Use a special kind of selector renderer to elide replica sets from pod nodes
and directly reference deployment parents instead.
Do the inverse (replace replica sets with pods) during the mapping from pod to deployment.
Note we can no longer use renderParents since we're using a non-standard Selector
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.
the information is constant and already present in the id, so we can
extract it from there.
That reduces the report size and improves report encoding/decoding
performance. It should reduce memory usage too and improve report
merging performance too.
NB: Probes with this change are incompatible with old apps.
The default sets the node label to the node ID.
This is likely to not look very good, but the intent is that it creates an obvious problem,
ie. that the node ID is being used as the label, rather than a silent omission or more subtle problem.
Possible future work:
* For single-component IDs, extract the component automatically and use that instead.
* Instead of functions, in simple cases just have a LUT by topology with common behaviours like
'stack = true or false', 'label = this key in node.Latest'
The latter opens up to eventually moving this info inside the report itself ala topology templates,
or at least centralizing it in the source.
This gives us a single source of truth in a variety of situations where we want to
know what view to direct a user to in order to 'open' a particular node.
I wanted to put this in app/api_topologies where the views are defined, but that creates
a circular import.
This is important for two reasons:
* It prevents nasty false-equality bugs when two different services from different ECS clusters
are present in the same report
* It allows us to retrieve the cluster and service name - all the info we need to look up the service -
using only the node ID. This matters, for example, when trying to handle a control request.
- Rename DataType -> Datatype for consisitency
- Remove Datatype from connection.metadata
- Change UI to read dataType from connection.columns rather than from
connection.metadata (col header rather than "cell")
This preserves the existing behaviour of showing count=0 as a plural
rather than empty. However, I rather suspect we cannot actually
encounter that case in the current code.
In the "no count found" case we now *set* the base.LabelMinor to empty
instead of leaving it alone. This is ok since it is empty to begin with.
- Change scrolling behaviour to lock headers in place
- Enable filtering (hitting enter in the search bar) in grid-mode
- Little more top-margin for k8s (can have 3 topos) + taller rows.
- Trying out rank-color + node.relatives in the grid-mode
- First pass at selecting rows.
- Needs a bit more a fiddle, colors + click areas
- Store grid sort direction (asc/desc) in url state
- Simplify node selection to one method. (over-ride existing card)
- Remove clicking on name directly (links) to overlay new cards for now.
- Playing w/ grid-mode-toggle icons and labels
- Improves rendering in ff, change of shortcut keys for grid-mode-toggle
- Playing w/ clearer selection colors for grid-mode
- Slight change to selection-ui
- Fixes showNodeInTopology button visibility on the details-panel
- Was using an old heuristic. Table-mode allows you to open child cards
before the parent.
- Make it clear what the default sort is in tables
- E.g. always show a sorting caret
- Sort grid-mode columns, first meta then metrics
- dancing-nodes rememdy #1: pause updates onRowHover
- Splits relatives out into their own columns
- Take into account scrollbar width for grid-mode col header position
- Tooltips on table column headers
- grid-mode: fixes first column headers (proc/container/c-by-image)
- Disable pause-on-hover, too aggresive
- reduce label column width a bit (33pc -> 25pc) for big tables
- Filter grid-mode onSearchChange
- Rather than previous behaviour of waiting for an <enter>
- Show label_minor on pseudo nodes, that might not have much other info
- grid-mode: further reduce width of id column.
- Fixes go tests, properly moves parents into node-summary
- Fixes sorting of string columns w/ missing fields.
- E.g. uptime. Where -1e-10 > '3days' doesn't work.
* Show k8s labels and container env vars in the details panel.
* Add show more bar to the env vars and labels
* React key was in the wrong place; empty tables render section labels.
* Collect container environment variales (close#1314)
* Containers managed by Marathon can have meaningful names (close#1309)
* Make linter happy and fix tests
Squash of:
* Include plugins in the report
* show plugin list in the UI
* moving metric and metadata templates into the probe reports
* update js for prime -> priority
* added retry to plugin handshake
* added iowait plugin
* review feedback
* plugin documentation
Squash of:
- including children in topologies_test.go
- report.Node.Prune should prune children also
- rewrote ShortLivedInternetConnections test to express its intent
- adding tests for detail Summary rendering
Squash of:
- use detailed.Summaries to render topology nodes
- ban merging nodes of different topologies (they should be mapped)
- need to prune parents when mapping node types
- render container images by id if they have no name
- remove separate render ids and prune parents in NewDerived*
- don't render metrics/metadata for groups of nodes
- fixing up tests
- removing pending unit tests (for mapping.go, for now)
- updating experimental dir for RenderableNode removal
* Bump github.com/ugorji/go/codec, to get fixes which make
github.com/2opremio/go-1/codec unnecessary
* Remove github.com/2opremio/go-1/codec
* Remove type embeddings to avoid the problem explained at
https://github.com/ugorji/go/issues/141#issuecomment-185450157
* New encoding format:
* Msgpack reports between probe<->app (smaller representation, faster to
encode/decode).
* Still use JSON between app<->UI (try to avoid making javascript deal with
mspack).
The app still suports publishing reports in both gob and JSON, not braking
backwards compatibility.
* Use compile-time generated marshallers/unmarshallers for higher performance. In
order to be able to skip code-generation for certain types, I included
https://github.com/2opremio/go-1/tree/master/codec/codecgen instead of
upstream until https://github.com/ugorji/go/pull/139 is merged.
* Encode/decode intermediate types using github.com/ugorji/go/codec.Selfer
for higher performance and reducing garbage collection (no temporary buffers).
