8 ms·

Microservices – Combinatorial Explosion of Versions

78 points by taleodor 7y ago

iblaine 7y ago
This is an interesting void for micro services that’s not yet being met.
dgrin91 7y ago
Don't monorepos solve this? Lets presume I'm not a FAANG company (though some of those have monorepos) with a bazillion services, I'm just a startup with say 10 services. At that scale I think its pretty realistic to maintain a monorepo for all 10 services. That way every release only upgrades the version count by one.
- grahamannett 7y ago
  what? i don’t understand how this has anything to do with micro services vs monorepo, if you are using either and you update a particular something what is different between the two
  - simongray 7y ago
    All of the content in a monorepo is versioned together, while microservices have their own independent versioning schemes and for this reason their interfaces can drift.
- taleodor 7y ago
  Thank you for the comment - monorepos indeed make things better to a degree. From one perspective, you now can point your Product (package) to the whole monorepo. But in reality, you're still facing same issue - let's say your components X and Y are updated in monorepo. You want X in production, but not Y - so now you have to invent something on top of monorepo to capture this. Another words back to square 1. Regarding startups, in my experience they start feeling this pain when they already have 5-6 microservices. But YMMV - depends on a lot of factors.
  - dgrin91 7y ago
    Yes, perhaps a better way to think about it is to version the set of services as a whole, and deploy them as a holistic, singular product. Then you don't run into the combinatorics problems.... Just challenges with roll backs.
- gravypod 7y ago
  They do in part but developers need to be aware of two major things: 1. Always be backwards compatible in protocols and data formats. If you store data on disk or in a table you V1 and V2 tables need to be generally compatible. V2 should always be able to read V1 tables. This helps with rollouts. In an ideal case V1 should gracefully handle V2 tables. This helps with rollbacks. 2. Service to service communications work between V1 and V2. If you don't want downtime you have to gracefully handle the situation in which V2 is rolling out. In a zero-downtime deployment config like blue/green V2 instances can get requests from V1 instances and vice versa. Monorepos don't help with these situations. These are just engineering practices that have to be upheld by the team. Monorepos will make the atomic source code updates applied across the codebase easy.
  - thedance 7y ago
    Every place you said "should" is actually "must" otherwise how can anything work? Clearly the next version MUST be able to read the stored data of the current version and just as clearly the next version MUST write data that can be read by the current version. If not then releasing the software is totally impossible.
- akuse 7y ago
  Don't monorepos have an adjacent problem as well? Feature-flags tend to create a combinatorial explosion of potential code-paths in your system.
- jennyyang 7y ago
  A startup has no business running microservices. Microservices are only when the engineering team can't scale properly because they keep stepping on top of each other. A startup should be using a monolith until it becomes a victim of its own success and then should migrate thoughtfully to microservices only when it has to. It also requires a very heavy investment in devops and tools since microservices will fail in production in every which way possible. It also requires a huge investment in metrics and alerts and logging otherwise you will have no idea what is wrong with your now too-complicated production system.
  - danielovichdk 7y ago
    Do you know the domain of the business? Otherwose it sounds like you should reconsider the advice you give out.
    - jennyyang 7y ago
      My statement applies across the board. It doesn't matter what the domain of the business is. Microservices cost more and have a higher impact on productivity and is only worth it if your engineering team needs the flexibility, at the cost of increase devops investment and more down time.
      - danielovichdk 7y ago
        No.
        jennyyang 7y ago
        Ok. Once you have real world experience, then you should continue commenting.
  - speedplane 7y ago
    > A startup has no business running microservices. Microservices are only when the engineering team can't scale properly because they keep stepping on top of each other. While I generally agree with you in the vast majority of practical cases, it can make sense in some circumstances, namely, if there is a very clear logical abstraction between one service and another. A microservice is just another layer of abstraction, similar to functions, classes, files, libraries, and programs, but even higher up the food-chain, and it comes with its own characteristics and peculiarities. If you have high confidence that one component is (1) logically extremely different from all of the other components, (2) has no side effects, (3) is not I/O bound as it's going over a network, and critically, (4) is on a separate development timeline, then you have a solid candidate for a microservice.
- mattlondon 7y ago
  I guess it would only help if the dependencies can be statically analyzed somehow - e.g. a change in X causes a compiler error in Y etc etc - or you have really really really comprehensive integration tests. In my experience this is not always the case and the problems creep in in other places that are not easily statically checked or tested, but which come up at runtime, e.g. specific runtime content/data, or serialisation of some message/object that happens at runtime.
- thedance 7y ago
  No, monorepos don't solve this at all. It's not even related. RPC services are described by IDL and you can't ever break the semantics of the service for any client that was ever using any released version of the IDL. Full stop. The is the golden rule of RPC services and it's also the only rule! You can do anything else, you can completely replace the service with a scratch rewrite in Erlang as long as you don't break the one rule.
- fetbaffe 7y ago
  Monorepo can help you in tracking dependencies between services at build time, however services are still executed independently in production, thus existing running services will have a dependency to older versions. To solve that without versioning you then have to shutdown all of them & wait until all off them are online again with the latest commit of your monorepo. Now you have created a network distributed monolith.
- antoinealb 7y ago
  An issue you might face is that during rollouts, you will still have some servers running the old binary and some running the new binary. So you must either: 1. Stop all old servers before starting the new ones -> downtime, might be acceptable 2. Have your clients handle the fact that they might hit vN or vN+1 of your services.
sudeepj 7y ago
Few points based on my experiences with (Micro) services: Note: None of these are cast in stone. These have worked well for us 1. Decide in your team/org what constitutes a new version and what it means. Does every change means version change? What if an API adds an optional parameter; do I need to update the version? 2. Not everything has to be a service. Think if the same functionality can be consumed in the form of a library/sdk instead of a full service. 3. The dependency amongst services should resemble acyclic graph and not cycles. This limits the version change impact. 4. Think about abstraction. Can we logically group set of services A, B & C and provide unified API via pass-through service D? Only D needs to handle API breakage most of the time.
bethesque 7y ago
I was at realestate.com.au when they started writing microservices in 2013 and we were all facing this exact issue. Out of that experience, the (now open source) contract testing tool called "Pact" was written (I am now one of the maintainers). I wrote a blog about the combinatorial explosion problem here many years ago now! https://www.rea-group.com/blog/enter-the-pact-matrix-or-how-to-decouple-the-release-cycles-of-your-microservices/ https://www.rea-group.com/blog/enter-the-pact-matrix-or-how-... From the Pact docs: 'Contract testing is a technique for testing an integration point by checking each application in isolation to ensure the messages it sends or receives conform to a shared understanding that is documented in a "contract".' By focussing just on the messages we get tests which are fast, give us quick feedback, and scale linearly instead of combinatorially. Some good resources are: https://pact.io https://pact.io (for information about contract testing and the Pact tool itself) https://pactflow.io/how-pact-works/ https://pactflow.io/how-pact-works/ (explains how Pact works) https://docs.pact.io/faq/convinceme https://docs.pact.io/faq/convinceme (answers the question of why you would want to do contract testing) https://slack.pact.io https://slack.pact.io (a friendly 1000+ member community which is very experienced dealing with these kinds of issues) https://docs.pact.io/pact_broker/can_i_deploy https://docs.pact.io/pact_broker/can_i_deploy (addresses how we handle and channel this combinatorial explosion for good instead of evil!)
- hinkley 7y ago
  I did a stupid simple thing for a system with a small fanout footprint. Not sure this scales. The organizing principal was that if my code passed the acceptance tests once, the very same code should still pass with your service/library/data added to the mix. If they don’t, it’s likely your code and not mine. This might not seem like much but it cuts hours of finger pointing out of a three party system. I expect this effect would be multiplied for a five or ten party scenario. And in particular, latency has a huge negative impact on behavioral modification. People will keep doing things that they get yelled at over three weeks later. They will stop doing things they get called on hours later. Which I would hope offsets the additional test matrix complexity.
- taleodor 7y ago
  Thank you, great read and Pact looks interesting.
bcrosby95 7y ago
Consumer driven tests can help with this too. Rather than just service providers writing tests, service consumers write them too, which are then given to the service provider as part of their test suite.
simonw 7y ago
A lesson I learned working with microservices that are deployed independently of each other is that you need a policy that ALL changes must be backwards compatible with existing clients. This sounds crazy, but it's essential if you can't guarantee that you can atomically deploy a new service version AND any other services that call it. Creating backwards compatible releases means sticking to some rules, things like: - you can add fields but you can never delete them - you can add new methods but you can't remove old ones Having really detailed logging helps a lot. If you want to deprecate a method you can remove it from calling clients first and then use the logs to confirm it isn't being called any more before removing it from the service. It's a lot if work, and teaching a large engineering team how to be productive in this kind of environment is decidedly non trivial.
- daotoad 7y ago
  Never say never. Adding fields or methods is easy. You can do it at any time. Removing them is difficult. You must first go through a long deprecation cycle and ensure ALL clients are updated. You need good logging AND good monitoring and analysis of the logs in both client and server to be able to detect access to removed features.
  - simonw 7y ago
    A coworker pointed out that this is a huge point in favour of GraphQL: because it forces clients to explicitly ask for every field you can log those queries and use them to spot when a field is no longer in use.
- zubspace 7y ago
  Actually, if the server knows the client version, he can downgrade returned data to the client in the version he expects. That way, adding, renaming and removing fields should not be a showstopper, assuming methods stay as generic as possible. Alternatively you could provide different API versions at the same time and keep the old ones running, as long as they are called.
  - 7y ago
TeeWEE 7y ago
if you: Deploy new versions of new services and quickly rollout the number of combinations get manageable. Making he interface of services backwards compatible, either by testing and/or by using stricter contracts like grpc (which had a way to deal with compatibility) Then you're mostly covered. However integration testing indeed makes sense
atrandom 7y ago
in my experience, most service API grow by accretion and need to break backward compatibility very very rarely. Naturally, clients should always be Tolerant Readers. So in many case, you will have only a single version per service. Moreover, imo, when we are talking about internal apis, there should be nothing preventing the service owner from updating the consumers if he wishes to converge more quickly after breaking compatibility - just like he would when refactoring a monolith. The culture should allow and encourage this kind of collaboration.
jennyyang 7y ago
The diagram of the service in the blog post is absurd. Anyone who designs a microservice environment like this should be fired. The interdependencies between services in the picture make the entire system fragile and everything will fail if a single service goes down. A real production microservice environment should be designed so that interdependencies are limited and system wide failures won't occur if a service or two go down. Once you limit the interdependencies then the combinatorial explosion doesn't exist anymore. You might have some services that have a wide range of version used by clients, but you don't have the interdependency that complicates things. Then, versioning isn't such a big deal.
- rapsey 7y ago
  Unfortunately however most developers (even senior ones) follow fads without proper homework and planning.
- danielovichdk 7y ago
  You cannot build software without dependencies. The dependency pattern around microservices is either a contract on a queue/stream/topic or a version of an API from another depened service. Don't care how many times you tell me that services should be interdepended and bla bla bla... In the end a service has dependencies and those dependencies must then have a versioning strategy so humans are not in doubt. Versioning a microservice is no different than versioning a linked library or a Rest API. Documentation is key. Tell people what versions break.
  - jennyyang 7y ago
    You didn't understand what I wrote. I said that having interdependent services the way the diagram was drawn leads to an extremely fragile system. If one service goes down, they all go down. It's not a realistic environment or it's a very poorly designed one. If you simplify the diagram to one that is more realistic, with more 1:1 or N:1 relationships, then the combinatorial factor decreases, and then there is no article to write about.
    - 7y ago
mattlondon 7y ago
The typical approach to catching this that I have seen is to deploy the new version of Service X to only a small percentage of all machines for a certain amount of time. E.g. 1% of machines get v1.1 while all others stay on v1.0. You then monitor error rates in v1.1 (and its dependencies) and stop if there are elevated levels, or crank the release up slowly if the error levels aren't elevated. Obviously some automation in your release management and monitoring tools helps a lot with this. Of course this does not "solve" the problem of incompatibilities themselves - for that the simple solution that I've seen used is just bog standard versioning of APIs (/v1/... /v2/... /v2019-03-04/... etc) and some procedures for managing the supported versions, e.g. not doing breaking changes in an existing version, only supporting a set number of versions (i.e. previous, current, and next), and having proper sunset-periods for older versions before they are turned off so clients have time to update. Slows velocity to do all this management but that is life anyway in a complex system
fxtentacle 7y ago
Turns out the latest hype is a dud for most companies ^_^ There are good reasons to use microservices, like when you have 100+ engineers working on your planet-scale cloud or social network. At that scale, you need to have internal API documentation to ensure people that never met can still work together productively. But if you are a normal company, you will probably never have enough engineers to make microservices worthwhile in the first place. So just put everything into one big git repo and call it a day. Also, the ability to do rolling restarts and no downtime upgrades by running multiple versions in parallel adds a lot of work and complexity. But for most small to medium companies, a planned 5 minute downtime in the middle of the night is completely no problem, so all that zero-downtime-upgrade work is just wasted effort. I mean, even my bank has a fixed offline maintenance window every night from 3:00 to 3:30 am. As does Amazon RDS. So don't solve problems that you don't have :) Most companies do fine without microservices.
- kqr 7y ago
  > But for most small to medium companies, a planned 5 minute downtime in the middle of the night is completely no problem, so all that zero-downtime-upgrade work is just wasted effort. The first half of that is true, but the conclusion overlooks a couple of much bigger issues: 1. There are few indicators of effective software engineering so strong as deployment frequency and lead time for changes. Being able to go from issue to fix in production in minutes rather than days opens up completely new paths for feedback and allows you to operate your service much more efficiently. (See the yearly state of devops reports for further research about this.) 2. Not all downtime is planned downtime. Being able to handle a true crash gracefully is very important to achieve reasonable response times in the high nines. I'm not a proponent of service-oriented architectures -- my opinion is that enforcing API boundaries is useful, but can be done at the source level rather than application level. I do however think there is great value to being able to deploy/restart your application with virtually no downtime.
  - funcDropShadow 7y ago
    > 1. There are few indicators of effective software engineering so strong as deployment frequency and lead time for changes. Being able to go from issue to fix in production in minutes rather than days opens up completely new paths for feedback and allows you to operate your service much more efficiently This is only true under certain assumptions. E.g. if your business is not a public facing website, you often don't have the immediate feedback in case of errors, to make the above observation applicable. I have a customer with an application that gather date for a whole year, and then does alot of stuff with it. Then, there are no short feedback cycles in production, no matter how often a day you deploy. Then, it is important to assure quality before a change reaches production. Otherwise it might replace your data with background noise and you only discover it at the end of the next year. This is a symptom of the general tendency in developer circles to assume that personal findings apply to everyone at every time in every situation. Which is obviously ridicoulous. But somehow completely acceptable if you generalize from something one Google employer has posted about.
jillesvangurp 7y ago
Microservices only have two versions you should worry about: the one that's currently live and the next one. Two for each micro service you are responsible for. IMHO roll backs are not a thing. Instead git revert and roll forward when you need to deploy a fix. So once a service has been fully deployed, you permanently stop caring about the previous version. Microservices should not be mass deployed together but be managed separately. That vastly simplifies the combinatorial explosion of versions you need the worry about: the currently live ones. The rest is just applying SOLID principles to your microservices and avoid having services with poor cohesiveness or tight coupling (most of the SOLID principles boil down to affecting these two metrics). Bad service design with tight coupling where deploying service A also requires redeploying B,C,and D, is basically a design problem and not a micro services problem. There's a lot of bad design in our industry. Monoliths allow you to get away with that but it's also the reason that breaking them up is a hard problem. Just because you are getting away with it does not mean it is not a problem though. You can further mitigate integration issues by deploying using modern practices like blue green deployments where you gradually move traffic to a new versions and adapt on things like error rates and other metrics, AB testing, etc. So if it breaks, you don't end up breaking it for everyone and you can fix the problem and try deploying that in a controlled way. Basically your goal is to keep your customers (aka. dependees) happy and make sure you don't negatively affect them (which you should be actively monitoring). Likewise if you have dependencies (to whom you are a customer) and they stop working when you deploy something new, you probably want to detect and fix that before you break all your customers. If that is a regular thing, consider having integration tests, contract tests, etc. Staging environments are a controversial topic in this context. My view is that they don't make much sense in a properly run micro service deployment since you are not testing in a real environment with real users, real data, and lots of things happening concurrently and features interacting. And of course your customers doing real things they care about. The only realistic environment that has that in most complex micro service deployments is called production. Once you add serverless and edge computing to the mix, these things become even more true. The bigger the organization, the less feasible it becomes to have a staging environment. Update. I forgot to add this but doing continuous deployment means small deltas that are low risk. Any bigger change can be behind a feature flag. There are a few more strategies. Also, I'm not actually a microservices proponent for small teams. It just creates deployment and operational overhead (read go to market bottlenecks).
pachico 7y ago
Although it's an interesting calculation, you rarely have all your microservices communicating with eachother since you tend to group by areas. And even if that was the case, if good practices are in place (semantic versioning, good test cases, etc) I don't see why that should be a problem.
jayd16 7y ago
Is this a real problem people deal with? People actually build these spaghetti services and deploy several versions of them at once? Each of these versions are somehow routable to and from each other version with no abstraction? Surely this doesn't actually happen.
ewindal 7y ago
I don’t see how this is a problem. I see that the combinatorial explosion exists, but that isn’t an actual problem. The entire point of microservices is independent deploys, meaning you literally only have two permutations per deploy, new and old. If you think deploying 10 services at once is feasible, you don’t understand the appral of microservices, and should never have opted for them in the first place.
- taleodor 7y ago
  Deploying one at a time may be fine if you always test that each one is working in backward compatible manner. However you still have same search space and same problem as before. Imagine a - you have 10 microservices, 1 update per each. Each supposed to be backward-compatible. You start rolling out. One by one. 5 go fine, 6th breaks. You end up in a weird state where 5 out of 10 are updated. You hope it's fine due to backward compatibility but you never really tested this config exactly. (Which is why I might prefer either converging to deploying all 10 or rolling back fully - if that was a known good state - and having canary cluster rather than canary microservice in many cases). b - same as above but now you try to catch this behaviour on test / staging. You still have same hard problem at hands. Key here is you clearly can't try every possible variation of what may break, so need to make conscious decisions about what to do in the case of failure.
BerislavLopac 7y ago
If you have this problem, you don't have microservices -- you have a distributed monolith. If each service can talk to each other, the number of possible communication paths grows exponentially -- this is manageable up to a point, but even the largest teams will have to start investing into orchestration sooner or later. There are many orchestration patterns and mechanisms, and this is not the place to list them all -- which one is the right choice will depend on the complexity and needs of the system.
- taleodor 7y ago
  This is not about microservices talking to each other, but rather about seeing customer-facing product as a whole with all its different components and their versions. And yes we can agree that there are various (usually "greedy") strategies to simplify this - but this problem is a fact of life and it makes thinking process easier to accept it as such. Same as accepting that it is impossible to solve consensus in async system, but there are algorithms that perform fairly well. + I summed this up in more details in the comments down below.

78 points by taleodor 7y ago

This is an interesting void for micro services that’s not yet being met.

Don't monorepos solve this? Lets presume I'm not a FAANG company (though some of those have monorepos) with a bazillion services, I'm just a startup with say 10 services. At that scale I think its pretty realistic to maintain a monorepo for all 10 services. That way every release only upgrades the version count by one.

grahamannett 7y ago
what? i don’t understand how this has anything to do with micro services vs monorepo, if you are using either and you update a particular something what is different between the two
- simongray 7y ago
  All of the content in a monorepo is versioned together, while microservices have their own independent versioning schemes and for this reason their interfaces can drift.
taleodor 7y ago
Thank you for the comment - monorepos indeed make things better to a degree. From one perspective, you now can point your Product (package) to the whole monorepo. But in reality, you're still facing same issue - let's say your components X and Y are updated in monorepo. You want X in production, but not Y - so now you have to invent something on top of monorepo to capture this. Another words back to square 1. Regarding startups, in my experience they start feeling this pain when they already have 5-6 microservices. But YMMV - depends on a lot of factors.
- dgrin91 7y ago
  Yes, perhaps a better way to think about it is to version the set of services as a whole, and deploy them as a holistic, singular product. Then you don't run into the combinatorics problems.... Just challenges with roll backs.
gravypod 7y ago
They do in part but developers need to be aware of two major things: 1. Always be backwards compatible in protocols and data formats. If you store data on disk or in a table you V1 and V2 tables need to be generally compatible. V2 should always be able to read V1 tables. This helps with rollouts. In an ideal case V1 should gracefully handle V2 tables. This helps with rollbacks. 2. Service to service communications work between V1 and V2. If you don't want downtime you have to gracefully handle the situation in which V2 is rolling out. In a zero-downtime deployment config like blue/green V2 instances can get requests from V1 instances and vice versa. Monorepos don't help with these situations. These are just engineering practices that have to be upheld by the team. Monorepos will make the atomic source code updates applied across the codebase easy.
- thedance 7y ago
  Every place you said "should" is actually "must" otherwise how can anything work? Clearly the next version MUST be able to read the stored data of the current version and just as clearly the next version MUST write data that can be read by the current version. If not then releasing the software is totally impossible.
akuse 7y ago
Don't monorepos have an adjacent problem as well? Feature-flags tend to create a combinatorial explosion of potential code-paths in your system.
jennyyang 7y ago
A startup has no business running microservices. Microservices are only when the engineering team can't scale properly because they keep stepping on top of each other. A startup should be using a monolith until it becomes a victim of its own success and then should migrate thoughtfully to microservices only when it has to. It also requires a very heavy investment in devops and tools since microservices will fail in production in every which way possible. It also requires a huge investment in metrics and alerts and logging otherwise you will have no idea what is wrong with your now too-complicated production system.
- danielovichdk 7y ago
  Do you know the domain of the business? Otherwose it sounds like you should reconsider the advice you give out.
  - jennyyang 7y ago
    My statement applies across the board. It doesn't matter what the domain of the business is. Microservices cost more and have a higher impact on productivity and is only worth it if your engineering team needs the flexibility, at the cost of increase devops investment and more down time.
    - danielovichdk 7y ago
      No.
      - jennyyang 7y ago
        Ok. Once you have real world experience, then you should continue commenting.
- speedplane 7y ago
  > A startup has no business running microservices. Microservices are only when the engineering team can't scale properly because they keep stepping on top of each other. While I generally agree with you in the vast majority of practical cases, it can make sense in some circumstances, namely, if there is a very clear logical abstraction between one service and another. A microservice is just another layer of abstraction, similar to functions, classes, files, libraries, and programs, but even higher up the food-chain, and it comes with its own characteristics and peculiarities. If you have high confidence that one component is (1) logically extremely different from all of the other components, (2) has no side effects, (3) is not I/O bound as it's going over a network, and critically, (4) is on a separate development timeline, then you have a solid candidate for a microservice.
mattlondon 7y ago
I guess it would only help if the dependencies can be statically analyzed somehow - e.g. a change in X causes a compiler error in Y etc etc - or you have really really really comprehensive integration tests. In my experience this is not always the case and the problems creep in in other places that are not easily statically checked or tested, but which come up at runtime, e.g. specific runtime content/data, or serialisation of some message/object that happens at runtime.
thedance 7y ago
No, monorepos don't solve this at all. It's not even related. RPC services are described by IDL and you can't ever break the semantics of the service for any client that was ever using any released version of the IDL. Full stop. The is the golden rule of RPC services and it's also the only rule! You can do anything else, you can completely replace the service with a scratch rewrite in Erlang as long as you don't break the one rule.
fetbaffe 7y ago
Monorepo can help you in tracking dependencies between services at build time, however services are still executed independently in production, thus existing running services will have a dependency to older versions. To solve that without versioning you then have to shutdown all of them & wait until all off them are online again with the latest commit of your monorepo. Now you have created a network distributed monolith.
antoinealb 7y ago
An issue you might face is that during rollouts, you will still have some servers running the old binary and some running the new binary. So you must either: 1. Stop all old servers before starting the new ones -> downtime, might be acceptable 2. Have your clients handle the fact that they might hit vN or vN+1 of your services.

sudeepj 7y ago

Few points based on my experiences with (Micro) services: Note: None of these are cast in stone. These have worked well for us 1. Decide in your team/org what constitutes a new version and what it means. Does every change means version change? What if an API adds an optional parameter; do I need to update the version? 2. Not everything has to be a service. Think if the same functionality can be consumed in the form of a library/sdk instead of a full service. 3. The dependency amongst services should resemble acyclic graph and not cycles. This limits the version change impact. 4. Think about abstraction. Can we logically group set of services A, B & C and provide unified API via pass-through service D? Only D needs to handle API breakage most of the time.

bethesque 7y ago

I was at realestate.com.au when they started writing microservices in 2013 and we were all facing this exact issue. Out of that experience, the (now open source) contract testing tool called "Pact" was written (I am now one of the maintainers). I wrote a blog about the combinatorial explosion problem here many years ago now! https://www.rea-group.com/blog/enter-the-pact-matrix-or-how-to-decouple-the-release-cycles-of-your-microservices/ https://www.rea-group.com/blog/enter-the-pact-matrix-or-how-... From the Pact docs: 'Contract testing is a technique for testing an integration point by checking each application in isolation to ensure the messages it sends or receives conform to a shared understanding that is documented in a "contract".' By focussing just on the messages we get tests which are fast, give us quick feedback, and scale linearly instead of combinatorially. Some good resources are: https://pact.io https://pact.io (for information about contract testing and the Pact tool itself) https://pactflow.io/how-pact-works/ https://pactflow.io/how-pact-works/ (explains how Pact works) https://docs.pact.io/faq/convinceme https://docs.pact.io/faq/convinceme (answers the question of why you would want to do contract testing) https://slack.pact.io https://slack.pact.io (a friendly 1000+ member community which is very experienced dealing with these kinds of issues) https://docs.pact.io/pact_broker/can_i_deploy https://docs.pact.io/pact_broker/can_i_deploy (addresses how we handle and channel this combinatorial explosion for good instead of evil!)

hinkley 7y ago
I did a stupid simple thing for a system with a small fanout footprint. Not sure this scales. The organizing principal was that if my code passed the acceptance tests once, the very same code should still pass with your service/library/data added to the mix. If they don’t, it’s likely your code and not mine. This might not seem like much but it cuts hours of finger pointing out of a three party system. I expect this effect would be multiplied for a five or ten party scenario. And in particular, latency has a huge negative impact on behavioral modification. People will keep doing things that they get yelled at over three weeks later. They will stop doing things they get called on hours later. Which I would hope offsets the additional test matrix complexity.
taleodor 7y ago
Thank you, great read and Pact looks interesting.

bcrosby95 7y ago

Consumer driven tests can help with this too. Rather than just service providers writing tests, service consumers write them too, which are then given to the service provider as part of their test suite.

simonw 7y ago

A lesson I learned working with microservices that are deployed independently of each other is that you need a policy that ALL changes must be backwards compatible with existing clients. This sounds crazy, but it's essential if you can't guarantee that you can atomically deploy a new service version AND any other services that call it. Creating backwards compatible releases means sticking to some rules, things like: - you can add fields but you can never delete them - you can add new methods but you can't remove old ones Having really detailed logging helps a lot. If you want to deprecate a method you can remove it from calling clients first and then use the logs to confirm it isn't being called any more before removing it from the service. It's a lot if work, and teaching a large engineering team how to be productive in this kind of environment is decidedly non trivial.

daotoad 7y ago
Never say never. Adding fields or methods is easy. You can do it at any time. Removing them is difficult. You must first go through a long deprecation cycle and ensure ALL clients are updated. You need good logging AND good monitoring and analysis of the logs in both client and server to be able to detect access to removed features.
- simonw 7y ago
  A coworker pointed out that this is a huge point in favour of GraphQL: because it forces clients to explicitly ask for every field you can log those queries and use them to spot when a field is no longer in use.
zubspace 7y ago
Actually, if the server knows the client version, he can downgrade returned data to the client in the version he expects. That way, adding, renaming and removing fields should not be a showstopper, assuming methods stay as generic as possible. Alternatively you could provide different API versions at the same time and keep the old ones running, as long as they are called.
- 7y ago

TeeWEE 7y ago

if you: Deploy new versions of new services and quickly rollout the number of combinations get manageable. Making he interface of services backwards compatible, either by testing and/or by using stricter contracts like grpc (which had a way to deal with compatibility) Then you're mostly covered. However integration testing indeed makes sense

atrandom 7y ago

in my experience, most service API grow by accretion and need to break backward compatibility very very rarely. Naturally, clients should always be Tolerant Readers. So in many case, you will have only a single version per service. Moreover, imo, when we are talking about internal apis, there should be nothing preventing the service owner from updating the consumers if he wishes to converge more quickly after breaking compatibility - just like he would when refactoring a monolith. The culture should allow and encourage this kind of collaboration.

jennyyang 7y ago

The diagram of the service in the blog post is absurd. Anyone who designs a microservice environment like this should be fired. The interdependencies between services in the picture make the entire system fragile and everything will fail if a single service goes down. A real production microservice environment should be designed so that interdependencies are limited and system wide failures won't occur if a service or two go down. Once you limit the interdependencies then the combinatorial explosion doesn't exist anymore. You might have some services that have a wide range of version used by clients, but you don't have the interdependency that complicates things. Then, versioning isn't such a big deal.

rapsey 7y ago
Unfortunately however most developers (even senior ones) follow fads without proper homework and planning.
danielovichdk 7y ago
You cannot build software without dependencies. The dependency pattern around microservices is either a contract on a queue/stream/topic or a version of an API from another depened service. Don't care how many times you tell me that services should be interdepended and bla bla bla... In the end a service has dependencies and those dependencies must then have a versioning strategy so humans are not in doubt. Versioning a microservice is no different than versioning a linked library or a Rest API. Documentation is key. Tell people what versions break.
- jennyyang 7y ago
  You didn't understand what I wrote. I said that having interdependent services the way the diagram was drawn leads to an extremely fragile system. If one service goes down, they all go down. It's not a realistic environment or it's a very poorly designed one. If you simplify the diagram to one that is more realistic, with more 1:1 or N:1 relationships, then the combinatorial factor decreases, and then there is no article to write about.
  - 7y ago

mattlondon 7y ago

The typical approach to catching this that I have seen is to deploy the new version of Service X to only a small percentage of all machines for a certain amount of time. E.g. 1% of machines get v1.1 while all others stay on v1.0. You then monitor error rates in v1.1 (and its dependencies) and stop if there are elevated levels, or crank the release up slowly if the error levels aren't elevated. Obviously some automation in your release management and monitoring tools helps a lot with this. Of course this does not "solve" the problem of incompatibilities themselves - for that the simple solution that I've seen used is just bog standard versioning of APIs (/v1/... /v2/... /v2019-03-04/... etc) and some procedures for managing the supported versions, e.g. not doing breaking changes in an existing version, only supporting a set number of versions (i.e. previous, current, and next), and having proper sunset-periods for older versions before they are turned off so clients have time to update. Slows velocity to do all this management but that is life anyway in a complex system

fxtentacle 7y ago

Turns out the latest hype is a dud for most companies ^_^ There are good reasons to use microservices, like when you have 100+ engineers working on your planet-scale cloud or social network. At that scale, you need to have internal API documentation to ensure people that never met can still work together productively. But if you are a normal company, you will probably never have enough engineers to make microservices worthwhile in the first place. So just put everything into one big git repo and call it a day. Also, the ability to do rolling restarts and no downtime upgrades by running multiple versions in parallel adds a lot of work and complexity. But for most small to medium companies, a planned 5 minute downtime in the middle of the night is completely no problem, so all that zero-downtime-upgrade work is just wasted effort. I mean, even my bank has a fixed offline maintenance window every night from 3:00 to 3:30 am. As does Amazon RDS. So don't solve problems that you don't have :) Most companies do fine without microservices.

kqr 7y ago
> But for most small to medium companies, a planned 5 minute downtime in the middle of the night is completely no problem, so all that zero-downtime-upgrade work is just wasted effort. The first half of that is true, but the conclusion overlooks a couple of much bigger issues: 1. There are few indicators of effective software engineering so strong as deployment frequency and lead time for changes. Being able to go from issue to fix in production in minutes rather than days opens up completely new paths for feedback and allows you to operate your service much more efficiently. (See the yearly state of devops reports for further research about this.) 2. Not all downtime is planned downtime. Being able to handle a true crash gracefully is very important to achieve reasonable response times in the high nines. I'm not a proponent of service-oriented architectures -- my opinion is that enforcing API boundaries is useful, but can be done at the source level rather than application level. I do however think there is great value to being able to deploy/restart your application with virtually no downtime.
- funcDropShadow 7y ago
  > 1. There are few indicators of effective software engineering so strong as deployment frequency and lead time for changes. Being able to go from issue to fix in production in minutes rather than days opens up completely new paths for feedback and allows you to operate your service much more efficiently This is only true under certain assumptions. E.g. if your business is not a public facing website, you often don't have the immediate feedback in case of errors, to make the above observation applicable. I have a customer with an application that gather date for a whole year, and then does alot of stuff with it. Then, there are no short feedback cycles in production, no matter how often a day you deploy. Then, it is important to assure quality before a change reaches production. Otherwise it might replace your data with background noise and you only discover it at the end of the next year. This is a symptom of the general tendency in developer circles to assume that personal findings apply to everyone at every time in every situation. Which is obviously ridicoulous. But somehow completely acceptable if you generalize from something one Google employer has posted about.

jillesvangurp 7y ago

Microservices only have two versions you should worry about: the one that's currently live and the next one. Two for each micro service you are responsible for. IMHO roll backs are not a thing. Instead git revert and roll forward when you need to deploy a fix. So once a service has been fully deployed, you permanently stop caring about the previous version. Microservices should not be mass deployed together but be managed separately. That vastly simplifies the combinatorial explosion of versions you need the worry about: the currently live ones. The rest is just applying SOLID principles to your microservices and avoid having services with poor cohesiveness or tight coupling (most of the SOLID principles boil down to affecting these two metrics). Bad service design with tight coupling where deploying service A also requires redeploying B,C,and D, is basically a design problem and not a micro services problem. There's a lot of bad design in our industry. Monoliths allow you to get away with that but it's also the reason that breaking them up is a hard problem. Just because you are getting away with it does not mean it is not a problem though. You can further mitigate integration issues by deploying using modern practices like blue green deployments where you gradually move traffic to a new versions and adapt on things like error rates and other metrics, AB testing, etc. So if it breaks, you don't end up breaking it for everyone and you can fix the problem and try deploying that in a controlled way. Basically your goal is to keep your customers (aka. dependees) happy and make sure you don't negatively affect them (which you should be actively monitoring). Likewise if you have dependencies (to whom you are a customer) and they stop working when you deploy something new, you probably want to detect and fix that before you break all your customers. If that is a regular thing, consider having integration tests, contract tests, etc. Staging environments are a controversial topic in this context. My view is that they don't make much sense in a properly run micro service deployment since you are not testing in a real environment with real users, real data, and lots of things happening concurrently and features interacting. And of course your customers doing real things they care about. The only realistic environment that has that in most complex micro service deployments is called production. Once you add serverless and edge computing to the mix, these things become even more true. The bigger the organization, the less feasible it becomes to have a staging environment. Update. I forgot to add this but doing continuous deployment means small deltas that are low risk. Any bigger change can be behind a feature flag. There are a few more strategies. Also, I'm not actually a microservices proponent for small teams. It just creates deployment and operational overhead (read go to market bottlenecks).

pachico 7y ago

Although it's an interesting calculation, you rarely have all your microservices communicating with eachother since you tend to group by areas. And even if that was the case, if good practices are in place (semantic versioning, good test cases, etc) I don't see why that should be a problem.

jayd16 7y ago

Is this a real problem people deal with? People actually build these spaghetti services and deploy several versions of them at once? Each of these versions are somehow routable to and from each other version with no abstraction? Surely this doesn't actually happen.

ewindal 7y ago

I don’t see how this is a problem. I see that the combinatorial explosion exists, but that isn’t an actual problem. The entire point of microservices is independent deploys, meaning you literally only have two permutations per deploy, new and old. If you think deploying 10 services at once is feasible, you don’t understand the appral of microservices, and should never have opted for them in the first place.

taleodor 7y ago
Deploying one at a time may be fine if you always test that each one is working in backward compatible manner. However you still have same search space and same problem as before. Imagine a - you have 10 microservices, 1 update per each. Each supposed to be backward-compatible. You start rolling out. One by one. 5 go fine, 6th breaks. You end up in a weird state where 5 out of 10 are updated. You hope it's fine due to backward compatibility but you never really tested this config exactly. (Which is why I might prefer either converging to deploying all 10 or rolling back fully - if that was a known good state - and having canary cluster rather than canary microservice in many cases). b - same as above but now you try to catch this behaviour on test / staging. You still have same hard problem at hands. Key here is you clearly can't try every possible variation of what may break, so need to make conscious decisions about what to do in the case of failure.

BerislavLopac 7y ago

If you have this problem, you don't have microservices -- you have a distributed monolith. If each service can talk to each other, the number of possible communication paths grows exponentially -- this is manageable up to a point, but even the largest teams will have to start investing into orchestration sooner or later. There are many orchestration patterns and mechanisms, and this is not the place to list them all -- which one is the right choice will depend on the complexity and needs of the system.

taleodor 7y ago
This is not about microservices talking to each other, but rather about seeing customer-facing product as a whole with all its different components and their versions. And yes we can agree that there are various (usually "greedy") strategies to simplify this - but this problem is a fact of life and it makes thinking process easier to accept it as such. Same as accepting that it is impossible to solve consensus in async system, but there are algorithms that perform fairly well. + I summed this up in more details in the comments down below.