There's currently no way to know whether an error is for SCTP or
UDP, for example:
Jul 24 09:55:54.469: INFO: netserver-0[e2e-nettest-3476].container[webserver].log
2020/07/24 09:53:52 Started UDP server
2020/07/24 09:53:52 Error occurred. error:protocol not supported
In this case the "Error occurred. error:protocol not supported" is
actually for the SCTP socket. Make that more apparent.
bprashanth hasn't be active since the release of v1.11. Removing them
from test/images/pets/OWNERS would leave mkumatag as the sole approver.
But mkumatag is also an approver for test/images/OWNERS so this commit
removes the test/images/pets/OWNERS completely.
Note: we should try to find more OWNERS for test/images/pets instead,
but this cleanup is a short term solution to avoid the bot suggesting
inactive members for reviews and approval.
Currently, the jessie-dnsutils image fails to build for arm64 arch with the following
error:
GPG error: http://archive.debian.org jessie Release: The following signatures were invalid: KEYEXPIRED 1587841717
We can bypass this issue by adding a --force-yes flag when installing the needed dnsutils.
Adds reviewers to the OWNERS files in the kubernetes/test/images folder.
The reviewers are added automatically, based on their contributions on
an image (>= 20% code churn).
Note that the code churn is taken into account for authors, and not committers.
Adds OWNERS files for: ipc-utils, node-perf, nonroot, regression-issue-74839,
resource-consumer, sample-device-plugin.
The google cloud builder job is launched without the required Windows Image Builder nodes
certificates that are needed for authentication when building the Windows container images.
Adds a step in test/images/cloudbuild.yaml that fetches a secret containing the certificates.
Removes the fatal error from getIP and moves it to
retry loop so that application will not immediately
crash on failure.
Signed-off-by: hasheddan <georgedanielmangum@gmail.com>
This updates the error messages when registering a
node to be more explicit about what error occurred
and how long it will wait to retry.
Signed-off-by: hasheddan <georgedanielmangum@gmail.com>
Currently the guestbook application will fail if unable
to resolve TCP address on first attempt. If pod networking
is not setup when the application starts then it will be
unable to resolve, leading to frequent failures. This moves
the address resolution into the retry block so it will try
again if unsuccessful on first attempt.
Signed-off-by: hasheddan <georgedanielmangum@gmail.com>
The current /exit method is not sufficient to test graceful shutdown
behaviors within Kube that allow services to remain available during
rolling restarts. Add support for `wait=DURATION` and
`timeout=DURATION` to the exit handler and wire that to the Go http
server's graceful termination.
With these methods netexec can be used in a pod to simulate graceful
shutdown by adding a preStop handler that hits the exit endpoint with
a timeout and wait period.
