Starting a new NOC and looking for a switch in my Switch Centric design. My current setup includes 2 Dell 4048s-on switches. Should I go with those or is there a viable Mikrotik option. I am looking to run CCR2118 as border and nat routers. I don’t need 48 Ports but need the performance. What hardwa...
I have a subnet I have advertised in bgp /routing bgp network print Flags: X - disabled # NETWORK SYNCHRONIZE 0 100.110.12.0/22 no I want to filter(?) out and not advertise 100.110.12.36/30 How would I do that? I have moved the client to a different Tower and am unable to change the client IP. Still...
Love answering my own questions. 1.Copied the Software ID in System -> license 2. I went to the mikrotik.com site. Logged in. 3. Purchased a level 4 Key. ($45) 4. Copied the Key off the order details. 5. Pasted the key into the unit under System -> License Thanks for thinking kind thoughts. Rick
I have a CCR2044-1G-12S+2XS. Am passing traffic and in production. I have no Link lights or 10G lights lit on any ports. Never have. The patch cable to the other device (CRS309-1G-8S+IN and CCR-1036 12G-4S) all have Link lights on and also are passing traffic. What do I need for the Link lights to c...
So I am on the verge of setting up a BGP with my 1st peer and he mentioned using a /31. I am going to assume he is using a Cisco. I am a Mikrotik shop.
You are saying that Cisco and Mikrotik are not able to use BGP on a /31?
I'm wanting to upgrade the time server settings on all my routers with a script. /system ntp client set enabled=yes primary-ntp=ipAddress secondary-ntp=ipAddress But this will not take. No errors when I execute. But no change in the GUI in winbox. At command line: /system ntp client print shows the ...
This is more then a beginner question. I would restate the question with information on your routing configuration and post it in at least the General area if not the Forwarding Protocols area. That is where your routing experts will lurk.
I would expect if you plug into the 2 ethernet port you would get an IP from the unit (depending on how it is configured.) Once connected, take a look and see what your IP and gateway are. You may be able to open the webfig screen by pointing your browser at the gateway IP. The default user name is ...
Thank you for that insight about EOIP. I believe that may be the smoking gun in my case.
I have used eoip for various access situations and the scenario fits with my problems with https.
I have a RB CCR1009-8G-1S-1S+ Last Thursday I upgraded from 6.37.1 to 6.38.5 Now I have no web access to the router / no queue graphing from the router. I have port 80 in services enabled. (never changed it) [rboucher@LookoutCCR] > ip service print Flags: X - disabled, I - invalid # NAME PORT ADDRES...
I have the same problem on some of my routers. Not all. https sites like https://wellsfargo.com can not be rendered. Other sites like https://crucial.com are very slow to render. I do not have a router workaround. The problem is exacerbated by some third party routers at the client location. Like a ...
I should set this on the Gateway router? The client is a 3rd party router. (netgear, dlink, ) Set the MAX MTU/MRU to 1492 on the third party or on my central distribution router?
Test my network I noted that an in-between router 10.100.2.5 drops 100%. Yet I can ping just fine. [sjwrick@main] /tool traceroute 10.100.17.1 # ADDRESS LOSS SENT LAST AVG BEST WORST STD-DEV STATUS ...
Core router is set this way /ip dns print servers: 8.8.8.8,8.8.4.4 dynamic-servers: allow-remote-requests: yes max-udp-packet-size: 4096 query-server-timeout: 2s query-total-timeout: 10s cache-size: 2048KiB cache-max-ttl: 1w cache-used: 1972KiB I was told that Google DNS can be flaky so I can use an...
Running ROS 6.35.2 RB 2011UiAS (though have seen issue on other routers) Customer connects and is able to ping out and open some websites. Other sites (which are almost always predictable across the client base i.e.: crucial.com, eyeonwater.com, msn.com, download apple app store items) DNS times out...
Running ROS 6.34.4 on a CCR1009-8G-1S-1S+ with Google DNS 8.8.8.8, 8.8.4.4 Using Hotspot with radius authentication. Ubiquity Radios (AP and Client) Client have Netgear Routers some Apple Airports, some Dlink all running Google DNS 8.8.8.8, 8.8.44 Clients are having trouble with DNS on some sites. F...
I have a RB1100AH at both ends. A Ubiquiti Airfiber24 link and a Ubiquiti PowerBridgeM5 link I want to bond these two to provide link redundancy. The primary traffic would run through the AirFiber24 link. When heavy rains hit and that link drops the traffic would move to the PowerBridgeM5 link 1. Is...
Workaround? I started including the ntp package in my routers and the client started synchronizing again. Mind you, I did not enable the ntp server. Just uploaded the ntp package and rebooted. Now my systems are show correct date and time. This has fixed my issue on all platforms mipsbe ppc tile Rick
I'm using the advice from this site http://aacable.wordpress.com/tag/mikrotik-fail-over/ Tested this morning as a matter of fact. Worked for me. We have a wireless 5m feed for a resort and a backup 1.5 m feed. If the wireless goes down it auto switches to the DSL feed then switches back after the wi...
I have filters that block 25 to my input chain. If I have a public ip dst-nated to an internal private ip, does to filter to my input effect that traffic? ie On my router: /ip address add address=x.x.x.x/24 comment="Public IP - for cust" interface=WAN add address=y.y.y.1/30 comment="P...
Can someone point me to how I might do "state tracking of TCP connections" with ROS I have a Content Delivery Network (CDN) chewing up bandwidth with repeated icmp traffic. The traffic is bouncing back and forth between my gateway router and the source causing headaches with my upstream pr...
/ip firewall nat chain=dst-nat dst-address=publicipaddress action=dst-nat to-address=internalIpAddress You can specify ports also thus limiting exposure. So if it is just a web server direct port 80 to the script by adding protocol=tcp dst-port=80 Be sure to add the public ip to the WAN interface. F...
I believe ou have to set the address pool to "none" in the hotspot server. If you don't the captive portal causes havoc with the static addresses you assign which are not hot spotted. Leave the DHCP server using a pool alone, but turn the hotspot server address-pool=none /ip hotspot set 0 ...
OK so the l2mtu is 1598 for the error prone port3 and 1600 for port11 (see below) name="eth3" mtu=1500 l2mtu=1598 mac-address=D4:CA:6D:3F:27:CC arp=enabled auto-negotiation=no full-duplex=yes speed=100Mbps master-port=none bandwidth=unlimited/unlimited switch=switch2 name="eth11"...
We have a RB11ah (ROS 5.19) running as a bandwidth box. UbiquityPowerBridge <---> switch <--> RB1100AH port3/Queueing/Port4 <--> RB1100 (ROS 5.19) Also have a x86 (ROC 5.13) slaved off the switch for hotspot customers. We are only able to produce a variable 3-8 MMbps through port 3. When inserting a...
jcem: It seems like this x86 is suggish and I wonder if HDD space is the culprit. Can't figure out why 80 GB is used up: Usermanager maint: Database size: 595.0 Kib In use: 100% Last rebuild: 06/01/2012 21:58:17 Last backup: 06/01/2012 21:52:47 Last restore: Never Free disk space: 249.4 Mib Rick
cbrown: I do not use the dude. I previously did and have removed it. I also did a netinstall on this x86. It was a while ago. I think I wiped the drive and reinstalled. After the re-install I added the usermanager package and then used my backup config to restore settings. Space was eaten up fairly ...
I have a hotspot machine which has an 80 GB harddrive. I am using Usermanager. ROS 5.14 x86. 350+ users.. I have done maint to the UserManager and can't figure out what is taking up all the space. Usernamager database only takes up 584KB tool user-manager database> pri size: 584 kB in-use: 100 % log...
I have a hotspot machine which has an 80 GB harddrive. I am using Usermanager. ROS 5.14 x86. 350+ users.. I have done maint to the UserManager and can't figure out what is taking up all the space. Usernamager database only takes up 584KB tool user-manager database> pri size: 584 kB in-use: 100 % log...
I have a usermanager install with an 80GB hd. ROS 5.9 The router was upgraded long ago from ros3, then ros4, now on ros5.9 The hard drive is full. I've deleted logs and sessions etc to no improvement. Verry little space was freed up by that process. What little space that was recovered filled right ...
Have two other RouterBoards a RB750G(ros 5.9) and a RB1100 (ros 5.9) that both report ethernet connections fluctuating 10M/1000M a few times and then the RouterMaxx reboots.
What should I be looking for to resolve this? Is ros 5.9 stable?
I want to forward a number of different Public IPs to specific private ips behind my router. One to one. Each public IP needs to be mapped to a specific private IP. I have the basic functionality in place, however my non-public ip clients also reflect that they are coming from the public IP assigned...
We use /ip hotspot ip-binding to have direct access to some ns2 aps on the hotspot network.
ip hotspot ip-binding print detail
Flags: X - disabled, P - bypassed, B - blocked
0 P mac-address=00:15:6D:F4:91:1F address=10.5.56.160 server=RoseMeadow_HS type=bypassed
Doing thia for a school. We are using dansguardian with squid along with mikrotik transparent proxy. I use port 8080 for dansguardian. If you don't use dansguardian I think you'd use squid port 3128 ofcourse /ip address print 0 10.1.1.1/24 10.1.1.0 10.1.1.255 Lan 1 10.1.2.1/24 10.1.2.0 10.1.2.255 Pr...
UM 4.9 Is there a way to set the profile after the user has been created in the command line interface? I see that I can establish a "template" profile in the GUI before I create a slug of predetermined users. Then I use the copy-from command from the add command to connect a user to that ...
I'm using Usermanager on ROS 3.27 with 6 different locations of Hotspots and all is working well! However, I want to limit time to not usage but from 1st login to the limit irregardless of usage. EX: 1. User buys a 24 Hour login. 2. User logs in at 10am and uses for 2 hours that day. 3. User tries t...
Upgraded to ROS 3.30 from 3.10(???) x86 In '/queue simple' when I set max-limit to unlimited/unlimited no tracking or accumulation of bytes occurs. At all in any stat. I set the max-limit to something huge like 50M/50M and accumulation works EXCEPT in the 'Total Bytes'. Graphing has no data - flatli...
OK - I was looking for "Universal Client" or something similar. Will try out adjusting adress pool and see how it goes and will read up on "one-to-one". I'm actually more interested in why the network crashes but maybe the MAC address translation via ARP is causing the issue. Tha...
I have an established wireless/wired network using MT 3.1. Multiple static /30 subnets for my customers. Trying to migrate from our former Hotspot solution to a MT hotspot solution. Got things working great with the MT hotspot using radius and user-manager with a separate MT router UNATTACHED to my ...
Running a powerrouter ROS 3.22 4 different classes of clients on DSL running PPPoe PPPoe rate limits looking like they limit things per username just fine. I only have a 10 MB pipe and want to run PCQ to balance the load. 200+ customers and growing. I have 3000/512 mbps, 1500/256 mbps, 768/128mbps a...
I too am having no problem with static ips and graphing. In my situation we wanted to charge for high usage so we set up a second MTbox with a bridge, Then I have unlimited queues setup for each customer IP. (the real BW limits are on the the gateway router) Graphing works very well. I have the data...
I'm running this for a Chistian school here in town I run the squid/dansgaurdian on the same physical port as the schools lan- but a different network I set up the webproxy on the MT box: port:8080 parent proxy: 10.1.2.2 (ip of squid box) Parent Proxy Port: 8080 (that is for dansguardian you might u...
We use http://www.digital-loggers.com/lpc.html
We can cycle via a webpage or it also comes with autoping. If the ip does not ping after x minutes it cycles the power on the unit.
We have had real good luck with our units (8 in the field) but have seen some others complain about em.
Sorry - I know it is not a Tranzeo forum. Response from Tranzeo was: What you are experiencing here is from the layer 2.5 bridging. What happens is when traffic passes from the CPE to the AP it will use the MAC address of the CPE. The only way to avoid the layer 2.5 bridging is to have the radios in...
Situation 1 INET <--> MTRouter <--> AP1 <--> CPE1 <--> Client1 ....................... | .................... X ............... Y ....................Client2 Have networking issue that occurs when I switch the AP and CPE roles. The above diagram works just fine. The AP and CPE are Tranzeo 900 gear i...
Running v3.10 Have a public IP for customer base which masqs just fine. Public ip is x.y.z.57 WhatIsMyIP says I'm ip address x.y.z.57 Fine, works right, all are happy. I then add in a customer who wants a public ip forwarded to his router. I assign the public ip x.y.z.69 on the public interface Setu...
I am moving to a different ISP and need to redirect my customers that have static ips in their home routers to a new ip address for DNS. Example: Customer has a static IP route through my MT 2.49 gateway router. Their current DNS queries 10.0.0.1 Without going to every single customer I want to redi...
Looking to manage Bandwidth using protocol matching I want to use Queue Tree and PCQ to do this. However I really like to see the graphing based on Simple Queues So I would like to shape the BW like this Mangle and mark P2P and http traffic and non-P2P Setup Queue Tree to prioritize traffic (http pr...