Fixes: 73b24ec5659f ("Icmp6Ping test") Signed-off-by: Ido Schimmel idosch@mellanox.com --- v1->v2: * No change. --- test_modules/Icmp6Ping.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/test_modules/Icmp6Ping.py b/test_modules/Icmp6Ping.py index 9197b42..405c756 100644 --- a/test_modules/Icmp6Ping.py +++ b/test_modules/Icmp6Ping.py @@ -38,7 +38,7 @@ class Icmp6Ping(TestGeneric): cmd = self.compose_cmd() logging.debug("%s" % cmd)
- limit_rate = self.get_opt("limit_rate ", default=80) + limit_rate = self.get_opt("limit_rate", default=80) data_stdout = exec_cmd(cmd, die_on_err=False)[0]
stat_pttr1 = r'(\d+) packets transmitted, (\d+) received'
We are seeing failures in our nightly regression where ping fails due to rate being lower than limit. The machine logs suggest the first few packets are dropped (while the links stabilize), but the rest go through just fine.
Solve that by lowering the limit rate in order to take into account the initial packet drop.
Signed-off-by: Ido Schimmel idosch@mellanox.com --- v1->v2: * Rename 'rate' to 'limit_rate'. --- recipes/switchdev/TestLib.py | 7 ++++--- recipes/switchdev/l2-005-bridge_bond_failover.py | 8 ++++---- recipes/switchdev/l2-007-bridge_team_failover.py | 8 ++++---- 3 files changed, 12 insertions(+), 11 deletions(-)
diff --git a/recipes/switchdev/TestLib.py b/recipes/switchdev/TestLib.py index 1159a30..68333d8 100644 --- a/recipes/switchdev/TestLib.py +++ b/recipes/switchdev/TestLib.py @@ -57,7 +57,8 @@ class TestLib:
m2.run(linkneg_mod, desc=desc)
- def ping_simple(self, if1, if2, fail_expected=False, desc=None): + def ping_simple(self, if1, if2, fail_expected=False, desc=None, + limit_rate=90): if not desc: desc = self._generate_default_desc(if1, if2)
@@ -73,7 +74,7 @@ class TestLib: "count": 100, "interval": 0.2, "iface" : if1.get_devname(), - "limit_rate": 90}) + "limit_rate": limit_rate})
ping_mod6 = self._ctl.get_module("Icmp6Ping", options={ @@ -81,7 +82,7 @@ class TestLib: "count": 100, "interval": 0.2, "iface" : if1.get_ip(1), - "limit_rate": 90}) + "limit_rate": limit_rate})
if self._ipv in [ 'ipv6', 'both' ]: m1.run(ping_mod6, fail_expected=fail_expected, desc=desc) diff --git a/recipes/switchdev/l2-005-bridge_bond_failover.py b/recipes/switchdev/l2-005-bridge_bond_failover.py index c7b4eca..f5c7c63 100644 --- a/recipes/switchdev/l2-005-bridge_bond_failover.py +++ b/recipes/switchdev/l2-005-bridge_bond_failover.py @@ -33,22 +33,22 @@ def do_task(ctl, hosts, ifaces, aliases): tl.ping_simple(m1_lag1, m2_lag1)
sw_if1.set_link_down() - tl.ping_simple(m1_lag1, m2_lag1) + tl.ping_simple(m1_lag1, m2_lag1, limit_rate=50)
sw_if1.set_link_up() sw_if2.set_link_down() - tl.ping_simple(m1_lag1, m2_lag1) + tl.ping_simple(m1_lag1, m2_lag1, limit_rate=50)
sw_if2.set_link_up() sw_if1.set_link_down() sw_if3.set_link_down() - tl.ping_simple(m1_lag1, m2_lag1) + tl.ping_simple(m1_lag1, m2_lag1, limit_rate=50)
sw_if1.set_link_up() sw_if3.set_link_up() sw_if2.set_link_down() sw_if4.set_link_down() - tl.ping_simple(m1_lag1, m2_lag1) + tl.ping_simple(m1_lag1, m2_lag1, limit_rate=50)
do_task(ctl, [ctl.get_host("machine1"), ctl.get_host("machine2"), diff --git a/recipes/switchdev/l2-007-bridge_team_failover.py b/recipes/switchdev/l2-007-bridge_team_failover.py index 23d1a7c..d3a8ee5 100644 --- a/recipes/switchdev/l2-007-bridge_team_failover.py +++ b/recipes/switchdev/l2-007-bridge_team_failover.py @@ -39,22 +39,22 @@ def do_task(ctl, hosts, ifaces, aliases): tl.ping_simple(m1_lag1, m2_lag1)
sw_if1.set_link_down() - tl.ping_simple(m1_lag1, m2_lag1) + tl.ping_simple(m1_lag1, m2_lag1, limit_rate=50)
sw_if1.set_link_up() sw_if2.set_link_down() - tl.ping_simple(m1_lag1, m2_lag1) + tl.ping_simple(m1_lag1, m2_lag1, limit_rate=50)
sw_if2.set_link_up() sw_if1.set_link_down() sw_if3.set_link_down() - tl.ping_simple(m1_lag1, m2_lag1) + tl.ping_simple(m1_lag1, m2_lag1, limit_rate=50)
sw_if1.set_link_up() sw_if3.set_link_up() sw_if2.set_link_down() sw_if4.set_link_down() - tl.ping_simple(m1_lag1, m2_lag1) + tl.ping_simple(m1_lag1, m2_lag1, limit_rate=50)
do_task(ctl, [ctl.get_host("machine1"), ctl.get_host("machine2"),
Mon, Mar 14, 2016 at 02:07:42PM CET, idosch@mellanox.com wrote:
We are seeing failures in our nightly regression where ping fails due to rate being lower than limit. The machine logs suggest the first few packets are dropped (while the links stabilize), but the rest go through just fine.
Solve that by lowering the limit rate in order to take into account the initial packet drop.
Signed-off-by: Ido Schimmel idosch@mellanox.com
Both applied, thanks!
lnst-developers@lists.fedorahosted.org
-
Ido Schimmel -
Jiri Pirko