New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Bug 1967317: Do not create systemd update-rps service for veth devices #659
Bug 1967317: Do not create systemd update-rps service for veth devices #659
Conversation
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: cynepco3hahue The full list of commands accepted by this bot can be found here. The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
Pull Request Test Coverage Report for Build 1563
💛 - Coveralls |
39b8960
to
ab0e39e
Compare
@cynepco3hahue: This pull request references Bugzilla bug 1967317, which is invalid:
Comment In response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. |
/bugzilla refresh |
@cynepco3hahue: This pull request references Bugzilla bug 1967317, which is valid. The bug has been moved to the POST state. The bug has been updated to refer to the pull request using the external bug tracker. 3 validation(s) were run on this bug
Requesting review from QA contact: In response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. |
netns_link_indexes=$(ip netns exec "${ns}" ip -j link | jq ".[] | select(.link_index != null) | .link_index") | ||
for link_index in ${netns_link_indexes}; do | ||
container_veth=$(ip -j link | jq ".[] | select(.ifindex == ${link_index}) | .ifname" | tr -d '"') | ||
echo ${mask} > /sys/devices/virtual/net/${container_veth}/queues/rx-0/rps_cpus |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
is that the only path to be added ? rx-0 ?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
yep
/lgtm |
Yes, I think we want to backport it. Has it been tested yet? |
I tested it on my cluster. @browsell Did you have a chance to run it on top of the SNO cluster? |
ab0e39e
to
09d666c
Compare
The start and shutdown of big amount of pods will initiate creation of the systemd service that should update the new interfaces `rps_cpus` mask and can create an additional CPU load under the cluster. The PR introduces to changes that should prevent it: 1. The OCI hook will update the pod virtual interfaces RPS mask under the node. 2. Exclude veth devices from the udev rule. Signed-off-by: Artyom Lukianov <alukiano@redhat.com>
09d666c
to
9fecd2a
Compare
/lgtm |
We can not really on |
/hold cancel |
@cynepco3hahue: All pull requests linked via external trackers have merged: Bugzilla bug 1967317 has been moved to the MODIFIED state. In response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. |
/cherry-pick release-4.8 |
@cynepco3hahue: #659 failed to apply on top of branch "release-4.8":
In response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. |
…i_hook Bug 1967317: Do not create systemd update-rps service for veth devices (cherry picked from commit 33b9640) Signed-off-by: Artyom Lukianov <alukiano@redhat.com>
…i_hook Bug 1967317: Do not create systemd update-rps service for veth devices (cherry picked from commit 33b9640) Signed-off-by: Artyom Lukianov <alukiano@redhat.com>
RPS handling on pod container level using crio-hooks causes long delay times when running the low latency script to set the RPS mask (https://bugzilla.redhat.com/show_bug.cgi?id=2109965) For RAN low latency solution it might be sufficient only to set the RPS on the host level and avoid setting it on the container level while utilizing RSS behavior. In the past the low latency hook was added with RPS additional settings on virtual devices since there was an issue where the start and shutdown of big amount of pods will initiate the creation of the systemd service that should update the new interfaces rps_cpus mask and can create an additional CPU load under the cluster (openshift-kni/performance-addon-operators#659) This might not be the case any more thus we need to examine how the revert of the aforementioned PR will behave now. Co-authored-by: Yanir Quinn <yquinn@redhat.com> Signed-off-by: Talor Itzhak <titzhak@redhat.com>
RPS handling on pod container level using crio-hooks causes long delay times when running the low latency script to set the RPS mask (https://bugzilla.redhat.com/show_bug.cgi?id=2109965) For RAN low latency solution it might be sufficient only to set the RPS on the host level and avoid setting it on the container level while utilizing RSS behavior. In the past the low latency hook was added with RPS additional settings on virtual devices since there was an issue where the start and shutdown of big amount of pods will initiate the creation of the systemd service that should update the new interfaces rps_cpus mask and can create an additional CPU load under the cluster (openshift-kni/performance-addon-operators#659) This might not be the case any more thus we need to examine how the revert of the aforementioned PR will behave now. Co-authored-by: Yanir Quinn <yquinn@redhat.com> Signed-off-by: Talor Itzhak <titzhak@redhat.com>
RPS handling on pod container level using crio-hooks causes long delay times when running the low latency script to set the RPS mask (https://bugzilla.redhat.com/show_bug.cgi?id=2109965) For RAN low latency solution it might be sufficient only to set the RPS on the host level and avoid setting it on the container level while utilizing RSS behavior. In the past the low latency hook was added with RPS additional settings on virtual devices since there was an issue where the start and shutdown of big amount of pods will initiate the creation of the systemd service that should update the new interfaces rps_cpus mask and can create an additional CPU load under the cluster (openshift-kni/performance-addon-operators#659) This might not be the case any more thus we need to examine how the revert of the aforementioned PR will behave now. Co-authored-by: Yanir Quinn <yquinn@redhat.com> Signed-off-by: Talor Itzhak <titzhak@redhat.com>
* set RPS for veth on host level only RPS handling on pod container level using crio-hooks causes long delay times when running the low latency script to set the RPS mask (https://bugzilla.redhat.com/show_bug.cgi?id=2109965) For RAN low latency solution it might be sufficient only to set the RPS on the host level and avoid setting it on the container level while utilizing RSS behavior. In the past the low latency hook was added with RPS additional settings on virtual devices since there was an issue where the start and shutdown of big amount of pods will initiate the creation of the systemd service that should update the new interfaces rps_cpus mask and can create an additional CPU load under the cluster (openshift-kni/performance-addon-operators#659) This might not be the case any more thus we need to examine how the revert of the aforementioned PR will behave now. Co-authored-by: Yanir Quinn <yquinn@redhat.com> Signed-off-by: Talor Itzhak <titzhak@redhat.com> * set-rps-mask: remove `find` Since we're dealing only with virtual devices here which has a single queue, we can set the RPS mask directly. Signed-off-by: Talor Itzhak <titzhak@redhat.com> Signed-off-by: Talor Itzhak <titzhak@redhat.com> Co-authored-by: Yanir Quinn <yquinn@redhat.com>
RPS handling on pod container level using crio-hooks causes long delay times when running the low latency script to set the RPS mask (https://bugzilla.redhat.com/show_bug.cgi?id=2109965) For RAN low latency solution it might be sufficient only to set the RPS on the host level and avoid setting it on the container level while utilizing RSS behavior. In the past the low latency hook was added with RPS additional settings on virtual devices since there was an issue where the start and shutdown of big amount of pods will initiate the creation of the systemd service that should update the new interfaces rps_cpus mask and can create an additional CPU load under the cluster (openshift-kni/performance-addon-operators#659) This might not be the case any more thus we need to examine how the revert of the aforementioned PR will behave now. Co-authored-by: Yanir Quinn <yquinn@redhat.com> Signed-off-by: Talor Itzhak <titzhak@redhat.com>
RPS handling on pod container level using crio-hooks causes long delay times when running the low latency script to set the RPS mask (https://bugzilla.redhat.com/show_bug.cgi?id=2109965) For RAN low latency solution it might be sufficient only to set the RPS on the host level and avoid setting it on the container level while utilizing RSS behavior. In the past the low latency hook was added with RPS additional settings on virtual devices since there was an issue where the start and shutdown of big amount of pods will initiate the creation of the systemd service that should update the new interfaces rps_cpus mask and can create an additional CPU load under the cluster (openshift-kni#659) This might not be the case any more thus we need to examine how the revert of the aforementioned PR will behave now. Co-authored-by: Yanir Quinn <yquinn@redhat.com> Signed-off-by: Talor Itzhak <titzhak@redhat.com>
RPS handling on pod container level using crio-hooks causes long delay times when running the low latency script to set the RPS mask (https://bugzilla.redhat.com/show_bug.cgi?id=2109965) For RAN low latency solution it might be sufficient only to set the RPS on the host level and avoid setting it on the container level while utilizing RSS behavior. In the past the low latency hook was added with RPS additional settings on virtual devices since there was an issue where the start and shutdown of big amount of pods will initiate the creation of the systemd service that should update the new interfaces rps_cpus mask and can create an additional CPU load under the cluster (openshift-kni#659) This might not be the case any more thus we need to examine how the revert of the aforementioned PR will behave now. Co-authored-by: Yanir Quinn <yquinn@redhat.com> Signed-off-by: Talor Itzhak <titzhak@redhat.com>
RPS handling on pod container level using crio-hooks causes long delay times when running the low latency script to set the RPS mask (https://bugzilla.redhat.com/show_bug.cgi?id=2109965) For RAN low latency solution it might be sufficient only to set the RPS on the host level and avoid setting it on the container level while utilizing RSS behavior. In the past the low latency hook was added with RPS additional settings on virtual devices since there was an issue where the start and shutdown of big amount of pods will initiate the creation of the systemd service that should update the new interfaces rps_cpus mask and can create an additional CPU load under the cluster (openshift-kni/performance-addon-operators#659) This might not be the case any more thus we need to examine how the revert of the aforementioned PR will behave now. Signed-off-by: Talor Itzhak <titzhak@redhat.com>
RPS handling on pod container level using crio-hooks causes long delay times when running the low latency script to set the RPS mask (https://bugzilla.redhat.com/show_bug.cgi?id=2109965) For RAN low latency solution it might be sufficient only to set the RPS on the host level and avoid setting it on the container level while utilizing RSS behavior. In the past the low latency hook was added with RPS additional settings on virtual devices since there was an issue where the start and shutdown of big amount of pods will initiate the creation of the systemd service that should update the new interfaces rps_cpus mask and can create an additional CPU load under the cluster (openshift-kni#659) This might not be the case any more thus we need to examine how the revert of the aforementioned PR will behave now. Signed-off-by: Talor Itzhak <titzhak@redhat.com>
RPS handling on pod container level using crio-hooks causes long delay times when running the low latency script to set the RPS mask (https://bugzilla.redhat.com/show_bug.cgi?id=2109965) For RAN low latency solution it might be sufficient only to set the RPS on the host level and avoid setting it on the container level while utilizing RSS behavior. In the past the low latency hook was added with RPS additional settings on virtual devices since there was an issue where the start and shutdown of big amount of pods will initiate the creation of the systemd service that should update the new interfaces rps_cpus mask and can create an additional CPU load under the cluster (openshift-kni/performance-addon-operators#659) This might not be the case any more thus we need to examine how the revert of the aforementioned PR will behave now. Signed-off-by: Talor Itzhak <titzhak@redhat.com> Signed-off-by: Talor Itzhak <titzhak@redhat.com>
* set RPS for veth on host level only RPS handling on pod container level using crio-hooks causes long delay times when running the low latency script to set the RPS mask (https://bugzilla.redhat.com/show_bug.cgi?id=2109965) For RAN low latency solution it might be sufficient only to set the RPS on the host level and avoid setting it on the container level while utilizing RSS behavior. In the past the low latency hook was added with RPS additional settings on virtual devices since there was an issue where the start and shutdown of big amount of pods will initiate the creation of the systemd service that should update the new interfaces rps_cpus mask and can create an additional CPU load under the cluster (openshift-kni/performance-addon-operators#659) This might not be the case any more thus we need to examine how the revert of the aforementioned PR will behave now. Co-authored-by: Yanir Quinn <yquinn@redhat.com> Signed-off-by: Talor Itzhak <titzhak@redhat.com> * set-rps-mask: remove `find` Since we're dealing only with virtual devices here which has a single queue, we can set the RPS mask directly. Signed-off-by: Talor Itzhak <titzhak@redhat.com> Signed-off-by: Talor Itzhak <titzhak@redhat.com> Co-authored-by: Yanir Quinn <yquinn@redhat.com>
* set RPS for veth on host level only RPS handling on pod container level using crio-hooks causes long delay times when running the low latency script to set the RPS mask (https://bugzilla.redhat.com/show_bug.cgi?id=2109965) For RAN low latency solution it might be sufficient only to set the RPS on the host level and avoid setting it on the container level while utilizing RSS behavior. In the past the low latency hook was added with RPS additional settings on virtual devices since there was an issue where the start and shutdown of big amount of pods will initiate the creation of the systemd service that should update the new interfaces rps_cpus mask and can create an additional CPU load under the cluster (openshift-kni/performance-addon-operators#659) This might not be the case any more thus we need to examine how the revert of the aforementioned PR will behave now. Co-authored-by: Yanir Quinn <yquinn@redhat.com> Signed-off-by: Talor Itzhak <titzhak@redhat.com> * set-rps-mask: remove `find` Since we're dealing only with virtual devices here which has a single queue, we can set the RPS mask directly. Signed-off-by: Talor Itzhak <titzhak@redhat.com> Signed-off-by: Talor Itzhak <titzhak@redhat.com> Co-authored-by: Yanir Quinn <yquinn@redhat.com>
The start and shutdown of big amount of pods will initiate the creation of the systemd
service that should update the new interfaces
rps_cpus
mask and can create an additionalCPU load under the cluster.
The PR introduces to changes that should prevent it:
Signed-off-by: Artyom Lukianov alukiano@redhat.com