LNST-developers October 2016

lnst-developers@lists.fedorahosted.org

8 participants
32 discussions

[RFC LNST 0/4] create functions without logic in run time
by Nogah Frankel 10 Nov '16

10 Nov '16

When one want to add a new function to the slave, one needs to add it also to the slave communicator (NetTestSlave), and to the controller side that control it (Macine) and to Task, that controls it. While some of this functions have logic, most of them are just calling a function of an object that they holds, usually, with the same name. Those function can, and I think they should, be manufactured in run time for the following reasons: * it shorten the code considerably (new code is 25% of the old one). * it makes adding new function much shorter. * it singles out the function that do have logic in them. There are however some down sides to it: * code is a bit harder to read at first. * Inheriting function is a bit more tricky. This is just a start and it can be expanded for: * more classes with functions with this problem * setter-getters without inner logic * function with arguments with default values I know that LNST is now in freeze and I don't know how much of it will be relevant after it. But in case it is, I will be glad to hear your opinions on it in the meantime. Nogah Frankel (4): Machine: Interface: create RPC calls functions in run time Slave: NetTestSlave: SlaveMethods: create dev functions calls in run time Slave: NetTestSlave: SlaveMethods: create bridge function calls in run time Task: InterfaceAPI: create interface calls functions in run time lnst/Controller/Machine.py | 106 ++++----------- lnst/Controller/Task.py | 93 ++++--------- lnst/Slave/NetTestSlave.py | 326 +++++++++------------------------------------ 3 files changed, 113 insertions(+), 412 deletions(-) -- 2.4.3

2 6

[PATCH v2 0/3] Netperf message size
by Kamil Jerabek 03 Nov '16

03 Nov '16

Patch 1 rename udp_size option to msg_size in Netperf module. Patch 2 rename udp_size to msg_size in regression_tests/phase1-2. Patch 3 adds set msg_size to tcp netperf tests in regression_tests/phase1-2. Changes in v2: * fixed rebase conflicts Kamil Jerabek (3): Netperf: rename udp_size option to msg_size regression_tests: rename udp_size to msg_size in netperf tests regression_tests: add set msg_size to tcp netperf tests recipes/regression_tests/phase1/3_vlans.py | 26 +++++++++----- .../regression_tests/phase1/3_vlans_over_bond.py | 26 +++++++++----- recipes/regression_tests/phase1/bonding_test.py | 26 +++++++++----- recipes/regression_tests/phase1/simple_netperf.py | 24 ++++++++----- .../phase1/virtual_bridge_2_vlans_over_bond.py | 26 +++++++++----- .../phase1/virtual_bridge_vlan_in_guest.py | 26 +++++++++----- .../phase1/virtual_bridge_vlan_in_host.py | 26 +++++++++----- .../regression_tests/phase2/3_vlans_over_team.py | 26 +++++++++----- recipes/regression_tests/phase2/team_test.py | 42 +++++++++++++++------- ...l_ovs_bridge_2_vlans_over_active_backup_bond.py | 26 +++++++++----- .../phase2/virtual_ovs_bridge_vlan_in_guest.py | 26 +++++++++----- .../phase2/virtual_ovs_bridge_vlan_in_host.py | 26 +++++++++----- test_modules/Netperf.py | 10 +++--- 13 files changed, 231 insertions(+), 105 deletions(-) -- 2.5.5

2 4

[PATCH 0/5] Add switchdev SPAN recipe
by Yotam Gigi 31 Oct '16

31 Oct '16

Add the SPAN port mirroring recipe for switchdev, and several lnst features that were missing and were needed for that recipe. Among the features added: - The lnst slave now cleans the tc qdiscs and filters at the clean procedure. - The PacketAssert module supports setting promiscuos flag. - The Switchdev TestLib IcmpPing module supports setting ping parameters. The SPAN recipe consist of two machines connected via a bridge in the switch, and pinging each other, while ingress/egress port mirrors are added and tested. Yotam Gigi (5): Controller: Task: Add documentation for one of the flags in run func. switchdev: TestLib: Add controllable parameters to IcmpPing module. PacketAssert: Add the promiscuous optional param. Slave: Add basic support for tc. recipes: switchdev: Add the SPAN recipe. lnst/Controller/Task.py | 2 + lnst/Slave/InterfaceManager.py | 28 +++++++++ recipes/switchdev/TestLib.py | 10 ++-- recipes/switchdev/l2-021-span.py | 119 ++++++++++++++++++++++++++++++++++++++ recipes/switchdev/l2-021-span.xml | 26 +++++++++ test_modules/PacketAssert.py | 6 +- 6 files changed, 185 insertions(+), 6 deletions(-) create mode 100644 recipes/switchdev/l2-021-span.py create mode 100644 recipes/switchdev/l2-021-span.xml -- 2.4.11

3 16

Re: LNST upstream meeting reschedule
by Jan Tluka 31 Oct '16

31 Oct '16

Mon, Oct 31, 2016 at 09:56:43AM CET, jiri(a)mellanox.com wrote: >Mon, Oct 31, 2016 at 09:53:14AM CET, idosch(a)mellanox.com wrote: >>On Mon, Oct 31, 2016 at 09:39:50AM +0100, Ondrej Lichtner wrote: >>> Hi, >>> >>> we have an LNST upstream meeting scheduled for today, 15.00 - 16.00 CET >>> time. However I have to be in school during that so I'd like to >>> reschedule. >>> >>> I'm ok with Tue, Wed, Thu for the same time slot, we can also move it to >>> Monday next week, or decide on a different time. >>> >>> What works for you guys? >> >>Wed is OK for Elad, Nogah, Yotam and me. Jiri, works for you as well? > >Works for me. Ok, thanks for response. I moved the meeting to Wed same time 15-16:00. -Jan

1 0

Re: LNST upstream meeting reschedule
by Jiri Prochazka 31 Oct '16

31 Oct '16

Sorry, I'll be in hospital on MRI. Please, do a recording and send it to me afterwards. Thanks, Jiri 2016-10-31 9:56 GMT+01:00 Jiri Pirko <jiri(a)mellanox.com>: > Mon, Oct 31, 2016 at 09:53:14AM CET, idosch(a)mellanox.com wrote: > >On Mon, Oct 31, 2016 at 09:39:50AM +0100, Ondrej Lichtner wrote: > >> Hi, > >> > >> we have an LNST upstream meeting scheduled for today, 15.00 - 16.00 CET > >> time. However I have to be in school during that so I'd like to > >> reschedule. > >> > >> I'm ok with Tue, Wed, Thu for the same time slot, we can also move it to > >> Monday next week, or decide on a different time. > >> > >> What works for you guys? > > > >Wed is OK for Elad, Nogah, Yotam and me. Jiri, works for you as well? > > Works for me. >

1 0

LNST upstream meeting reschedule
by Ondrej Lichtner 31 Oct '16

31 Oct '16

Hi, we have an LNST upstream meeting scheduled for today, 15.00 - 16.00 CET time. However I have to be in school during that so I'd like to reschedule. I'm ok with Tue, Wed, Thu for the same time slot, we can also move it to Monday next week, or decide on a different time. What works for you guys? -Ondrej

3 2

[PATCH] test_modules: report LOCAL_SEND_SIZE and REMOTE_RECV_SIZE in Netperf module
by Jan Tluka 27 Oct '16

27 Oct '16

For omni tests this information is not displayed anywhere so let's include it in netperf's output. Signed-off-by: Jan Tluka <jtluka(a)redhat.com> --- test_modules/Netperf.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/test_modules/Netperf.py b/test_modules/Netperf.py index 3c07922..023bd9a 100644 --- a/test_modules/Netperf.py +++ b/test_modules/Netperf.py @@ -134,7 +134,7 @@ class Netperf(TestGeneric): # Print only relevant output if self._is_omni(): - cmd += ' -- -k "THROUGHPUT, LOCAL_CPU_UTIL, REMOTE_CPU_UTIL, CONFIDENCE_LEVEL, THROUGHPUT_CONFID"' + cmd += ' -- -k "THROUGHPUT, LOCAL_CPU_UTIL, REMOTE_CPU_UTIL, CONFIDENCE_LEVEL, THROUGHPUT_CONFID, LOCAL_SEND_SIZE, REMOTE_RECV_SIZE"' if self._testoptions: if self._is_omni(): -- 2.7.4

2 2

[PATCH] test_modules: fix cpu_util comparison in Netperf non-omni output parsing
by Jan Tluka 27 Oct '16

27 Oct '16

Test module get_opt() returns string representation of None rather than None value so we need to compare the attribute to 'None'. Without this the module will always return failure and throughput value will be silently ignored. Fixes: 6ec96ab7 ("Netperf: fix SCTP_STREAM test") Signed-off-by: Jan Tluka <jtluka(a)redhat.com> --- test_modules/Netperf.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/test_modules/Netperf.py b/test_modules/Netperf.py index 5ab0815..3c07922 100644 --- a/test_modules/Netperf.py +++ b/test_modules/Netperf.py @@ -218,7 +218,7 @@ class Netperf(TestGeneric): # pattern for SCTP streams and other tests # decimal decimal decimal float (float) pattern = "\d+\s+\d+\s+\d+\s+\d+\.\d+\s+(\d+(?:\.\d+){0,1})" - if self._cpu_util: + if self._cpu_util != 'None': # cpu utilization data in format: float float pattern += "\s+(\d+(?:\.\d+){0,1})\s+(\d+(?:\.\d+){0,1})" @@ -228,7 +228,7 @@ class Netperf(TestGeneric): rate_in_kb = 0.0 else: rate_in_kb = float(r2.group(1)) - if self._cpu_util: + if self._cpu_util != 'None': res_val["LOCAL_CPU_UTIL"] = float(r2.group(2)) res_val["REMOTE_CPU_UTIL"] = float(r2.group(3)) -- 2.7.4

2 2

[PATCH] regression_tests phase3: add driver alias
by Ondrej Lichtner 27 Oct '16

27 Oct '16

From: Ondrej Lichtner <olichtne(a)redhat.com> Recipes in phase3 were missing the driver alias used to control which interfaces should be matched for a specific test run. This commit fixes that. Signed-off-by: Ondrej Lichtner <olichtne(a)redhat.com> --- recipes/regression_tests/phase3/2_virt_ovs_vxlan.xml | 7 +++++++ recipes/regression_tests/phase3/novirt_ovs_vxlan.xml | 7 +++++++ recipes/regression_tests/phase3/simple_macsec.xml | 7 +++++++ recipes/regression_tests/phase3/vxlan_multicast.xml | 10 +++++++++- recipes/regression_tests/phase3/vxlan_remote.xml | 7 +++++++ 5 files changed, 37 insertions(+), 1 deletion(-) diff --git a/recipes/regression_tests/phase3/2_virt_ovs_vxlan.xml b/recipes/regression_tests/phase3/2_virt_ovs_vxlan.xml index 009d648..2861acb 100644 --- a/recipes/regression_tests/phase3/2_virt_ovs_vxlan.xml +++ b/recipes/regression_tests/phase3/2_virt_ovs_vxlan.xml @@ -13,6 +13,7 @@ <alias name="net" value="192.168.2"/> <alias name="vxlan_net" value="192.168.100"/> <alias name="vxlan_net6" value="fc00:0:0:0"/> + <alias name="driver" value=""/> </define> <network> <host id="h1"> @@ -21,6 +22,9 @@ </params> <interfaces> <eth id="if1" label="n1"> + <params> + <param name="driver" value="{$driver}"/> + </params> <addresses> <address value="{$net}.1/24"/> </addresses> @@ -83,6 +87,9 @@ </params> <interfaces> <eth id="if1" label="n1"> + <params> + <param name="driver" value="{$driver}"/> + </params> <addresses> <address value="{$net}.2/24"/> </addresses> diff --git a/recipes/regression_tests/phase3/novirt_ovs_vxlan.xml b/recipes/regression_tests/phase3/novirt_ovs_vxlan.xml index 98f5f1b..bf2d387 100644 --- a/recipes/regression_tests/phase3/novirt_ovs_vxlan.xml +++ b/recipes/regression_tests/phase3/novirt_ovs_vxlan.xml @@ -13,11 +13,15 @@ <alias name="net" value="192.168.2"/> <alias name="vxlan_net" value="192.168.100"/> <alias name="vxlan_net6" value="fc00:0:0:0"/> + <alias name="driver" value=""/> </define> <network> <host id="test_host1"> <interfaces> <eth id="if1" label="n1"> + <params> + <param name="driver" value="{$driver}"/> + </params> <addresses> <address value="{$net}.1/24"/> </addresses> @@ -51,6 +55,9 @@ <host id="test_host2"> <interfaces> <eth id="if1" label="n1"> + <params> + <param name="driver" value="{$driver}"/> + </params> <addresses> <address value="{$net}.2/24"/> </addresses> diff --git a/recipes/regression_tests/phase3/simple_macsec.xml b/recipes/regression_tests/phase3/simple_macsec.xml index 83fbccc..2209af6 100644 --- a/recipes/regression_tests/phase3/simple_macsec.xml +++ b/recipes/regression_tests/phase3/simple_macsec.xml @@ -11,11 +11,15 @@ <alias name="nperf_max_dev" value="20%"/> <alias name="mapping_file" value="simple_macsec.mapping" /> <alias name="net" value="192.168.0" /> + <alias name="driver" value=""/> </define> <network> <host id="machine1"> <interfaces> <eth id="eth" label="tnet"> + <params> + <param name="driver" value="{$driver}"/> + </params> <addresses> <address>{$net}.1/24</address> </addresses> @@ -25,6 +29,9 @@ <host id="machine2"> <interfaces> <eth id="eth" label="tnet"> + <params> + <param name="driver" value="{$driver}"/> + </params> <addresses> <address>{$net}.2/24</address> </addresses> diff --git a/recipes/regression_tests/phase3/vxlan_multicast.xml b/recipes/regression_tests/phase3/vxlan_multicast.xml index 734c854..5021652 100644 --- a/recipes/regression_tests/phase3/vxlan_multicast.xml +++ b/recipes/regression_tests/phase3/vxlan_multicast.xml @@ -13,6 +13,7 @@ <alias name="net" value="192.168.0"/> <alias name="vxlan_net" value="192.168.100"/> <alias name="vxlan_net6" value="fc00:0:0:0"/> + <alias name="driver" value=""/> </define> <network> <host id="testmachine1"> @@ -20,7 +21,11 @@ <param name="machine_type" value="baremetal"/> </params> <interfaces> - <eth id="eth1" label="tnet"/> + <eth id="eth1" label="tnet"> + <params> + <param name="driver" value="{$driver}"/> + </params> + </eth> <eth id="tap1" label="to_guest1"/> <bridge id="br0"> <slaves> @@ -74,6 +79,9 @@ </params> <interfaces> <eth id="eth1" label="tnet"> + <params> + <param name="driver" value="{$driver}"/> + </params> <addresses> <address value="{$net}.3/24" /> </addresses> diff --git a/recipes/regression_tests/phase3/vxlan_remote.xml b/recipes/regression_tests/phase3/vxlan_remote.xml index 040479e..824c839 100644 --- a/recipes/regression_tests/phase3/vxlan_remote.xml +++ b/recipes/regression_tests/phase3/vxlan_remote.xml @@ -13,11 +13,15 @@ <alias name="net" value="192.168.0"/> <alias name="vxlan_net" value="192.168.100"/> <alias name="vxlan_net6" value="fc00:0:0:0"/> + <alias name="driver" value=""/> </define> <network> <host id="testmachine1"> <interfaces> <eth id="eth" label="tnet"> + <params> + <param name="driver" value="{$driver}"/> + </params> <addresses> <address value="{$net}.1/24" /> </addresses> @@ -40,6 +44,9 @@ <host id="testmachine2"> <interfaces> <eth id="eth" label="tnet"> + <params> + <param name="driver" value="{$driver}"/> + </params> <addresses> <address value="{$net}.2/24" /> </addresses> -- 2.10.1

1 0

[PATCH v2] regression_tests: add set_mtu to virtual ovs bridge tests
by Kamil Jerabek 27 Oct '16

27 Oct '16

This commit adds missing set_mtu to virtual ovs bridge tests in our regression tests in phase2. Signed-off-by: Kamil Jerabek <kjerabek(a)redhat.com> --- ...ual_ovs_bridge_2_vlans_over_active_backup_bond.py | 20 ++++++++++++++++++++ ...al_ovs_bridge_2_vlans_over_active_backup_bond.xml | 1 + .../phase2/virtual_ovs_bridge_vlan_in_guest.py | 12 ++++++++++++ .../phase2/virtual_ovs_bridge_vlan_in_guest.xml | 1 + .../phase2/virtual_ovs_bridge_vlan_in_host.py | 11 +++++++++++ .../phase2/virtual_ovs_bridge_vlan_in_host.xml | 1 + 6 files changed, 46 insertions(+) diff --git a/recipes/regression_tests/phase2/virtual_ovs_bridge_2_vlans_over_active_backup_bond.py b/recipes/regression_tests/phase2/virtual_ovs_bridge_2_vlans_over_active_backup_bond.py index 409cb88..c2f3b78 100644 --- a/recipes/regression_tests/phase2/virtual_ovs_bridge_2_vlans_over_active_backup_bond.py +++ b/recipes/regression_tests/phase2/virtual_ovs_bridge_2_vlans_over_active_backup_bond.py @@ -33,6 +33,7 @@ g4.sync_resources(modules=["IcmpPing", "Icmp6Ping", "Netperf"]) # TESTS # ------ +mtu = ctl.get_alias("mtu") ipv = ctl.get_alias("ipv") netperf_duration = int(ctl.get_alias("netperf_duration")) nperf_reserve = int(ctl.get_alias("nperf_reserve")) @@ -230,6 +231,25 @@ ping_mod6_bad2 = ctl.get_module("Icmp6Ping", "interval" : 0.1 }) + +# configure mtu +h1_nic1.set_mtu(mtu) +h1_nic2.set_mtu(mtu) +h1.get_interface("tap1").set_mtu(mtu) +h1.get_interface("tap2").set_mtu(mtu) +h1.get_interface("bridge").set_mtu(mtu) + +h2_nic1.set_mtu(mtu) +h2_nic2.set_mtu(mtu) +h2.get_interface("tap1").set_mtu(mtu) +h2.get_interface("tap2").set_mtu(mtu) +h2.get_interface("bridge").set_mtu(mtu) + +g1.get_interface("guestnic").set_mtu(mtu) +g2.get_interface("guestnic").set_mtu(mtu) +g3.get_interface("guestnic").set_mtu(mtu) +g4.get_interface("guestnic").set_mtu(mtu) + ctl.wait(15) for setting in offload_settings: diff --git a/recipes/regression_tests/phase2/virtual_ovs_bridge_2_vlans_over_active_backup_bond.xml b/recipes/regression_tests/phase2/virtual_ovs_bridge_2_vlans_over_active_backup_bond.xml index 32b4c06..01c4291 100644 --- a/recipes/regression_tests/phase2/virtual_ovs_bridge_2_vlans_over_active_backup_bond.xml +++ b/recipes/regression_tests/phase2/virtual_ovs_bridge_2_vlans_over_active_backup_bond.xml @@ -7,6 +7,7 @@ <alias name="nperf_max_runs" value="5" /> <alias name="nperf_mode" value="default"/> <alias name="nperf_num_parallel" value="2"/> + <alias name="mtu" value="1500" /> <alias name="nperf_debug" value="0"/> <alias name="nperf_max_dev" value="20%"/> <alias name="mapping_file" value="virtual_ovs_bridge_2_vlans_over_active_backup_bond.mapping" /> diff --git a/recipes/regression_tests/phase2/virtual_ovs_bridge_vlan_in_guest.py b/recipes/regression_tests/phase2/virtual_ovs_bridge_vlan_in_guest.py index b0dc574..812bc27 100644 --- a/recipes/regression_tests/phase2/virtual_ovs_bridge_vlan_in_guest.py +++ b/recipes/regression_tests/phase2/virtual_ovs_bridge_vlan_in_guest.py @@ -27,6 +27,7 @@ h2.sync_resources(modules=["IcmpPing", "Icmp6Ping", "Netperf"]) # TESTS # ------ +mtu = ctl.get_alias("mtu") ipv = ctl.get_alias("ipv") netperf_duration = int(ctl.get_alias("netperf_duration")) nperf_reserve = int(ctl.get_alias("nperf_reserve")) @@ -178,6 +179,17 @@ if nperf_udp_size is not None: netperf_cli_udp.update_options({"udp_size" : nperf_udp_size}) netperf_cli_udp6.update_options({"udp_size" : nperf_udp_size}) +#set mtu +h1.get_interface("nic").set_mtu(mtu) +h1.get_interface("tap").set_mtu(mtu) +h1.get_interface("br").set_mtu(mtu) + +h2.get_interface("nic").set_mtu(mtu) +h2.get_interface("vlan10").set_mtu(mtu) + +g1.get_interface("guestnic").set_mtu(mtu) +g1.get_interface("vlan10").set_mtu(mtu) + ctl.wait(15) for setting in offload_settings: diff --git a/recipes/regression_tests/phase2/virtual_ovs_bridge_vlan_in_guest.xml b/recipes/regression_tests/phase2/virtual_ovs_bridge_vlan_in_guest.xml index ae85997..786b9d3 100644 --- a/recipes/regression_tests/phase2/virtual_ovs_bridge_vlan_in_guest.xml +++ b/recipes/regression_tests/phase2/virtual_ovs_bridge_vlan_in_guest.xml @@ -9,6 +9,7 @@ <alias name="nperf_num_parallel" value="2"/> <alias name="nperf_debug" value="0"/> <alias name="nperf_max_dev" value="20%"/> + <alias name="mtu" value="1500" /> <alias name="mapping_file" value="virtual_ovs_bridge_vlan_in_guest.mapping" /> <alias name="vlan10_net" value="192.168.10"/> <alias name="vlan10_tag" value="10"/> diff --git a/recipes/regression_tests/phase2/virtual_ovs_bridge_vlan_in_host.py b/recipes/regression_tests/phase2/virtual_ovs_bridge_vlan_in_host.py index 25f19f1..69560b9 100644 --- a/recipes/regression_tests/phase2/virtual_ovs_bridge_vlan_in_host.py +++ b/recipes/regression_tests/phase2/virtual_ovs_bridge_vlan_in_host.py @@ -27,6 +27,7 @@ h2.sync_resources(modules=["IcmpPing", "Icmp6Ping", "Netperf"]) # TESTS # ------ +mtu = ctl.get_alias("mtu") ipv = ctl.get_alias("ipv") netperf_duration = int(ctl.get_alias("netperf_duration")) nperf_reserve = int(ctl.get_alias("nperf_reserve")) @@ -177,6 +178,16 @@ if nperf_udp_size is not None: netperf_cli_udp.update_options({"udp_size" : nperf_udp_size}) netperf_cli_udp6.update_options({"udp_size" : nperf_udp_size}) +#set mtu +h1.get_interface("nic").set_mtu(mtu) +h1.get_interface("tap").set_mtu(mtu) +h1.get_interface("ovs_br").set_mtu(mtu) + +h2.get_interface("nic").set_mtu(mtu) +h2.get_interface("vlan10").set_mtu(mtu) + +g1.get_interface("guestnic").set_mtu(mtu) + ctl.wait(15) for setting in offload_settings: diff --git a/recipes/regression_tests/phase2/virtual_ovs_bridge_vlan_in_host.xml b/recipes/regression_tests/phase2/virtual_ovs_bridge_vlan_in_host.xml index 6b2a271..5f8b9e5 100644 --- a/recipes/regression_tests/phase2/virtual_ovs_bridge_vlan_in_host.xml +++ b/recipes/regression_tests/phase2/virtual_ovs_bridge_vlan_in_host.xml @@ -9,6 +9,7 @@ <alias name="nperf_num_parallel" value="2"/> <alias name="nperf_debug" value="0"/> <alias name="nperf_max_dev" value="20%"/> + <alias name="mtu" value="1500" /> <alias name="mapping_file" value="virtual_ovs_bridge_vlan_in_host.mapping" /> <alias name="vlan10_net" value="192.168.10"/> <alias name="vlan10_tag" value="10"/> -- 2.5.5

3 2

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

LNST-developers October 2016