passt/test, branch 2025_08_05.309eefd

treewide: By default, don't quit source after migration, keep sockets open

2025-07-29T15:57:01+00:00

We are hitting an issue in the KubeVirt integration where some data is
still sent to the source instance even after migration is complete. As
we exit, the kernel closes our sockets and resets connections. The
resulting RST segments are sent to peers, effectively terminating
connections that were meanwhile migrated.

At the moment, this is not done intentionally, but in the future
KubeVirt might enable OVN-Kubernetes features where source and
destination nodes are explicitly getting mirrored traffic for a while,
in order to decrease migration downtime.

By default, don't quit after migration is completed on the source: the
previous behaviour can be enabled with the new, but deprecated,
--migrate-exit option. After migration (as source), the -1 / --one-off
option has no effect.

Also, by default, keep migrated TCP sockets open (in repair mode) as
long as we're running, and ignore events on any epoll descriptor
representing data channels. The previous behaviour can be enabled with
the new, equally deprecated, --migrate-no-linger option.

By keeping sockets open, and not exiting, we prevent the kernel
running on the source node to send out RST segments if further data
reaches us.

Reported-by: Nir Dothan 
Signed-off-by: Stefano Brivio

test: Deal with /bin, /sbin unification in Fedora

2025-07-22T21:49:43+00:00

Fedora is apparently merging /bin and /sbin together [0].  At least some of
this has landed in Fedora 42, with sshd and dhclient now being in /bin.
They are symlinked from /sbin, but mbuto doesn't preserve that symlink when
it builds its image, it just puts them in /bin.  That breaks later parts
of the script which expects to find sshd and dhclient in /sbin.  Other
parts of the test also expect sysctl in /sbin.

Use mbuto's LINKS feature to deal with these cases.

[0] https://fedoraproject.org/wiki/Changes/Unify_bin_and_sbin

Signed-off-by: David Gibson 
Signed-off-by: Stefano Brivio

test: Display count of skipped tests in status and summary

2025-05-15T16:06:19+00:00

This commit enhances test reporting by tracking and displaying the
number of skipped tests.

The skipped test count is now visible in the tmux status bar during
execution and included in the final test summary log. This provides
a more complete overview of test suite results.

Signed-off-by: Laurent Vivier 
Signed-off-by: Stefano Brivio

test/perf: Simplify iperf3 server lifetime management

2025-03-12T22:08:33+00:00

After we start the iperf3 server in the background, we have a sleep to
make sure it's ready to receive connections.  We can simplify this slightly
by using the -D option to have iperf3 background itself rather than
backgrounding it manually.  That won't return until the server is ready to
use.

Signed-off-by: David Gibson 
Signed-off-by: Stefano Brivio

test: Add migration tests

2025-02-17T07:29:36+00:00

PCAP=1 ./run migrate/bidirectional gives an overview of how the
whole thing is working.

Add 12 tests in total, checking basic functionality with and without
flows in both directions, with and without sockets in half-closed
states (both inbound and outbound), migration behaviour under traffic
flood, under traffic flood with > 253 flows, and strict checking of
sequences under flood with ramp patterns in both directions.

These tests need preparation and teardown for each case, as we need
to restore the source guest in its own context and pane before we can
test again. Eventually, we could consider alternating source and
target so that we don't need to restart from scratch every time, but
that's beyond the scope of this initial test implementation.

Trick: './run migrate/*' runs all the tests with preparation and
teardown steps.

Co-authored-by: David Gibson 
Reviewed-by: David Gibson 
Signed-off-by: Stefano Brivio

rampstream: Add utility to test for corruption of data streams

2025-02-12T18:48:17+00:00

Signed-off-by: David Gibson 
Signed-off-by: Stefano Brivio

dhcp, dhcpv6: Add hostname and client fqdn ops

2025-02-10T17:30:24+00:00

Both DHCPv4 and DHCPv6 has the capability to pass the hostname to
clients, the DHCPv4 uses option 12 (hostname) while the DHCPv6 uses option 39
(client fqdn), for some virt deployments like kubevirt is expected to
have the VirtualMachine name as the guest hostname.

This change add the following arguments:
 - -H --hostname NAME to configure the hostname DHCPv4 option(12)
 - --fqdn NAME to configure client fqdn option for both DHCPv4(81) and
   DHCPv6(39)

Signed-off-by: Enrique Llorente 
Signed-off-by: Stefano Brivio

debug: Add tcpdump to mbuto.img

2025-02-06T08:43:09+00:00

Signed-off-by: David Gibson 
Signed-off-by: Stefano Brivio

test/pasta_podman: Run Podman tests on a single CPU thread

2025-01-15T21:16:32+00:00

Increasingly often, I'm getting occasional failures of the same type
as https://github.com/containers/podman/issues/24147. I guess it
mostly depends on the system load.

It will be a while until I'll actually run tests on a kernel
including my fix for it, kernel commit a502ea6fa94b ("udp: Deal with
race between UDP socket address change and rehash"), so add a horrible
workaround using taskset(1), for the moment.

Signed-off-by: Stefano Brivio

perf/passt_vu_tcp: Make it shine

2024-11-28T14:06:44+00:00

Signed-off-by: Stefano Brivio