MTR is a useful tool but it is a somewhat common source of illusory issues since it generates so many icmp time exceeded packets that routers stop replying to other folks running traces. It's important, as others said, to understand that these aren't testing the data path of a network but instead the control plane path.
MTR has long been one of the first little tools that I install on workstations.
sudo apt install mtr-tiny
I also have a hotkey to pop it up in a window, pinging to some host that'll always be somewhere on the other side of any ISP from me. Whenever I suddenly suspect a networking problem from my laptop, I hit the hotkey as the first troubleshooting step. MTR starts to narrow down a few different problems very quickly.
> Is it indicating that the router is faster at forwarding packets than responding to ping requests?
Exactly this. In most “real” routers, forwarding (usually) happens in the “data plane”. It’s handled by an ASIC that has a routing table accessible to it in RAM. A packet comes in on an interface, a routing decision is made, and it goes out another interface - all of this happens with dedicated hardware. Pings (ICMP Echo requests), however, get forwarded by this ASIC to a local CPU, where they are handled by software (in the “control plane”).
You’re really seeing different response times from the two control planes - one may be more loaded or less powerful than another, regardless of the capacity of their data planes.
This is also why you may see packet loss at one particular hop but then responses from hops beyond it. The hop with packet loss in this case probably has an overwhelmed CPU, rather than indicating that a particular network link has packet loss. mtr reporting packet loss at a hop is only reliable if every hop after it has similar packet loss.
Maybe the only thing I've explained more in my career than this is why it's ok that your Linux box has no "free" memory.
Traceroute doesn't use ping requests except with the old Windows binary. Usually it uses "Time-to-live (TTL) exceeded in transit" messages.
Beyond that technicality, your guess is often right... Routers will frequently prioritize forwarding packets over sending the TTL exceeded packets tools like MTR use to measure response times.
Also you can easily have the TTL expired message going via a different route on the return path (and indeed the same applies with your normal connections, asymetric routing can be a pain - especially in networks with rpf issues (multicast ones are a particular pain point), and with stateful firewalls, but most of the time it's fine. You just need to be aware.
Obviously you know, but for anyone else reading, a modern traceroute tool (like mtr) can send icmp, udp or tcp, on generic or specific ports. Indeed the default for mtr on my laptop is to use icmp.
Most likely, it's as you described, router N forwards packets much faster than it generates icmp ttl exceeded, and router N+1 is nearby and generates icmp faster.
However, it could also be the case that the routing back to you is significantly different, so you can have a much longer path to you from router N than router N+1.
This is more likely to happen on routes that cross oceans. Say you're tracing from the US to Brazil. If router N and N+1 are both in Brazil, but N sends return packets through Europe and N+1 sends through Florida, N+1 returns will arrive significantly sooner.
The results aren't misleading, shockingly large numbers of "computer professionals" have no idea how networks work, but that's because they can't use the tools rather than the tools being misleading
>shockingly large numbers of "computer professionals" have no idea how networks work
Incidentally, if you suspect you yourself are this, I can't recommend any book more highly than Michael W. Lucas's Networking for Systems Administrators. Don't be fooled by the title - the whole idea is to get you to the level where you can talk to a network engineer without looking totally clueless, and no farther - an excellent stopping point.
I would recommend it handily over, say, my own Intro to Networking class in college. And yes, `mtr` is mentioned by name in it!
People familiar with networking underestimate how complicated networking actually is. A huge segment of programmers will learn about the existence of routing and BGP and end up in a career where HTTPS and maybe DNS is all they need to worry about.
I'm 100% sure the only reason so many programmers know how NAT works is because NAT breaks video games.
EDIT: Re-Reading. I think I am some degree of a networker underestimating network complexity. I'll stand by that. Please make fun of me for only speaking in IPs and Ports.
Yeh. There is a very achievable level of knowledge about networking that's enough to make a lot of practical problems solvable.
Like, my practically acquired patchwork of knowledge about subnets, routing, some DNS, some VPN tech, maybe some ideas of masquerading and NAT'ing is easily enough to run a multi-site production environment across a number of networking stacks. And I wouldn't really call these things hard. I don't like people who are like "I don't know networking" once you say "routing table". The hardest part there is to understand how things are often a very large amount of very local decisions and a bunch of crossed fingers to get a packet from A to B. Oh an no one thinks about return paths until they run a site to site VPN.
But just a few steps beyond that is a cliff dropping into a terrifying abyss of complexity. LIke I know acronyms like BGP, CGNAT, ideas like Anycast DNS and kinda what they do, but it turns into very dark and different magik rather quickly. I say if we need that, we need a networker.
I once interviewed the manage who built MSN messenger - and when I asked her what the most important thing to the growth was, she said it needed to be able to punch through NATs so kids could use it at high school and uni, because that was the segment they were trying to get it to take off in. (and from what I recall, that strategy indeed worked quite well)
One under-appreciated problem (except from MPLS fudging and multiple load-balancing routers) is that traceroute (including MTR) only shows the way from the sender to the recipient, but actual networks, especially non-peered connections, usually do not use the same paths for both directions. One example that I've encountered is network A sending its packets via then-Telia (now Arelion) but network B routing their packets through NTT instead, which is only shown if you have initiated traceroutes in both directions.
The way you write it makes it seems like you're blaming the tool for misleading results when that's the nature of traceroute itself.
MPLS don't have to hide routers though, up to the operator, even if they do it will give you idea of where things went wrong and you can contact the correct people. Load balancing links is either lacp or ecmp, first case doesn't really matter and in the second you'll just see multiple responses on a hop. Neither really had any impact on how useful traceroute is and doesn't really mislead.
Which is why any network engineer worth their salt with ask for a trace in both directions (if available). Asymmetric routing can be an issue especially when going through stateful devices like firewalls.
I'm not going to tell you how long I've at one time been searching for a missing route on the return path of a VPN connection... But damn the lights that went on when I realized that hurt by being too bright.
It is possible to to reveal the impacts of asymmetric routing through other tools, for instance ThousandEyes can do this by performing a time synchronized bidirectional trace (among other things it can do that MTR cannot). This can be very valuable.
That said, in practice for the majority of end users, they will not be directly impacted by asymmetric routing, if only because so many services are now cloud-based and the major cloud devices are direct peered with all of the major ISPs at regional meeting points in most countries. As an example, on my connection in Denver on Comcast, going to most applications in AWS will enter the AWS network /in Denver/ and without traversing any transit provider, meaning effectively my traffic never goes across "the Internet", it goes from Comcast (my provider) directly to AWS (the provider for the application).
While it's always good to be mindful of the complexities of real-world routing, for the vast majority of common use cases now, entry-points to the target application are so widely distributed that the most impactful routing is inside the private network of the cloud provider, not across the larger Internet.
The packet loss indicator is the biggest issue I have. I’m well aware that routers may deprioritise ICMP and lead to packet loss, and therefore if you’re not seeing cascading packet loss then it’s probably phantom. Also what really matters is end to end loss anyways.
The other issue with packet loss is the tool doesn’t handle ICMP properly in the first place. A ping flood to an end to end host like 1.1.1.1 shows 0% loss, but when I use mtr to do flood like pinging it shows my wifi router with 100% loss. If I ping flood my router I get 0%.
It’s genuinely a bad tool and you should really just be keeping ping and traceroute separate as they do completely different things.
They are only misleading if you allow yourself to be misled by them. It's an extremely informative measurement if you are aware of how it works and don't misinterpret the results.
None of these claims are mutually exclusive with one another.
"Great tool for misleading results." -> the results the tool provides are either mostly misleading (many are misleading), or are in large part misleading (a large part of each is misleading), potentially both
"Traceroute is easy to be misinterpreted" -> the results the tool provides are easy to misinterpret
"They are only misleading if you allow yourself to be misled by them" -> the results the tool provides require expertise to interpret, implying that otherwise they're (largely) misleading - the same thing the person said right above you
This is turning into a "well I like it and it has its place". Cool, it's just not what was being argued.
Yes, you can. It's basically a terminal case of something being unintuitive. Whether something is misleading is in the eye of the beholder.
Recently my mother felt misled by a car commercial. Her position was that saying things like "under this many years or that many miles" is misleading, because it suggests that it's a set of options she can pick from (which of course ended up not being the case).
Unfortunately for her, this is a natural language construct - whether she understands it correctly or not depends on how aligned her common sense regarding it is with people at large. She understood it differently and thus felt misled. But you may notice that ultimately it was her own mistaken understanding of the common parlance that misled her. So when she said this was misleading the only thing I could reasonably say was exactly this. That I did not find the phrasing misleading, and I'm sorry she'd been misled by it (irrespective of whether that was on her or on the world, as that doesn't really matter).
It's completely on people how they want to handle this. You can find people being misled by stuff like this to be unreasonable and just tell them so, or you can put out a disclaimer regardless. Depends completely per case. This goes all the way to having multiple mechanical interlocks at places with heavy duty xray sources, or preferring machine checked memory management.
I don’t trust the ICMP code in mtr. I’ve had an mtr to 1.1.1.1 which shows my wifi router as an intermediate hop showing 100% loss when doing pings at interval of 0.1ms. A flood ping to my router shows 0%. I’d rather just use time tested tools such as ping and trace route, which shouldn’t even be combined anyway since the loss indicator is usually unreliable unless there’s cascading loss (and even then can still be unreliable).
One thing I've not understood is why will some hops have consistently lower ping times than hops farther down the chain in the same trace?
Is it indicating that the router is faster at forwarding packets than responding to ping requests?
https://archive.nanog.org/sites/default/files/traceroute-201...
Exactly this. In most “real” routers, forwarding (usually) happens in the “data plane”. It’s handled by an ASIC that has a routing table accessible to it in RAM. A packet comes in on an interface, a routing decision is made, and it goes out another interface - all of this happens with dedicated hardware. Pings (ICMP Echo requests), however, get forwarded by this ASIC to a local CPU, where they are handled by software (in the “control plane”).
You’re really seeing different response times from the two control planes - one may be more loaded or less powerful than another, regardless of the capacity of their data planes.
Maybe the only thing I've explained more in my career than this is why it's ok that your Linux box has no "free" memory.
Beyond that technicality, your guess is often right... Routers will frequently prioritize forwarding packets over sending the TTL exceeded packets tools like MTR use to measure response times.
Obviously you know, but for anyone else reading, a modern traceroute tool (like mtr) can send icmp, udp or tcp, on generic or specific ports. Indeed the default for mtr on my laptop is to use icmp.
However, it could also be the case that the routing back to you is significantly different, so you can have a much longer path to you from router N than router N+1.
This is more likely to happen on routes that cross oceans. Say you're tracing from the US to Brazil. If router N and N+1 are both in Brazil, but N sends return packets through Europe and N+1 sends through Florida, N+1 returns will arrive significantly sooner.
I believe most of the time this is the reason indeed. Answering an ICMP error to a TTL expiration or to an echo request is very low priority.
This latency in error message generation may even be a better signal of the router load than the latency of the actualy trip through it.
Incidentally, if you suspect you yourself are this, I can't recommend any book more highly than Michael W. Lucas's Networking for Systems Administrators. Don't be fooled by the title - the whole idea is to get you to the level where you can talk to a network engineer without looking totally clueless, and no farther - an excellent stopping point.
I would recommend it handily over, say, my own Intro to Networking class in college. And yes, `mtr` is mentioned by name in it!
I'm 100% sure the only reason so many programmers know how NAT works is because NAT breaks video games.
Yeh. There is a very achievable level of knowledge about networking that's enough to make a lot of practical problems solvable.
Like, my practically acquired patchwork of knowledge about subnets, routing, some DNS, some VPN tech, maybe some ideas of masquerading and NAT'ing is easily enough to run a multi-site production environment across a number of networking stacks. And I wouldn't really call these things hard. I don't like people who are like "I don't know networking" once you say "routing table". The hardest part there is to understand how things are often a very large amount of very local decisions and a bunch of crossed fingers to get a packet from A to B. Oh an no one thinks about return paths until they run a site to site VPN.
But just a few steps beyond that is a cliff dropping into a terrifying abyss of complexity. LIke I know acronyms like BGP, CGNAT, ideas like Anycast DNS and kinda what they do, but it turns into very dark and different magik rather quickly. I say if we need that, we need a networker.
... and filesharing, from the days when bittorrent was huuuuge.
MPLS don't have to hide routers though, up to the operator, even if they do it will give you idea of where things went wrong and you can contact the correct people. Load balancing links is either lacp or ecmp, first case doesn't really matter and in the second you'll just see multiple responses on a hop. Neither really had any impact on how useful traceroute is and doesn't really mislead.
Packet traces do not lie, per se, but they represent only a certain perspective. More perspectives are needed for problems to come into focus.
That said, in practice for the majority of end users, they will not be directly impacted by asymmetric routing, if only because so many services are now cloud-based and the major cloud devices are direct peered with all of the major ISPs at regional meeting points in most countries. As an example, on my connection in Denver on Comcast, going to most applications in AWS will enter the AWS network /in Denver/ and without traversing any transit provider, meaning effectively my traffic never goes across "the Internet", it goes from Comcast (my provider) directly to AWS (the provider for the application).
While it's always good to be mindful of the complexities of real-world routing, for the vast majority of common use cases now, entry-points to the target application are so widely distributed that the most impactful routing is inside the private network of the cloud provider, not across the larger Internet.
Disclaimer: Opinions are my own.
The other issue with packet loss is the tool doesn’t handle ICMP properly in the first place. A ping flood to an end to end host like 1.1.1.1 shows 0% loss, but when I use mtr to do flood like pinging it shows my wifi router with 100% loss. If I ping flood my router I get 0%.
It’s genuinely a bad tool and you should really just be keeping ping and traceroute separate as they do completely different things.
https://movingpackets.net/2017/10/06/misinterpreting-tracero... (discussion at https://news.ycombinator.com/item?id=15474043 )
"Great tool for misleading results." -> the results the tool provides are either mostly misleading (many are misleading), or are in large part misleading (a large part of each is misleading), potentially both
"Traceroute is easy to be misinterpreted" -> the results the tool provides are easy to misinterpret
"They are only misleading if you allow yourself to be misled by them" -> the results the tool provides require expertise to interpret, implying that otherwise they're (largely) misleading - the same thing the person said right above you
This is turning into a "well I like it and it has its place". Cool, it's just not what was being argued.
Recently my mother felt misled by a car commercial. Her position was that saying things like "under this many years or that many miles" is misleading, because it suggests that it's a set of options she can pick from (which of course ended up not being the case).
Unfortunately for her, this is a natural language construct - whether she understands it correctly or not depends on how aligned her common sense regarding it is with people at large. She understood it differently and thus felt misled. But you may notice that ultimately it was her own mistaken understanding of the common parlance that misled her. So when she said this was misleading the only thing I could reasonably say was exactly this. That I did not find the phrasing misleading, and I'm sorry she'd been misled by it (irrespective of whether that was on her or on the world, as that doesn't really matter).
It's completely on people how they want to handle this. You can find people being misled by stuff like this to be unreasonable and just tell them so, or you can put out a disclaimer regardless. Depends completely per case. This goes all the way to having multiple mechanical interlocks at places with heavy duty xray sources, or preferring machine checked memory management.
https://www.youtube.com/watch?v=L0RUI5kHzEQ