| Message ID | 20240506114320.12178-1-fw@strlen.de (mailing list archive) |
|---|---|
| State | Accepted |
| Commit | 76508154d7da82c139416eee6d318fde4feae143 |
| Delegated to: | Netdev Maintainers |
| Headers | show |
| Series | [net-next] selftests: netfilter: conntrack_tcp_unreplied.sh: wait for initial connection attempt | expand |
Hello: This patch was applied to netdev/net-next.git (main) by Jakub Kicinski <kuba@kernel.org>: On Mon, 6 May 2024 13:43:16 +0200 you wrote: > Netdev CI reports occasional failures with this test > ("ERROR: ns2-dX6bUE did not pick up tcp connection from peer"). > > Add explicit busywait call until the initial connection attempt shows > up in conntrack rather than a one-shot 'must exist' check. > > Signed-off-by: Florian Westphal <fw@strlen.de> > > [...] Here is the summary with links: - [net-next] selftests: netfilter: conntrack_tcp_unreplied.sh: wait for initial connection attempt https://git.kernel.org/netdev/net-next/c/76508154d7da You are awesome, thank you!
diff --git a/tools/testing/selftests/net/netfilter/conntrack_tcp_unreplied.sh b/tools/testing/selftests/net/netfilter/conntrack_tcp_unreplied.sh index 1f862c089028..121ea93c0178 100755 --- a/tools/testing/selftests/net/netfilter/conntrack_tcp_unreplied.sh +++ b/tools/testing/selftests/net/netfilter/conntrack_tcp_unreplied.sh @@ -106,6 +106,23 @@ ip netns exec "$ns1" bash -c 'for i in $(seq 1 $BUSYWAIT_TIMEOUT) ; do sleep 0.1 done' & +wait_for_attempt() +{ + count=$(ip netns exec "$ns2" conntrack -L -p tcp --dport 80 2>/dev/null | wc -l) + if [ "$count" -gt 0 ]; then + return 0 + fi + + return 1 +} + +# wait for conntrack to pick the new connection request up before loading +# the nat redirect rule. +if ! busywait "$BUSYWAIT_TIMEOUT" wait_for_attempt; then + echo "ERROR: $ns2 did not pick up tcp connection from peer" + exit 1 +fi + ip netns exec "$ns2" nft -f - <<EOF table inet nat { chain prerouting { @@ -119,12 +136,6 @@ if [ $? -ne 0 ]; then exit 1 fi -count=$(ip netns exec "$ns2" conntrack -L -p tcp --dport 80 2>/dev/null | wc -l) -if [ "$count" -eq 0 ]; then - echo "ERROR: $ns2 did not pick up tcp connection from peer" - exit 1 -fi - wait_for_redirect() { count=$(ip netns exec "$ns2" conntrack -L -p tcp --reply-port-src 8080 2>/dev/null | wc -l) @@ -136,7 +147,7 @@ wait_for_redirect() } echo "INFO: NAT redirect added in ns $ns2, waiting for $BUSYWAIT_TIMEOUT ms for nat to take effect" -busywait $BUSYWAIT_TIMEOUT wait_for_redirect +busywait "$BUSYWAIT_TIMEOUT" wait_for_redirect ret=$? expect="packets 1 bytes 60"
Netdev CI reports occasional failures with this test ("ERROR: ns2-dX6bUE did not pick up tcp connection from peer"). Add explicit busywait call until the initial connection attempt shows up in conntrack rather than a one-shot 'must exist' check. Signed-off-by: Florian Westphal <fw@strlen.de> --- .../net/netfilter/conntrack_tcp_unreplied.sh | 25 +++++++++++++------ 1 file changed, 18 insertions(+), 7 deletions(-)