Note: This bug is displayed in read-only format because the product is no longer active in Red Hat Bugzilla.

RHEL Engineering is moving the tracking of its product development work on RHEL 6 through RHEL 9 to Red Hat Jira (issues.redhat.com). If you're a Red Hat customer, please continue to file support cases via the Red Hat customer portal. If you're not, please head to the "RHEL project" in Red Hat Jira and file new tickets here. Individual Bugzilla bugs in the statuses "NEW", "ASSIGNED", and "POST" are being migrated throughout September 2023. Bugs of Red Hat partners with an assigned Engineering Partner Manager (EPM) are migrated in late September as per pre-agreed dates. Bugs against components "kernel", "kernel-rt", and "kpatch" are only migrated if still in "NEW" or "ASSIGNED". If you cannot log in to RH Jira, please consult article #7032570. That failing, please send an e-mail to the RH Jira admins at rh-issues@redhat.com to troubleshoot your issue as a user management inquiry. The email creates a ServiceNow ticket with Red Hat. Individual Bugzilla bugs that are migrated will be moved to status "CLOSED", resolution "MIGRATED", and set with "MigratedToJIRA" in "Keywords". The link to the successor Jira issue will be found under "Links", have a little "two-footprint" icon next to it, and direct you to the "RHEL project" in Red Hat Jira (issue links are of type "https://issues.redhat.com/browse/RHEL-XXXX", where "X" is a digit). This same link will be available in a blue banner at the top of the page informing you that that bug has been migrated.

Description of problem: VM to VM over OVS-dpdk bond got "no route to host" while running multiple streams of netperf UDP_STREAM with mode=balance-slb and active-backup.
 
NOTE #1: Even in 300 seconds this issue happened.

NOTE #2: In a 24 hours 64 streams of netperf UDP_STREAM test with 64 bytes messages 24 the 64 streams encountered this error.

NOTE #3: It got worse with mode=balance-slb.


Version-Release number of selected component (if applicable):

3.10.0-327.13.1.el7.x86_64

dpdk-2.2.0-3

How reproducible: Reproducible


Steps to Reproduce:
1.OVS-dpdk bond between two Hosts, each with a 4-Q vhostuser guest
2.On each guest configure a regular OVS with eth0 and 64 internal ports.
3.Run multiple streams of netperf UDP_STREAM tests between them.

Actual results: Some streams got "no route to host" errors.


Expected results: All 64 streams should complete successfully in a 24 hour testing --- UDP_STREAM with 64 bytes messages.


Additional info:

Hi, Jean-Tsung.

Do you see that with a setup with bond and host-to-host? Or with no bond and multiqueue guest to multiqueue guest?

I am wondering if only one of those setups cause this, or if both things are really needed. Also, does the guest need to run OVS with that many internal ports as well?

If you don't have that data right now, can you test guest to guest, with bond configured on the host, but the guest using a single queue and no OVS on the guest?

Thanks a lot.
Cascardo.

Hi, Jean-Tsung.

Do you think this could be a duplicate of bug#1306466?

I would guess the cause is the same one.

Thanks.
Cascardo.

(In reply to Thadeu Lima de Souza Cascardo from comment #3)
> Hi, Jean-Tsung.
> 
> Do you think this could be a duplicate of bug#1306466?
> 
> I would guess the cause is the same one.
> 
> Thanks.
> Cascardo.

yes

*** This bug has been marked as a duplicate of bug 1306466 ***