Since there are multiple types in the same topology, displaying the type is important.
We do this in multiple places:
* Add node type to minor label
* Add node type as metadata and include in metadata template.
Even though this will always be the same for every node of that topology, this was
the easiest way to add it so it displays in the table view.
Note we can't control ordering of columns in table view, it's always alphabetical.
While we're there, adopt a consistent ordering for all places that shapes are listed
Order is least sides to most sides, with circle before polygons, and complex shapes (currently just Cloud) after.
On shape choices for topologies:
* Since the k8s logo is a heptagon, we want pods to be heptagons.
* Since triangle is 'a bit weird', we put it on the least-important type, replica sets.
* Pentagons look a little weirder than octogons (it's the lack of symmetry) so we put octogons on the most common (deployments)
The existing technique of "reducing" the two rendered graphs for daemonsets and deployments
had a glaring issue that no connections would ever be made between nodes of different types,
since that information would've been discarded earlier in the process.
It also makes it hard to identify "parentless" pods.
This commit extends the Map2Parent function, teaching it:
* To check multiple topologies for parents
* To pass through nodes with no parents found without modification
Since we already had two 'modes' for what to do with nodes without parents,
and it would've been clunky to try to encode the third option into the existing PseudoNodeID
arg in some way, we instead split it into two args, with the first being an enum specifying
either the old pseudo node behaviour, the old drop behaviour, or the new keep behaviour.
We then use the new Map2Parent to map pods to:
* A replica set, if it has one
* A daemonset, if it has one
* Itself, if neither of the above
and then map again from the results to any deployment, leaving as-is any nodes that
don't map to a deployment. Hence we are left with:
* Deployments
* Daemonsets
* Replica sets, but only if they map to no deployment
* Pods, but only if they map to none of the above
and connections between all these will be calculated correctly.
ProcNet.Next does not allocate Connection structs, for efficiency.
Instead it always returns a *Connection pointing to the same instance.
As a result, any mutations by the caller to struct elements that
aren't actually set by ProcNet.Next, in particular Connection.Proc,
are carried across to subsequent calls.
This had hilarious consequences: connections referencing an inode
which we hadn't come across during proc walking would be associated
with the process corresponding to the last successfully looked up
inode.
The fix is to clear out the garbage left over from previous calls.
Fixes#2638.
* Rendering both nodes and edges from one component.
* Organized the render layers in a more readable manner.
* Rerendering optimization.
* Finer edge rendering layers.
* Moved the constants out of the component file.
* Typo fix.
* Include edge thickness in <Motion /> instead of fade-in effect.
ProcessRenderer was coloring connected nodes because we need that info
for rendering details panels. However, the main process topology view
renderers depending on ProcessRenderer were also doing coloring
themselves. For the 'processes' topology that was literally
duplicating work. For the 'processes-by-name' topology that was
throwing away the process coloring, and then coloring at the name
level.
Solution: remove the coloring from the ProcessRenderer, thus
eliminating the duplicate/thrown-away work, and introduce a
ColorConnectedProcessRenderer which is only used in places that
populate details panels.
In union option groups, the list of selected ids is encoded as a comma-delimited string.
For example, if 'foo' and 'bar' are selected, the value 'foo,bar' is sent, ie ["foo", "bar"] -> "foo,bar"
Under this scheme, with nothing selected, the empty string should be sent, ie. [] -> ""
Before this change, the frontend code called the "none option" by id 'none'.
There were several issues with this:
* The frontend would send 'none' when nothing was selected, not ''. The backend ignored this as it ignores
junk values in the options, treating them as though they hadn't been given.
* The backend would attempt to set the default value of an option to "nothing selected", ie. [], by sending ''.
The frontend would interpret this as nothing selected, *not even the 'none' option*, which caused a visual bug.
* Everything would break if one of the legitimate options had the id 'none', which could easily happen eg.
if a user had a 'none' k8s namespace. This is perhaps an unusual name, but our code shouldn't break when
a particular arbitary string is used as an input.
With this change, the none option is called '', which fixes all the above problems:
* The frontend encodes [''] as ''
* The frontend decodes '' as [''], and therefore shows the '' option as selected
* The string '' is not a valid k8s namespace name and is a reasonable "prohibited value" for all other use cases.
The backend already couldn't handle a value with this id correctly prior to this change anyway.
The rendering code checks whether endpoint IPs are part of
cluster-local networks. Due to the prevalence of endpoints - medium
sized reports can contain many thousands of endpoints - this is
performance critical. Alas the existing code performs the check via a
linear scan of a list of networks. That is slow when there are more
than a few, which will be the case in the context of k8s, since there
the probes register service IPs as local /32 networks.
Here we change representation of the set of networks to a prefix
tree (aka trie), which is well-suited for IP network membership checks
since networks are in fact a bitstring prefixes.
The specific representation is a crit-bit tree, but that choice was
purely based on implementation convenience - the chosen library is the
only one I could find that directly supports IP networks.
The rendering code checks whether endpoint IPs are part of
cluster-local networks. Due to the prevalence of endpoints - medium
sized reports can contain many thousands of endpoints - this is
performance critical. Alas the existing code performs the check via a
linear scan of a list of networks. That is slow when there are more
than a few. Unfortunately in some common k8s network setups, e.g. on
AWS, a cluster can contain hundreds of networks, due to /32 networks
derived from interfaces with multiple IPs.
Here we change representation of the set of networks to a prefix
tree (aka trie), which is well-suited for IP network membership checks
since networks are in fact a bitstring prefixes.
The specific representation is a crit-bit tree, but that choice was
purely based on implementation convenience - the chosen library is the
only one I could find that directly supports IP networks.
...by removing them. It was a ridiculous amount of contorted code to
test some utterly trivial functionality that is largely provided by
the golang stdlib.
* Node details fetching reports at proper timestamp.
* Corrected all the relevant timestamps in the UI.
* Renamed some state variables.
* Time travel works for topologies list.
* Added a whole screen overlay for time travel.
* Polished the backend.
* Make time travel work also with the Resource View.
* Fixed the jest tests.
* Fixed the empty view message for resource view.
* Some naming polishing.
* Addressed the comments.
