MikroTik

AlohaSpark

Quick question. Is there a way to add prefixes received via BGP to an address list?

I would like to add the prefixes of a specific ASN for example, to an address list.

Alternatively, is there a way I can filter/mangle traffic for a specific AS?

AlohaSpark

You don't have an aggregation-layer where you terminate such (very high-speed) customers ? Straight onto your core switch doesn't sound like a very good plan to me. I don't know what CPU-power you have in your coreswitch-device, but perhaps you can iterate from 1Gbps shaping profiles to see howmuch...

AlohaSpark

Hello, I have a client that will avail of a 35Gbps line from us with BGP. Is simple queue still the way to go with this? I'm planning to patch them into our core switch but then what? I've only ever tried shaping up to 1Gbps. Our core does not have connection tracking. How do we shape these kinds of...

AlohaSpark

Did you get to the bottom of this? I'm seeing the same when we try to walk OID 1.3.6.1.4.1.14988 on our CRS354. Unfortunately not. I just outright skipped those slow OIDs. In my case, I am more concerned about the health of the switch (temps, fan speeds, etc.) rather than the actual interface stats...

AlohaSpark

Hello, I have a situation I have not encountered before. I have a fairly simple network VLAN setup. A CHR instance has an interface VLAN numbered 123. It has a LAN Address and a DHCP server to establish this VLAN network. CHR (Interface VLAN 123) -> CRS Switch -> CRS Switch -> hAP Lite -> Port 3 (Un...

AlohaSpark

I created a few raw firewall rules with Content set to a domain name (Firewall -> Raw -> Advanced -> Content). For example: /ip firewall raw add action=add-dst-to-address-list address-list=REDIR_FB address-list-timeout=none-static chain=prerouting content=.facebook.com src-address-list=LAN Initially...

AlohaSpark

I would like to enable/disable firewall rules based on criteria, such as how much traffic is passing through an interface (e.g. mark-routing). All other rules are already in place, such that users experience doesn't get affected when it happens, and traffic still flows smoothly (via another interfac...

AlohaSpark

I'm collecting statistics using Prometheus and snmp_exporter . I've also hand-picked OIDs to scrape in order to optimize scraping time. So far, all is great. Scraping on my CHRs take milliseconds, CCRs around 2 seconds. The second slowest , CRS326-24G-2S+, takes 4-5 seconds to scrape. The slowest of...

AlohaSpark

I am implementing sticky connection with guidance from this MUM presentation . In slide 27, it says to exempt mangling LAN routes by adding an accept rule for it above all other mangle rules. The effect of this is if a connection is within LAN only, simply use the main routing table. If not, add a c...

AlohaSpark

I am reading Edge Router & BNG Optimisation Guide for ISPs and implementing some of its advice. However, under QoS and Bufferbloat control , it says This means configuration wise, you apply the FQ_Codel queueing to all your physical ports and wireless interfaces across all your network devices. ...

AlohaSpark

I have a CCR2116-12G-4S+. My traffic levels are 3-4Gbit/s during peak hours. My current setup is that I have my downstream and upstreams running on a single SFP+ interface using VLANs. No bridge - this is a core router. Would there be any advantages (e.g. performance, lower CPU load, etc.) if I spli...

AlohaSpark

We get the problem so go around the issue. At these speeds you are not a domestic customer anymore and you don't want a PPPOE connection because you know there is going to be overheads and MTU clamps. You want a transit link from your upstream provider like any normal ISP would take and in this cas...

AlohaSpark

I would like directly connected and static, and OSPF routes on main table to also be reflected in table rtab1 . How would I do that? So far, I have tried making two local BGP instances (connecting to 127.0.0.1) and specifying the routing tables, but the session won't establish. Am I doing this right...

AlohaSpark

Have you found a solution by any chance? I'm facing the same issue.

AlohaSpark

For reference, I am having this exact same issue: viewtopic.php?t=200771#p1031293

AlohaSpark

I'm using RouterOS 7.13. My WAN connections are four 1Gbit PPPoE connections as well as two lightly-used BGP full-table upstreams. The issue I'm facing is that whenever the PPPoE connections are loaded to around 2Gbps total, one of the CPU cores is heavily loaded around 70% on one core while the ave...

AlohaSpark

I have resolved the issue. Thanks for sharing the solution, you b–d. Oh, right, the solution :lol: I made a new certificate with a 1024-bit key. It's more than sufficient, overkill even, for my use case. Initially the key for the SSL certificate was 4096-bit RSA, which I believed was fine coming fr...

AlohaSpark

I have resolved the issue.

AlohaSpark

The title says it all. I am issuing REST commands via www-ssl. Whenever I am using the SSL REST API, such as listing DHCP clients (~1k) or editing a comment on one of them, CPU usage spikes across all 16 cores (around additional 20%) with one or two pinned at 100%. SSL is also using 100% CPU. Is the...

AlohaSpark

RTFM has indeed clarified everything. Thanks

AlohaSpark

Hello, I am about to have several new eBGP upstreams with full routing tables. I unfortunately have little experience on this since for the longest time I only had one. I would like outbound traffic to go via the closest route i.e., best path. Otherwise, if the paths are of the same cost, do ECMP. D...

AlohaSpark

Hello, I have a script which runs every second that monitors bandwidth, then selects which appropriate WAN it should go through based on used bandwidth capacity. So far it works great. The only issue though is it has a 1-second reaction time between runs. That is, it takes up to 1 second for the swi...

AlohaSpark

I can see that the scheduler can only go down to executing a script every second.

How do I run my script every 0.25 or 0.2 seconds?

It's an experimental monitoring script.

AlohaSpark

I'm looking to transfer a CHR license as a birthday gift to a friend of mine. How do I do that?

AlohaSpark

Is the peering iBGP or eBGP between the routers? What address are you trying to ping and where does it stop if you traceroute? It's eBGP between ISP my Mikrotik (on their premises). I then tried to set a static route to that remote MT and my local MT. All good now, it seems to have been a Windows i...

AlohaSpark

I don't buy the ether2 route table entry: DAc 192.168.1.1/29 ether2 0 I expect it to look like: DAc 192.168.1.0/29 ether2 0 Take a look at corresponding /ip address add network= All good now. I manually edited the entries for privacy; this is related to this other post: https://forum.mikrotik.com/v...

AlohaSpark

Hello, I have set up static IPs on my Mikrotik hEX, set up a DHCP server, and connected my laptop. All is well. My laptop can ping other devices on the LAN (192.168.1.3, etc.) as well as the MT itself (192.168.1.1). However, when I add an IP address on another interface, the laptop cannot ping that ...

AlohaSpark

Do you have a firewall configuration?

Currently, no. Both MikroTiks have zero firewall/NAT rules.

AlohaSpark

NOTE: All the IP addresses below are examples, and are not my actual public IP addresses. I have a Mikrotik located on my ISP's premises. I also have a /24 block and ASN, which is announced just fine. [admin@MikroTik] > /routing/bgp/advertisements/print 0 peer=bgp1 dst=49.123.123.0/24 afi=ip nexthop...

AlohaSpark

NOTE: All the IP addresses below are examples, and are not my actual public IP addresses. I have a Mikrotik located on my ISP's premises. I also have a /24 block and ASN, which is announced just fine. [admin@MikroTik] > /routing/bgp/advertisements/print 0 peer=bgp1 dst=49.123.123.0/24 afi=ip nexthop...

AlohaSpark

Since you got BGP on all devices, why even bother running ospf (except for maybe IGP)?
You could attribute a better local preference to routes learned via your favorite router and prefer these routes.
hth, hk

That's true. You're right.

I set up a static route and got rid of OSPF. All good now.

AlohaSpark

Hello, My main CHR has 3 main uplinks. Two are via direct BGP, and one is via another Mikrotik (which I have control over) on the ISP's premises. All 3 uplinks receive full table BGP. The issue is, the two providers who are direct BGP peers have high latency. The one that's sharing routes via OSPF i...

AlohaSpark

I just found out Mikrotik has a chart for this exact question. It's over at https://wiki.mikrotik.com/wiki/MikroTik_wired_interface_compatibility In my case, the answer is yes. Not sure about the Mellanox though, but I'm willing to bet it's a yes as well. https://i.ibb.co/9Vvz1vy/Screenshot-2023-08-...

AlohaSpark

I have a CRS326-24S+2Q+RM connected to a Mellanox ConnectX-3 (two QSFP+ ports) via a copper cable, specifically a Q+DA0001. I was looking to get another Q+DA0001 to connect the other pair of 40Gbps ports. The problem is, my go-to shop only has XQ+DA0001 available, which is a 40G/100G QSFP28 cable. C...

AlohaSpark

It seems like my Netwatch script can't modify a global variable that was created from another script. In my research on this I found that Netwatch runs as another user... that's it. I haven't found any straightforward solutions, only complaints since 2012. https://forum.mikrotik.com/viewtopic.php?p=...

AlohaSpark

I have a bandwidth monitor that changes mangle rules based on traffic amount.

However, the logs are now filled with "mangle rule changed".

Is there any way to disable this specific message?

Time: Mar/04/2023 18:51:13
Buffer: memory
Topics: system, info
Message: mangle rule changed

AlohaSpark

You should turn off loose TCP tracking when you want to burn your CPU and performance. I had it off to begin with. I turned it on 2 days ago. CPU usage did not decrease. Actually, turning on loose TCP tracking seems to have solved my RDP/Remote Desktop issues. The connection doesn't drop anymore (w...

AlohaSpark

I am implementing sticky connections using connection and routing marks. I'm wondering if loose TCP tracking has any effect on mangles.

1. What is it for?
2. When should I turn off loose TCP tracking?
3. Does it impact performance? By how much?

AlohaSpark

Bumping this a bit. Is there something somewhere documenting supported NICs by RouterOS/CHR? We want to see what actually is supported in passthru mode and what type of NIC offloading CHR actually performs. Unfortunately I haven't found any such document. As a small data point, Mellanox ConnectX-3 ...

AlohaSpark

I am trying to figure out why my CCR1009 isn't pushing more than 500Mbps. CPU seems to be lightly loaded according to profiling during peak hours (5-10% total CPU usage, 15% on the highest-loaded core). One advice according to this reply (https://forum.mikrotik.com/viewtopic.php?t=149701#p736982) is...

AlohaSpark

For example, general advice is to have the related/established firewall rule at the very top.

In my case however, I have 20+ firewall rules before that, but all of them are disabled.

Are disabled rules evaluated? Do disabled rules affect performance?

AlohaSpark

I have a CCR1009 that I plan to decommission permanently. Can I convert its L6 license to a P-unlimited CHR license?

AlohaSpark

This test has one important application. It shows the maximum capability of the processor to generate traffic. And if in the UDP everything is of little use in life, the TCP everything is clear. Many users, when trying to perform such tests, get bogged down by the performance of the processor and s...

AlohaSpark

Every time I get big numbers - (UDP) around 80Gbps for my CCR1009 and above 500Gbps for my Xeon E5-2403 server. But what is it testing? How is it significant? How does it translate into the real world? For the above, even though Xeon is more than 5x faster, in production (no FW, just simple cake que...

AlohaSpark

I'm planning to build a VM server then put CHR on it among other things.

1. Would it be better to have hardware interfaces as PCIE passthrough, or bridged via virtio?

2. If it's better to have PCIe passthrough, is there a list somewhere of supported Ethernet/SFP+ cards?

AlohaSpark

I would like to know how long it takes to fetch a URL. Is there a way to do so within a script?

AlohaSpark

Does it matter if I have a parent queue or not in terms of CPU usage?

Would it be better if I have everything (i.e. all clients) as a global queue?

AlohaSpark

Hello, Is there a way to add a comment to a simple queue created by DHCP server? While testing my script, the command /queue simple set dhcp-ds<XX:XX:XX:XX:XX:XX> comment="hello" gave me the error failure: cannot change dynamic I just want to show the host name on the list of queues. A whi...

Search found 48 matches