Tuesday, 2018-08-07

*** longkb has joined #openstack-fwaas00:37
*** SridarK has joined #openstack-fwaas00:41
*** yushiro4 has joined #openstack-fwaas02:32
yushiro4At last I joined due to some trouble..02:33
yushiro4Hi SridarK.  Are you online?02:33
SridarKyushiro4: hi02:33
*** yushiro4 is now known as yushiro02:34
SridarKyushiro4: are u still having trouble in joining ?02:34
yushiroIdentifying has cleared but connection to freenode are still in trouble :(   I don't know why...02:35
yushiroSo, thanks for your reviewing SridarK and xgerman_ for logging patches.02:35
yushiroI read your comment and now double-checking in devstack environment.02:35
SridarKyushiro: sigh yes i joined easily today but few days ago had issues02:36
SridarKyushiro: no worries - i did not +A02:36
yushiroSridarK, Yeah,  you are waiting for my final judgement :)  Thanks.02:36
SridarKyushiro: i want to be sure we dont land up with some issue in case we are not able to get in all the patches02:36
SridarKyushiro: yes - just being a bit careful - so we dont have to do reverts etc. I see that first 2 patches in the list:02:37
SridarK#link https://review.openstack.org/#/c/53279202:37
SridarK#link https://review.openstack.org/#/c/53071502:38
SridarKstill need some work02:38
yushiroAh, yes.  Thanks for pointing.02:39
yushiroThese patches depend on validator patch in neutron-side.  However, these patches still got -1 from zuul even if validator patch has been merged..02:39
SridarKyushiro: ah yes that confused me since all 3 neutron patches are in02:40
yushiro3 patches don't refer validator patch in neutron-side.  So, their tests has passed.02:41
SridarKokay02:42
yushiroOther 5 patches ( in other words, related logging validator patch) refers Validator in neutron side...02:42
SridarKok02:42
yushiroSridarK, I think that to solve these error, we need to wait cutting branch in neutron.02:42
SridarKyushiro: hmm ok maybe u shd discuss with mlavalle ?02:43
SridarKso from our side we will not +A02:44
SridarKeither xgerman_ or I will wait for u to +A02:44
yushiroSridarK, Maybe yes until new neutron RC-1 would be cut  except 3 patches.02:45
SridarKyushiro: ok02:46
yushiroTherefore, I'll ask mlavalle some question about schedule.02:46
SridarKyushiro: ok02:46
yushiroBTW,02:46
SridarKyushiro: also will try to sync with u around this time or a bit earlier tomorrow as well - in case something needs my attention02:47
yushirohttps://review.openstack.org/#/c/580552/02:47
yushiroSridarK, Thanks for your kindness :)02:47
SridarKyushiro: no issues at all02:47
SridarKyushiro: ah ok u removed the -1 workflow02:47
SridarKdid u get any feedback ?02:47
yushiroSridarK, Yes.  Let me sync up current status of this L3HA issue.02:48
SridarKyushiro: ok02:48
yushiroI found that the issue was we cannot insert iptables rules into 'active' router's namespace.  Even if we specified router-port for L3-HA, the rule has been set into 'standby' router's one.02:49
SridarKThat issue is a bit odd as to how it goes to standby router02:49
yushiroSridarK, Yes.  Maybe we need to verify the router's status by calling is_active() or some method in neutron..02:50
yushiroAh, or we can set firewall rules into both 'active' and 'standby'.02:51
SridarKyushiro: hmm ok - u think we are setting it on to standy02:51
SridarKyushiro: maybe setting in both is needed - so it is there when a switchover happens02:51
SridarKi am a bit unsure on the L3HA model tood02:52
SridarK*too02:52
yushiroBTW, I've checked switchover.  When switchover has occurred,  SSH was kept on connecting.02:52
yushiro(But there is no firewall rules)02:52
SridarKyushiro: oh02:53
SridarKshould we not hit the default deny if the fwg is active ?02:54
SridarKoh u mean there is no fwg at all02:54
yushiroSridarK, sorry.  Some word made you confuse..  My environment was as follows.   (standby router with default fwg rule)  and (active router with no default rule)02:56
SridarKok02:56
yushirovm0 --(SSH)--> vm1  :  When switchover, this connection is kept because default fwaas rules(from vm0) are allow02:57
yushirovm0 <----(ssh)--- vm1  : When switchover, this connection was reset beacuse default fwaas rules(from vm1) are deny.02:57
SridarKvm0 and vm1 are they on different networks so there is a path thru the router ?02:58
SridarKi think w.r.t default fwg (02:58
yushiroYes, different subnet:    vm0 ---( net0 ) ---- (router0) ---- (net1) ---- vm102:58
SridarKok02:58
SridarKand we really want to check the effect of the fwd on the router port02:59
SridarK*fwg on the router port02:59
yushirovm0 ---( net0 ) ----(fwg)(router0, standby) ---- (net1) ---- vm103:00
yushiro^^^^ my environment03:00
SridarKok03:00
yushirovm0 ---( net0 ) ----(router0, active) ---- (net1) ---- vm103:00
yushiro^^^^ switchover environment03:00
SridarKnow there is no fwg03:01
yushiroSridarK, Yes, you're right.  We need to test fwg on port with Active router and switchover don't you?03:01
SridarKi think it all goes thru because there is no fwg03:01
SridarKso at L3 port level there is no filtering at all03:01
yushiroSridarK, Ah sorry.  I draw wrong picture.03:02
yushirovm0 ---( net0 ) ----(router0, active) ---- (net1) ---- vm103:02
SridarKyushiro: yes i think we should test fwg on Active router and then do switchover - agree03:02
yushiro^^^ Initial status.03:02
SridarKok03:02
yushirovm0 ---( net0 ) ----(fwg)(router0, standby->active) ---- (net1) ---- vm103:03
yushiro^^^ switchover (after)03:03
SridarKhmm ok so the fwg was never applied on the active (before swover)03:04
yushiroSridarK, Yes!! that is current behavior..03:04
SridarKbecause it got appled on the standby ?03:04
yushiroYes03:04
SridarKso initially all traffic is passed because no fwg03:04
yushiroSridarK, Yes, that's correct.  No guard.03:05
SridarKbut after swover - the fwg is on the old standby and new active03:05
yushiroYes.  That was my result for current implemantation.03:05
SridarKyushiro: ok thx for the detailed explanation03:06
SridarKyushiro: i am not sure if we need to apply on both active and standby03:06
yushiroSridarK, NP, haha I need more practice to explain to you more clearly.03:06
SridarKyushiro: oh no worries at all - u were very detailed - sorry i got confused :-)03:07
yushiroSridarK, Yes, I'm considering it too.  I'm not sure this issue caused by fwaas side or L3-HA side03:07
SridarKyushiro: ok it seems this bug is a bit more tricky - which was my concern03:07
SridarKyushiro: yes - depends on how they sync features03:07
SridarKif we are responsible to do it03:08
yushiroSridarK, +10 Yeeeess..03:08
*** bbbbzhao_ has quit IRC03:08
SridarKyushiro: ok - lets see if we can get some info or we may have to address this in Stein if it gets complex03:09
yushiroSridarK, Sure.  Next, I'd like to test for sync behavior between active - standby router.  E.g. If we applied fwg rule into only active router and switchover, what would happen.03:10
SridarKah yes - we will know if the fwg gets synced over to standby03:11
yushiroAnd, we can ask L3-Ha guys for more detail impl as well :)03:11
SridarKthat should tell us if fwaas is responsible to manage the apply of fwg on active and standbys03:12
SridarKyushiro: +103:12
SridarKok we can sync up tomorrow - or shoot me an email if u have found some thing and i will look too03:12
yushiroSridarK, Yes.  We definitely should clarify about that.03:12
SridarKyushiro: ok agreed03:12
yushiroAh, SridarK , regarding fwg logging feature, what do you think to handle 3 patches which passed zuul job?  I tested and these patches are OK.03:13
SridarKyushiro: which ones ? The one with +2 ?03:14
*** jhesketh_ is now known as jhesketh03:14
yushiroSridarK, yes, the patches which +2 from me.03:14
SridarKhttps://review.openstack.org/#/c/57871803:15
SridarK?03:15
SridarKand the other 203:15
yushiroYes, this one and below 2 patches.03:15
SridarKhttps://review.openstack.org/#/c/57633803:16
yushiroYes, and https://review.openstack.org/#/c/574128/2503:16
yushiroHowever, other 5 patches are got an issue.03:17
SridarKah ok i have +2 on all these 303:17
SridarKif u think we are ok - pls +A03:17
SridarKbut if for some reason the other patches are not able to get in03:17
SridarKwhat will happen ?03:18
SridarKwill it cause an issue ?03:18
SridarKand i assume there is no dependency03:18
yushiroSridarK, yes, there is no dependency.  It is just applied some parts and no effect I think.03:19
SridarKyushiro: ok then if it is safe - i think it is ok to +A03:19
SridarKIf for some reason there is an issue and we are not able to merge some patches by the deadline03:20
SridarKit will be ok if these patches are in ?03:20
yushiroSridarK, Yes. Even if the only 3 patches are merged, there is 1 effect to add 'neutron-l3-agent-dropped/accepted' on router namespace but no issue.03:24
SridarKyushiro: ok03:24
SridarKthen i think it is ok to +A03:24
yushiroSridarK, OK, so I'll test it again and ask Miguel about cutting Rc-1 in neutron.03:24
SridarKyushiro: perfect03:25
yushiroAfter that, I'll put +A .03:25
SridarKsounds good03:25
yushiroIf there is no issue :p03:25
yushiroAnd will sync up with you.03:25
yushiroIf you are not on IRC, will send e-mail.03:25
SridarKyushiro: ok sounds good - i will step away now - but will check back here or just send an email03:26
SridarKyushiro: many thx03:26
yushiroSridarK, You too! :)03:26
SridarKok talk later03:27
yushiroya03:27
*** yushiro is now known as yushiro_lunch03:30
openstackgerritAkihiro Motoki proposed openstack/neutron-fwaas-dashboard master: Unblock Django 2.0 in requirements.txt  https://review.openstack.org/58933203:47
*** annp has joined #openstack-fwaas04:19
*** yushiro_lunch is now known as yushiro05:32
*** velizarx has joined #openstack-fwaas06:56
*** yushiro has quit IRC07:00
*** yushiro has joined #openstack-fwaas07:00
openstackgerritMerged openstack/neutron-fwaas-dashboard master: Unblock Django 2.0 in requirements.txt  https://review.openstack.org/58933207:04
*** velizarx has quit IRC07:13
*** reedip has joined #openstack-fwaas07:20
*** velizarx has joined #openstack-fwaas07:24
*** yamamoto has joined #openstack-fwaas08:00
*** yamamoto has quit IRC08:07
*** yushiro has quit IRC08:18
openstackgerritYushiro FURUKAWA proposed openstack/neutron-fwaas master: Add dependency for generating reno  https://review.openstack.org/58939708:29
*** yushiro has joined #openstack-fwaas08:40
*** yamamoto has joined #openstack-fwaas08:46
*** velizarx has quit IRC09:03
*** velizarx has joined #openstack-fwaas09:07
*** yamamoto has quit IRC09:11
openstackgerritKim Bao Long proposed openstack/neutron-fwaas master: Add log validator for FWaaS v2 logging  https://review.openstack.org/53279209:29
amotokiyushiro: hi, do we need reno dependency in tox venv now?09:31
yushiroamotoki, Ah, sorry.  I forgot to put workflow -1.  It's ok to land to next cycle.09:31
yushiroI wanted to generate reno.09:32
amotokiyushiro: if so, you can use 'docs' tox env09:32
amotokiyushiro: or you can add 'releasenotes' tox env rather than adding doc/requirements.txt to 'venv' env09:33
amotokithis is what we do in neutorn09:34
amotoki*neutron09:34
yushiroamotoki, I see.  So, I'll abandon this patch and follow neutron's way.  Thank you.09:35
yushiroI used 'tox -e venv -- reno new hogehoge' for a long time :p.09:35
amotokiyushiro: ah, i see. I use '.tox/releasenotes/bin/reno new foo' instead09:37
yushiroamotoki, BTW, why neutron defined 'reno' in docs/requirements.txt and test-requirements.txt ?09:38
amotokiyushiro: looking at the commit history, boden re-added reno to test-reqs.txt. this allows us to follow reno doc instruction [1]09:43
amotoki [1] https://docs.openstack.org/reno/latest/user/usage.html09:43
amotokibut I am not sure it is the right direction. perhaps we should update reno docs.09:44
amotokior your solution might be more straight forward09:44
*** reedip has quit IRC09:45
yushiroamotoki, Ah, I just referred ironic one and [1]'s page.  'tox -e venv -- reno new hoge'09:45
yushiroAnyway, I just wanted to generate reno :)  Thanks for your kindness.09:46
amotokiyushiro: you're welcom09:46
amotoki*e09:46
yushiroamotoki, Ah!! I wanted to ask you cutting branch in neutron.09:46
amotokiyushiro: go ahead09:47
yushiroCurrently, we're implementing fwaas logging feature.  Some codes are merged into neutron side.  In addition, some codes refers neutron patches from fwaas side.09:48
yushiroBasically, fwaas installs neutron package instead of pulling the master source code.  So, fwaas code cannot refer the master code of neutron.09:49
amotokiyushiro: do such patches depends on neutron features after RC3?09:50
amotokis/RC3/Rocky-3/09:50
amotokithere are several ways to address the situation09:51
amotokione is to consume neutron R-3 package if it is okay.09:51
amotokianother way is to specify required-projects in .zuul.yaml to consume the last neutron (master or stable branch head)09:52
yushiroamotoki, yes, after Rocky-3.  Merged last weekend...09:52
amotokiif so, it sounds reasonable to specify required-projects in .zuul.yaml09:54
amotokiit pulls neutron master09:54
amotokiyushiro: for example, https://review.openstack.org/#/c/530715/ ?09:55
amotokii might be wrong09:55
yushiroamotoki, yes, correct.  haha you read my mind.09:55
amotokiyou are talking about l-c job?09:56
yushiroamotoki, ahhh yes..09:57
amotokiin this case, there is no release which satisfies your need, so required-projects is the only solution to me.09:58
yushiroSorry, makes you confuse.  l-c job failed..09:58
hoangcxamotoki: yushiro I think We need neutron released with RC-1 (by this week?) and then remaining depended patch in fwaas can be land during RC-209:59
amotokihoangcx: RC2 is limited to bug fixes strictly, so I don't think it works for you09:59
hoangcxamotoki: Is it possible to land more patches in RC-2 for fwaas regarding to FFE?09:59
amotokiyushiro: hoangcx: neutron-fwaas-dashboard .zuul.yaml does similar thing http://git.openstack.org/cgit/openstack/neutron-fwaas-dashboard/tree/.zuul.yaml#n410:00
amotokiit always consumes the latest horizon10:00
amotokiI believe this is what you would like to do.10:00
yushiroamotoki, Aha.  That's good.  Yes, that is what I'd like to do :)10:01
hoangcxamotoki: Yes. I understand this solution :) Will try to invoke with your suggestion10:01
amotokineutron and fwaas are released together, so actually there is no need to care lower-constraints (lower bound of neutron)10:01
yushirohoangcx, Oh, you will ?? :p10:01
hoangcxyushiro: yeah :p10:01
yushirohoangcx, Thanks very very much.10:02
amotokiFYI: we will cut RC1 Thurday10:02
amotokiperhaps I will prepare a release patch around Thurday afternoon in Japan TZ and will wait some updates around Thurday in PCT TZ10:03
yushiroamotoki, Yes...  Tomorrow's morning, we'll tell Miguel about our progress and judge land or not..10:03
amotokiyushiro: makes sense :)10:04
yushiroganbaru10:09
*** yushiro is now known as yushiro_dinner10:09
openstackgerritCao Xuan Hoang proposed openstack/neutron-fwaas master: Add neutron to required-projects of l-c job  https://review.openstack.org/58942110:18
openstackgerritCao Xuan Hoang proposed openstack/neutron-fwaas master: Add neutron to required-projects of l-c job  https://review.openstack.org/58942110:20
openstackgerritCao Xuan Hoang proposed openstack/neutron-fwaas master: Add log validator for FWaaS v2 logging  https://review.openstack.org/53279210:20
* xgerman_ reading scrollback10:38
amotokihoangcx: could you explain more direct reason on your l-c change?10:40
*** yushiro_dinner has quit IRC10:40
hoangcxamotoki: Sure. I will do it right now10:41
openstackgerritCao Xuan Hoang proposed openstack/neutron-fwaas master: Add neutron to required-projects of l-c job  https://review.openstack.org/58942110:45
*** velizarx has quit IRC10:46
hoangcxamotoki: ^^10:47
hoangcxamotoki: Thanks for your time and your suggestion :)10:47
hoangcxamotoki: http://logs.openstack.org/92/532792/25/check/openstack-tox-lower-constraints/ea2d3bb/ the job passed :)10:48
*** velizarx has joined #openstack-fwaas10:59
hoangcxamotoki: Is it possible do the same for build-openstack-sphinx-docs job?11:00
hoangcxamotoki: the job failed caused by the latest head commit also: http://logs.openstack.org/92/532792/25/check/build-openstack-sphinx-docs/47d3575/job-output.txt.gz#_2018-08-07_10_42_02_52081011:00
*** yushiro has joined #openstack-fwaas11:06
yushirohoangcx, Thanks for your patch.11:08
yushiroI'll read IRC log.11:08
hoangcxyushiro: I have checked that it is the same with build-openstack-sphinx-docs job11:14
yushirohoangcx, Aha.  So, we need to add 'require-project' for build-openstack-sphinx-docs as well, right ?11:16
hoangcxyushiro: I'm waiting amotoki feedback11:16
yushirohoangcx, OK, thanks.  Ah, did you tell annp about that ?  Maybe he will rebase his patches.11:18
openstackgerritCao Xuan Hoang proposed openstack/neutron-fwaas master: Add neutron to required-projects of l-c job  https://review.openstack.org/58942111:19
openstackgerritCao Xuan Hoang proposed openstack/neutron-fwaas master: Add neutron to required-projects of l-c job  https://review.openstack.org/58942111:24
openstackgerritCao Xuan Hoang proposed openstack/neutron-fwaas master: Add log validator for FWaaS v2 logging  https://review.openstack.org/53279211:24
hoangcxyushiro: yes. I talked to him11:24
hoangcxyushiro: I update the l-c patch to see if  build-openstack-sphinx-docs job passed11:25
*** yushiro has quit IRC12:29
*** velizarx has quit IRC12:34
*** velizarx has joined #openstack-fwaas12:35
*** yushiro has joined #openstack-fwaas13:10
yushiroping hoangcx13:11
*** longkb has quit IRC13:13
*** velizarx has quit IRC13:16
*** velizarx has joined #openstack-fwaas13:16
yushiroping SridarK xgerman_13:22
amotokiyushiro: do you know why sphinx job is related in https://review.openstack.org/#/c/589421/ ?13:33
openstackgerritYushiro FURUKAWA proposed openstack/neutron-fwaas master: FWaaS v2: L3 logging agent extension  https://review.openstack.org/57633813:34
openstackgerritYushiro FURUKAWA proposed openstack/neutron-fwaas master: Introduce accepted/dropped/rejected chains for future processing  https://review.openstack.org/57412813:34
openstackgerritYushiro FURUKAWA proposed openstack/neutron-fwaas master: Add notification callback events for FWaaS v2  https://review.openstack.org/57871813:34
openstackgerritYushiro FURUKAWA proposed openstack/neutron-fwaas master: Add neutron as required-projects  https://review.openstack.org/58942113:34
openstackgerritYushiro FURUKAWA proposed openstack/neutron-fwaas master: Add log validator for FWaaS v2 logging  https://review.openstack.org/53279213:34
yushiroamotoki, hi13:36
amotokiyushiro: hi13:36
yushiroamotoki, AFAIK, regarding sphinx job,  this job try to make some relation for object/class.  So, if we referred some module in master branch of neutron, it failed.13:38
amotokiyushiro: is this due to the module reference?13:40
yushiroamotoki, I think so.  Coudld you take a look http://logs.openstack.org/92/532792/26/check/build-openstack-sphinx-docs/4ca2757/job-output.txt.gz#_2018-08-07_11_33_43_72380013:42
yushirofwg_validate.py tried to call @validators.ResourceValidateRequest.register(log_const.FIREWALL_GROUP) and @validators is at neutron.13:44
amotokiyushiro: thanks for the pointer. At a quick look, there is no surprise if you refer to something from the neutron master in the module reference.13:44
yushiroamotoki, Thanks.13:46
yushiroReferring master branch of neutron, all patches will work fine.13:47
yushiroThank you so much.  It is not over, I'll relax, carefully review and merge existing patches :p13:48
amotokithanks. it makes sense. it's clear to me now13:49
yushiro:)14:33
*** longkb has joined #openstack-fwaas14:35
*** hongbin has joined #openstack-fwaas14:38
*** longkb has quit IRC15:05
openstackgerritKim Bao Long proposed openstack/neutron-fwaas master: Adding resources callback handler for FWaaS v2 logging  https://review.openstack.org/58097615:28
*** yushiro has quit IRC16:13
*** velizarx has quit IRC16:47
*** SridarK has quit IRC17:05
*** Swami has joined #openstack-fwaas18:12
*** openstackgerrit has quit IRC18:49
-openstackstatus- NOTICE: Due to a bug, Zuul has been unable to report on cherry-picked changes over the last 24 hours. This has now been fixed; if you encounter a cherry-picked change missing its results (or was unable to merge), please recheck now.20:42
*** openstackgerrit has joined #openstack-fwaas22:15
openstackgerritMerged openstack/neutron-fwaas master: FWaaS v2: L3 logging agent extension  https://review.openstack.org/57633822:15
openstackgerritMerged openstack/neutron-fwaas master: Introduce accepted/dropped/rejected chains for future processing  https://review.openstack.org/57412822:15
*** hongbin has quit IRC22:39
*** Swami has quit IRC23:46

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!