - Yieldbroker 06 May '16

06 May '16

Hi Trying to work out an issue with multicast routing. Trying to connect to 2 sources 233.71.185.130 233.71.185.146 /routing pim> export # may/06/2016 13:56:54 by RouterOS 6.33.3 # /routing pim interface add igmp-version=IGMPv3 interface=MCasx add igmp-version=IGMPv3 interface=Management /routing pim rp add address=203.0.119.247 comment="233.71.185.128/28 and 233.71.185.144/28" group=233.71.185.0/24 Inbound (upstream is MCasx) /routing pim> igmp-group print Flags: v1 - IGMPv1, v2 - IGMPv2, v3 - IGMPv3, I - include, E - exclude, F - forward, D - don't forward INTERFACE GROUP SOURCE TIMEOUT v3E Management 224.0.0.2 0.0.0.0 4m16s v3E Management 224.0.0.13 0.0.0.0 4m16s v3E Management 224.0.0.18 0.0.0.0 4m16s v3E Management 224.0.0.22 0.0.0.0 4m16s v3E Management 224.0.0.252 0.0.0.0 4m18s v3I Management 233.71.185.130 0.0.0.0 0s v3F Management 233.71.185.130 203.6.253.212 4m19s v3I Management 233.71.185.146 0.0.0.0 0s v3F Management 233.71.185.146 203.6.253.214 4m18s v3E Management 239.255.255.250 0.0.0.0 2m13s v3E MCasx 224.0.0.2 0.0.0.0 2m24s v3E MCasx 224.0.0.13 0.0.0.0 2m24s v3E MCasx 224.0.0.22 0.0.0.0 2m24s v3I MCasx 233.71.185.130 0.0.0.0 0s v3F MCasx 233.71.185.130 203.6.253.212 2m16s tcpdump on the MCasx shows packets coming in. tcpdump on the Management doesn't show any packets leaving /routing pim> join print Flags: RP - (*,*,RP), WC - (*,G), SG - (S,G), SG_rpt - (S,G,rpt) GROUP SOURCE RP WC 224.0.0.0 203.0.119.247 203.0.119.247 SG 224.2.2.3 0.0.0.0 SG 239.225.100.100 0.0.0.0 SG 239.255.255.250 0.0.0.0 SG_rpt 224.2.2.3 10.32.50.52 SG_rpt 233.71.185.130 203.6.253.212 203.0.119.247 SG_rpt 233.71.185.146 203.6.253.214 203.0.119.247 Any suggestions ??

1 1