- Add a defaultpreemption PostFilter plugin
- Make g.Preempt() stateless
- make g.Preempt() stateless
- make g.getLowerPriorityNominatedPods() stateless
- make g.processPreemptionWithExtenders() stateless
`DefaultPodTopologySpread` need't score when the `TopologySpreadConstraints`
is specified.
`PreScore` needn't do this as well, this cut off the cost of `PreScore` if
possible.
Signed-off-by: Dave Chen <dave.chen@arm.com>
iSCSI volume plugin persists volume metadata into global mount directory,
before it is mounted. Content of the directory is shadowed by the volume
mount. Therefore kubelet should not write metadata to the directory when a
second pod uses the same volume on the same node.
1. The metadata were already persisted before mounting the volume for the
first pod.
2. The global mount directory has the volume mounted, so any write there
would write to the volume, which is undesirable.
This make it easier to catch the issue during the compilation, also,
this also align with other plugins, i.e. plugin of "InterPodAffinity".
Signed-off-by: Dave Chen <dave.chen@arm.com>
Before this commit, containers which have both a `startupProbe` and a
`readinessProbe` are marked as `ready=false` during stratup, but
containers which have only a `startupProbe` are marked `ready=true`.
This doesn't make sense.
This commit only considers readiness if the container is considered to
have "started", which leaves `ready=false` while starting up.
refactor and add the following metrics to the cidr_sets used by the range
allocator:, under the subsystem: node_ipam_controller
cidrset_cidrs_allocations_total
cidrset_cidrs_releases_total
cidrset_usage_cidrs
cidrset_allocation_tries_per_request
It seems that if you set the packet mark on a packet and then route
that packet through a kernel VXLAN interface, the VXLAN-encapsulated
packet will still have the mark from the original packet. Since our
NAT rules are based on the packet mark, this was causing us to
double-NAT some packets, which then triggered a kernel checksumming
bug. But even without the checksum bug, there are reasons to avoid
double-NATting, so fix the rules to unmark the packets before
masquerading them.
- replace error with NodeToStatusMap in Preempt() signature
- eliminate podPreemptor interface and expose its functions statelessly
- move logic in scheduler.go#preempt to generic_scheduler.go#Preempt()
These changes allow to set FQDN as hostname of pods for pods
that set the new PodSpec field setHostnameAsFQDN to true. The PodSpec
new field was added in related PR.
This is PART2 (last) of the changes to enable KEP #1797 and addresses #91036
These changes add a new field, called setHostnameAsFQDN, to the PodSpec. This
field is a bool that will be used to indicate whether we would like
FQDN be set as hostname or not.
This is PART1 of the changes to enable KEP #1797 and addresses #91036
