3 ms·

A Kubernetes Query Language

fatliverfreddy 2y ago
Learn a new language to unlock the power of the Kubernetes graph. Query multiple resource kinds and compose custom JSON payloads Express complex CRUD operations in a compact, effective new way. Full-featured interactive shell includes completion, syntax highlighting, resource graph visualization in ASCII-art and more.
- zactato 2y ago
  I should know this, but does K8s actually store objects as a graph internally? Is this doing multiple api calls? How does it perform at scale?
  - fatliverfreddy 2y ago
    K8s does not store any graph model internally. The relationship engine in Cyphernetes is comparing fields that should either match/contain all/etc. between two resources - for example a ResplicaSet's .metadata.ownerReferences[].name matches a Deployment's .metadata.name, or a Deployment's .metadata.labels contains all of a Service's spec.selector. These rules are used both to read the resources but also for creating them. Each rule may also contain a set of default properties which are used in create operations - for instance: `MATCH (d:Deployment {name: "nginx"}) CREATE (d)->(s:Service)` Will create a clusterIp Service (similar to what `kubectl expose` does). Doing this with kubectl or programmatically without tuning for rate limiting, implementing caching etc. of course fails even at a very small scale - Cyphernetes has its own request throttling and caching layer to make this possible.
zaps 2y ago
Phern is terrifying.
- fatliverfreddy 2y ago
  It seems terrifying at first but once you get to know it you’ll get over it. It’s super cuddly and really into graphs.
  - muixoozie 2y ago
    Reminds me of fizzgig https://www.youtube.com/watch?v=lSnzoeZMljM https://www.youtube.com/watch?v=lSnzoeZMljM
    - fatliverfreddy 2y ago
      Oh I loved The Dark Crystal! Forgot about that little guy. Will use as inspiration for the next iteration :)
zug_zug 2y ago
If you want to expose kubernetes by a query language, why not expose an SQL interface?
- fatliverfreddy 2y ago
  I personally am a fan of Cypher and a Cypher-inspired language to describe graph operations in Kubernetes had been a dream of mine for several years until I eventually mustered the courage to create it myself. I just think it’s a great fit, much more than SQL! I’m aware picking up a new language will be a barrier for most to pick it up, and that’s fine. In any case I think this is an interesting tool worth sharing and I hope to find a few more that will be into it, who knows! While doing the research for this tool I did however come across two projects that do exactly what you describe: kubeql: https://github.com/saracen/kubeql https://github.com/saracen/kubeql kubesql: https://github.com/xuxinkun/kubesql https://github.com/xuxinkun/kubesql I’m not sure if either graduated past doing read operations, and both aren’t maintained anymore.

Learn a new language to unlock the power of the Kubernetes graph. Query multiple resource kinds and compose custom JSON payloads Express complex CRUD operations in a compact, effective new way. Full-featured interactive shell includes completion, syntax highlighting, resource graph visualization in ASCII-art and more.

zactato 2y ago
I should know this, but does K8s actually store objects as a graph internally? Is this doing multiple api calls? How does it perform at scale?
- fatliverfreddy 2y ago
  K8s does not store any graph model internally. The relationship engine in Cyphernetes is comparing fields that should either match/contain all/etc. between two resources - for example a ResplicaSet's .metadata.ownerReferences[].name matches a Deployment's .metadata.name, or a Deployment's .metadata.labels contains all of a Service's spec.selector. These rules are used both to read the resources but also for creating them. Each rule may also contain a set of default properties which are used in create operations - for instance: `MATCH (d:Deployment {name: "nginx"}) CREATE (d)->(s:Service)` Will create a clusterIp Service (similar to what `kubectl expose` does). Doing this with kubectl or programmatically without tuning for rate limiting, implementing caching etc. of course fails even at a very small scale - Cyphernetes has its own request throttling and caching layer to make this possible.

zaps 2y ago

Phern is terrifying.

fatliverfreddy 2y ago
It seems terrifying at first but once you get to know it you’ll get over it. It’s super cuddly and really into graphs.
- muixoozie 2y ago
  Reminds me of fizzgig https://www.youtube.com/watch?v=lSnzoeZMljM https://www.youtube.com/watch?v=lSnzoeZMljM
  - fatliverfreddy 2y ago
    Oh I loved The Dark Crystal! Forgot about that little guy. Will use as inspiration for the next iteration :)

zug_zug 2y ago

If you want to expose kubernetes by a query language, why not expose an SQL interface?

fatliverfreddy 2y ago
I personally am a fan of Cypher and a Cypher-inspired language to describe graph operations in Kubernetes had been a dream of mine for several years until I eventually mustered the courage to create it myself. I just think it’s a great fit, much more than SQL! I’m aware picking up a new language will be a barrier for most to pick it up, and that’s fine. In any case I think this is an interesting tool worth sharing and I hope to find a few more that will be into it, who knows! While doing the research for this tool I did however come across two projects that do exactly what you describe: kubeql: https://github.com/saracen/kubeql https://github.com/saracen/kubeql kubesql: https://github.com/xuxinkun/kubesql https://github.com/xuxinkun/kubesql I’m not sure if either graduated past doing read operations, and both aren’t maintained anymore.