StarlingX miscellaneous tools and utilities
Go to file
Tara Subedi d36ac91d90 Enrolment failure on enroll subcloud with different OAM subnet address
During subcloud enrollment, default route is always expected to be as
set by cloud-init. It is seen sometimes overwritten by OAM ifcfg
configuration even before oam-modify triggered.

This commit forces cloud-init route back i.e in case of oam-
reconfiguration, we are forcing to have default route as given by new
configuration through cloud-init, and making sure there is no old
default route given by existing oam configuration.

This provides OAM connection to subcloud through new OAM configuration
at early stage.

This also creates temporary file /var/run/.enroll-init-reconfigure,
before doing oam-modify, which later checked by network runtime script
to restate the default OAM route [1].
[1] https://review.opendev.org/c/starlingx/stx-puppet/+/937800

TEST PLAN:
  PASS: subcloud enrollment with oam-reconfig with interface change
        - OAM connection based on cloud-init's new IP/route/interface
  PASS: subcloud enrollment with oam-reconfig (oam-subnet) with vlan change
        - OAM connection based on cloud-init's new IP/route/vlan
  PASS: subcloud enrollment with oam-reconfig w/o interface/vlan change
        - OAM connection based on cloud-init's new IP/route
  PASS: subcloud enrollment without oam-reconfig
  PASS: test above subcloud enrollement with both IPv4 and IPv6 on OAM
    - check /var/log/cloud-init-output.log output to see
      oam default route through cloud-init dev before oam-modify
      triggered
    - check file created: /var/run/.enroll-init-reconfigure

Closes-bug: 2092151
Change-Id: I5b2af5610f43aa8cf76baff64abd94ec7cc411e0
Signed-off-by: Tara Nath Subedi <tara.subedi@windriver.com>
2024-12-19 13:32:27 -05:00
ceph Merge "Change stx-ceph-manager base image" 2024-08-05 15:02:58 +00:00
security Remove CentOS/OpenSUSE build support 2024-04-26 14:19:03 -04:00
tools Add more IPSec info to the collect tool 2024-12-13 11:59:08 -03:00
utilities Enrolment failure on enroll subcloud with different OAM subnet address 2024-12-19 13:32:27 -05:00
.gitignore Refactoring novaClient instantiation and unittests 2022-01-25 19:27:36 -03:00
.gitreview Add a .gitreview file to the new repo 2019-09-09 09:48:42 -05:00
.zuul.yaml Add python-ldap deps for zuul test 2023-09-15 18:41:49 +00:00
bindep.txt Add python-ldap deps for zuul test 2023-09-15 18:41:49 +00:00
CONTRIBUTING.rst Adding zuul jobs for new repo 2019-09-09 13:43:49 -05:00
debian_build_layer.cfg Add debian_build_layer.cfg file 2021-10-05 14:13:38 -04:00
debian_iso_image.inc debian: Remove debian-integration package 2022-12-06 08:01:23 -05:00
debian_pkg_dirs Add accel-config to stx-debian-tools-dev docker image 2024-03-21 12:20:34 -03:00
debian_stable_docker_images.inc Port stx-pci-irq-affinity-agent to stx-debian 2023-01-16 15:44:08 -03:00
debian_stable_wheels.inc Debian: Add build structure for utilities/pci-irq-affinity-agent 2023-01-16 15:43:53 -03:00
HACKING.rst Adding zuul jobs for new repo 2019-09-09 13:43:49 -05:00
pylint.rc Re-enable important py3k checks for utilities 2021-10-21 12:34:24 +00:00
README.rst Document PCI IRQ Affinity Agent operation 2021-11-04 16:39:08 -03:00
requirements.txt Turn off legacy resolver workaround in pip 2020-12-17 17:04:41 -06:00
test-requirements.txt Add flake8-import-order and use python3.9 on tox 2022-09-13 21:49:41 +00:00
tox.ini Add python-ldap deps for zuul test 2023-09-15 18:41:49 +00:00

utilities

This file serves as documentation for the components and features included on the utilities repository.

PCI IRQ Affinity Agent

While in OpenStack it is possible to enable instances to use PCI devices, the interrupts generated by these devices may be handled by host CPUs that are unrelated to the instance, and this can lead to a performance that is lower than it could be if the device interrupts were handled by the instance CPUs.

The agent only acts over instances with dedicated vCPUs. For instances using shared vCPUs no action will be taken by the agent.

The expected outcome from the agent operation is achieving a higher performance by assigning the instances core to handle the interrupts from PCI devices used by these instances and avoid interrupts consuming excessive cycles from the platform cores.

Agent operation

The agent operates by listening to RabbitMQ notifications from Nova. When an instance is created or moved to the host, the agent checks for an specific flavor spec (detailed below) and if it does then it queries libvirt to map the instance vCPUs into pCPUs from the host.

Once the agent has the CPU mapping, it determines the IRQ for each PCI device used by the instance, and then it loops over all PCI devices and determines which host NUMA node is associated with the device, the pCPUs that are associated with the NUMA node and finally set the CPU affinity for the IRQs of the PCI device based on the pCPU list.

There is also a periodic audit that runs every minute and loops over the existing IRQs, so that if there are new IRQs that weren't mapped before the agent maps them, and if there are PCI devices that aren't associated to an instance that they were before, their IRQ affinity is reset to the default value.

Flavor spec

The PCI IRQ Affinity Agent uses a specific flavor spec for PCI interrupt affining, that is used to determine which vCPUs assigned to the instance must handle the interrupts from the PCI devices:

  • hw:pci_irq_affinity_mask=<vcpus_cpulist>

Where vcpus_cpulist can assume a comma-separated list of values that can be expressed as:

  • int: the vCPU expressed by int will be assigned to handle the interruptions from the PCI devices
  • int1-int2: the vCPUs between int1 and int2 (inclusive) will be used to handle the interruptions from the PCI devices
  • ^int: the vCPU expressed by int will not be assigned to handle the interruptions from the PCI devices and shall be used to exclude a vCPU that was included in a previous range

NOTE: int must be a value between 0 and flavor.vcpus - 1

Example: hw_pci_irq_affinity_mask=1-4,^3,6 means that vCPUs with indexes 1,2,4 and 6 from the vCPU list that Nova allocates to the instance will be assigned to handle interruptions from the PCI devices.

Limitations

  • No CPU affining is performed for instances using shared CPUs (i.e., when using flavor spec hw:cpu_policy=shared)
  • No CPU affining will be performed when invalid ranges are specified on the flavor spec, the agent instead will log error messages indicating the problem

Agent packaging

The agent code resides on the starlingx/utilities repo, along with the spec and docker_image files that are used to build a CentOS image with the agent wheel installed on it.

The agent is deployed by Armada along with the other OpenStack helm charts; refer to PCI IRQ Affinity Agent helm chart on starlingx/openstack-armada-app repository.