Endpoints generated by the endpoints controller are in the canonical
form, however, custom endpoints can not be in canonical format
(there was a time they were canonicalized in the apiserver, but this
caused performance issues because the endpoint controller kept
updating them since the created endpoint were different than the
stored one due to the canonicalization)
There are cases where a custom endpoint may generate multiple slices
due to the controller, per example, when the same address is present
in different subsets.
The endpointslice mirroring controller should canonicalize the
endpoints subsets before start processing them to be consistent
on the slices generated, there is no risk of hotlooping because
the endpoint is only used as input.
Change-Id: I2a8cd53c658a640aea559a88ce33e857fa98cc5c
This ensures that the daemonset controller updates daemonset statuses in
a best-effort manner even if syncDaemonSet fails.
In order to add an integration test, this also replaces
`cmd/kube-apiserver/app/testing.StartTestServer` with
`test/integration/framework.StartTestServer` and adds
`setupWithServerSetup` to configure the admission control of the
apiserver.
This change will leverage the new PreFilterResult
to reduce down the list of eligible nodes for pod
using Bound Local PVs during PreFilter stage so
that only the node(s) which local PV node affinity
matches will be cosnidered in subsequent scheduling
stages.
Today, the NodeAffinity check is done during Filter
which means all nodes will be considered even though
there may be a large number of nodes that are not
eligible due to not matching the pod's bound local
PV(s)' node affinity requirement. Here we can
reduce down the node list in PreFilter to ensure that
during Filter we are only considering the reduced
list and thus can provide a more clear message to
users when node(s) are not available for scheduling
since the list only contains relevant nodes.
If error is encountered (e.g. PV cache read error) or
if node list reduction cannot be done (e.g. pod uses
no local PVs), then we will still proceed to consider
all nodes for the rest of scheduling stages.
Signed-off-by: Yibo Zhuang <yibzhuang@gmail.com>
This adds a new resource.k8s.io API group with v1alpha1 as version. It contains
four new types: resource.ResourceClaim, resource.ResourceClass, resource.ResourceClaimTemplate, and
resource.PodScheduling.
Also make some design changes exposed in testing and review.
Do not remove the ambiguous old metric
`apiserver_flowcontrol_request_concurrency_limit` because reviewers
though it is too early. This creates a problem, that metric can not
keep both of its old meanings. I chose the configured concurrency
limit.
Testing has revealed a design flaw, which concerns the initialization
of the seat demand state tracking. The current design in the KEP is
as follows.
> Adjustment is also done on configuration change … For a newly
> introduced priority level, we set HighSeatDemand, AvgSeatDemand, and
> SmoothSeatDemand to NominalCL-LendableSD/2 and StDevSeatDemand to
> zero.
But this does not work out well at server startup. As part of its
construction, the APF controller does a configuration change with zero
objects read, to initialize its request-handling state. As always,
the two mandatory priority levels are implicitly added whenever they
are not read. So this initial reconfig has one non-exempt priority
level, the mandatory one called catch-all --- and it gets its
SmoothSeatDemand initialized to the whole server concurrency limit.
From there it decays slowly, as per the regular design. So for a
fairly long time, it appears to have a high demand and competes
strongly with the other priority levels. Its Target is higher than
all the others, once they start to show up. It properly gets a low
NominalCL once other levels show up, which actually makes it compete
harder for borrowing: it has an exceptionally high Target and a rather
low NominalCL.
I have considered the following fix. The idea is that the designed
initialization is not appropriate before all the default objects are
read. So the fix is to have a mode bit in the controller. In the
initial state, those seat demand tracking variables are set to zero.
Once the config-producing controller detects that all the default
objects are pre-existing, it flips the mode bit. In the later mode,
the seat demand tracking variables are initialized as originally
designed.
However, that still gives preferential treatment to the default
PriorityLevelConfiguration objects, over any that may be added later.
So I have made a universal and simpler fix: always initialize those
seat demand tracking variables to zero. Even if a lot of load shows
up quickly, remember that adjustments are frequent (every 10 sec) and
the very next one will fully respond to that load.
Also: revise logging logic, to log at numerically lower V level when
there is a change.
Also: bug fix in float64close.
Also, separate imports in some file
Co-authored-by: Han Kang <hankang@google.com>
