Bug 1625179 - [OSP-12] Nova puts virtio RX/TX Queue Size on SRIOV ports failing VM creation
Summary: [OSP-12] Nova puts virtio RX/TX Queue Size on SRIOV ports failing VM creation
Keywords:
Status: CLOSED ERRATA
Alias: None
Product: Red Hat OpenStack
Classification: Red Hat
Component: openstack-nova
Version: 12.0 (Pike)
Hardware: Unspecified
OS: Unspecified
medium
medium
Target Milestone: async
: 12.0 (Pike)
Assignee: OSP DFG:Compute
QA Contact: OSP DFG:Compute
URL:
Whiteboard:
Depends On: 1620171 1625180 1638299
Blocks: 1625178
TreeView+ depends on / blocked
 
Reported: 2018-09-04 10:37 UTC by Stephen Finucane
Modified: 2023-03-21 18:59 UTC (History)
14 users (show)

Fixed In Version: openstack-nova-16.1.4-7.el7ost
Doc Type: Enhancement
Doc Text:
The hypervisor kernel thread can pre-empt Virtual CPUs (vCPUs) even with strong partitioning enabled (isolcpus, tuned). These pre-emptions are not frequent, but with 256 descriptors per virtio queue, even one single pre-emption of the vCPU can cause packet drop in network function virtualization (NFC) virtual machines that have a packet rate of 1 Mpps (1 million packets per second) or higher. With this update, there are two new tunable options for configuring the RX queue size and TX queue size of virtio NICs and reducing packet drop: - 'rx_queue_size' - 'tx_queue_size'
Clone Of: 1620171
Environment:
Last Closed: 2018-12-05 18:50:50 UTC
Target Upstream Version:
Embargoed:


Attachments (Terms of Use)


Links
System ID Private Priority Status Summary Last Updated
Red Hat Product Errata RHBA-2018:3791 0 None None None 2018-12-05 18:52:23 UTC

Comment 12 errata-xmlrpc 2018-12-05 18:50:50 UTC
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHBA-2018:3791


Note You need to log in before you can comment on or make changes to this bug.