| *** yamamoto has quit IRC | 01:19 | |
| *** yamamoto has joined #openstack-neutron-ovn | 01:26 | |
| *** gongysh has joined #openstack-neutron-ovn | 01:44 | |
| *** saphi has joined #openstack-neutron-ovn | 02:23 | |
| *** saphi has quit IRC | 02:24 | |
| *** saphi has joined #openstack-neutron-ovn | 02:25 | |
| *** saphi has quit IRC | 03:57 | |
| *** gongysh has quit IRC | 04:03 | |
| *** saphi has joined #openstack-neutron-ovn | 04:13 | |
| *** gongysh has joined #openstack-neutron-ovn | 04:39 | |
| *** zefferno has joined #openstack-neutron-ovn | 05:09 | |
| *** yamamoto has quit IRC | 05:19 | |
| *** yamamoto has joined #openstack-neutron-ovn | 05:31 | |
| *** janki has joined #openstack-neutron-ovn | 05:37 | |
| *** yamamoto_ has joined #openstack-neutron-ovn | 06:05 | |
| *** yamamoto has quit IRC | 06:09 | |
| *** livelace has quit IRC | 06:34 | |
| *** pcaruana has joined #openstack-neutron-ovn | 07:24 | |
| *** anilvenkata has joined #openstack-neutron-ovn | 07:38 | |
| *** dasanind has quit IRC | 08:47 | |
| *** ltomasbo has quit IRC | 08:47 | |
| *** yamamoto_ has quit IRC | 08:47 | |
| *** masber has quit IRC | 08:47 | |
| *** gizmoguy has quit IRC | 08:47 | |
| *** numans has quit IRC | 08:47 | |
| *** gongysh has quit IRC | 08:47 | |
| *** afranc has quit IRC | 08:47 | |
| *** mestery has quit IRC | 08:47 | |
| *** lucas-afk has quit IRC | 08:47 | |
| *** flaviof has quit IRC | 08:47 | |
| *** switchcade has quit IRC | 08:47 | |
| *** dalvarez has quit IRC | 08:47 | |
| *** rpb has quit IRC | 08:47 | |
| *** vaishali has quit IRC | 08:47 | |
| *** basilAB has quit IRC | 08:47 | |
| *** pcaruana has quit IRC | 08:47 | |
| *** carl_baldwin has quit IRC | 08:47 | |
| *** elurkki has quit IRC | 08:47 | |
| *** leifmadsen has quit IRC | 08:47 | |
| *** fkautz has quit IRC | 08:47 | |
| *** lrichard has quit IRC | 08:47 | |
| *** markmcclain has quit IRC | 08:47 | |
| *** jagter has quit IRC | 08:47 | |
| *** kevinbenton has quit IRC | 08:47 | |
| *** safchain has quit IRC | 08:47 | |
| *** anilvenkata has quit IRC | 08:47 | |
| *** zefferno has quit IRC | 08:47 | |
| *** saphi has quit IRC | 08:47 | |
| *** openstackgerrit has quit IRC | 08:47 | |
| *** igordcard has quit IRC | 08:47 | |
| *** amotoki has quit IRC | 08:47 | |
| *** openstack has joined #openstack-neutron-ovn | 13:53 | |
| *** janki has quit IRC | 14:00 | |
| *** anilvenkata has quit IRC | 14:17 | |
| *** anilvenkata has joined #openstack-neutron-ovn | 14:19 | |
| *** trinaths has joined #openstack-neutron-ovn | 14:39 | |
| otherwiseguy | russellb, The odd thing is, I don't think the string PKG_CHECK_MODULES should actually be in configure itself...it's as if the generation of the configure script is broken. but i have a xenial vm and am not reproducing. | 14:40 |
|---|---|---|
| trinaths | hi | 14:40 |
| russellb | otherwiseguy: right, it's a macro .. | 14:41 |
| russellb | autotools \o/ | 14:41 |
| otherwiseguy | yeah, it's quite odd. Running recheck just to see if it goes away (though it has been there for several runs). The check queue is at 235...so that should take a while. | 14:50 |
| *** yamamoto has quit IRC | 15:43 | |
| *** yamamoto has joined #openstack-neutron-ovn | 15:45 | |
| *** yamamoto has quit IRC | 15:45 | |
| *** yamamoto has joined #openstack-neutron-ovn | 15:46 | |
| *** yamamoto has quit IRC | 15:51 | |
| *** yamamoto has joined #openstack-neutron-ovn | 15:57 | |
| *** yamamoto has quit IRC | 16:01 | |
| *** trinaths has left #openstack-neutron-ovn | 16:11 | |
| *** lucasagomes is now known as lucas-afk | 16:14 | |
| *** mjblack has joined #openstack-neutron-ovn | 16:21 | |
| openstackgerrit | Miguel Angel Ajo proposed openstack/networking-ovn master: Support for L3 gateway HA https://review.openstack.org/486971 | 16:22 |
| *** yamamoto has joined #openstack-neutron-ovn | 16:58 | |
| *** yamamoto has quit IRC | 17:03 | |
| openstackgerrit | Merged openstack/networking-ovn master: Rename 'ns-metadata-proxy' config dir to 'ovn-metadata-proxy' https://review.openstack.org/486637 | 17:22 |
| *** yamamoto has joined #openstack-neutron-ovn | 17:59 | |
| russellb | otherwiseguy: any luck narrowing it down? | 18:00 |
| *** yamamoto has quit IRC | 18:04 | |
| openstackgerrit | venkata anil proposed openstack/networking-ovn master: schedule gateway on chassis with external connectivity https://review.openstack.org/485176 | 18:11 |
| otherwiseguy | russellb, looks like the latest run magically passed...so who knows. | 18:19 |
| russellb | ¯\_(ツ)_/¯ | 18:20 |
| otherwiseguy | russellb, actually, just saw this from commit yesterday: 5dc451b65fa567716b930da88ef440023df082fb | 18:21 |
| otherwiseguy | So apparently Ben fixed it. | 18:22 |
| russellb | ok | 18:22 |
| otherwiseguy | which would explain why I couldn't reproduce. :p | 18:22 |
| otherwiseguy | THIS SHOULD WORK! (because it is already fixed) | 18:22 |
| * otherwiseguy sighs | 18:22 | |
| * otherwiseguy has been reading over the syncing proposal | 18:23 | |
| russellb | cool - i like it | 18:23 |
| russellb | what i don't like: that gnome-terminal keeps crashing and i haven't figured out why | 18:23 |
| otherwiseguy | i'm pretty sure i like it. just trying to think of alternatives. | 18:23 |
| otherwiseguy | russellb, there are lots of things built-in to the OVSDB for handling some of the issues. the verify() call in the IDL, the NB_Global nb_cfg/sb_cfg/hv_cfg values, etc. | 18:24 |
| otherwiseguy | just trying to think of ways that they could be leveraged. | 18:25 |
| russellb | yes ... verify() should be used on the external-ids to ensure that the version number is still what we think it is | 18:25 |
| russellb | and i thought about the _cfg vars ... problem with syncing anything on those is that they are indeed global | 18:25 |
| russellb | and a global lock is far from ideal | 18:25 |
| otherwiseguy | yeah. | 18:26 |
| otherwiseguy | I mean, you could store the nb_cfg value that you increment as an external_id on the object you create... :p | 18:27 |
| otherwiseguy | russellb, re: syncing. since the solution is row-level locks, we'd still have the issue of order-dependent mutli-object operations occurring on different servers? e.g. network/port create? | 18:51 |
| otherwiseguy | russellb, like what you worked around here: https://review.openstack.org/#/c/404791/2/networking_ovn/ml2/mech_driver.py | 18:52 |
| russellb | yep, this doesn't help that | 18:53 |
| russellb | and i'm not sure journaling did either ... | 18:53 |
| russellb | i haven't thought of anything too great for that except "wait and retry" in the dependent cases where we know the IDL may just need to catch up | 18:53 |
| russellb | :-/ | 18:53 |
| otherwiseguy | russellb, a generic object dependency system :D | 18:55 |
| otherwiseguy | Anything with a foreign key relationship, require version matching for multiple objects? | 18:56 |
| * otherwiseguy is still not really up to date on how all of the oslo versioned object/neutron db stuff really works | 18:58 | |
| *** yamamoto has joined #openstack-neutron-ovn | 19:00 | |
| *** yamamoto has quit IRC | 19:05 | |
| russellb | otherwiseguy: yeah, something like that .. | 19:09 |
| otherwiseguy | russellb, In ovsdbapp, for testing, I created a RowEvent that would essentially do a wait() for an object to be created: https://github.com/openstack/ovsdbapp/blob/master/ovsdbapp/tests/functional/schema/ovn_southbound/test_impl_idl.py#L25 | 19:10 |
| otherwiseguy | Could do something like that. When creating a port, do a wait_for_network() call, etc. | 19:10 |
| otherwiseguy | russellb, used like https://github.com/openstack/ovsdbapp/blob/master/ovsdbapp/tests/functional/schema/ovn_southbound/test_impl_idl.py#L106 | 19:11 |
| russellb | yeah, though we're not really waiting on something to happen on the server side, we're waiting for our client to catch up | 19:12 |
| russellb | so seems odd to issue a server command | 19:13 |
| otherwiseguy | if anyone has time for a review: https://review.openstack.org/#/c/490199/5 needs a second +2 | 19:30 |
| russellb | dalvarez: https://review.openstack.org/#/c/486822/ metadata related patch | 19:35 |
| openstackgerrit | Merged openstack/ovsdbapp master: Add RowView versions of db_find and db_list https://review.openstack.org/490199 | 19:41 |
| otherwiseguy | russellb, thx | 19:44 |
| russellb | np | 19:44 |
| *** yamamoto has joined #openstack-neutron-ovn | 20:02 | |
| *** yamamoto has quit IRC | 20:06 | |
| *** yamamoto_ has joined #openstack-neutron-ovn | 21:03 | |
| *** yamamoto_ has quit IRC | 21:08 | |
| *** yamamoto_ has joined #openstack-neutron-ovn | 21:16 | |
| *** yamamoto_ has quit IRC | 21:21 | |
| *** yamamoto_ has joined #openstack-neutron-ovn | 21:57 | |
| *** mmichelson has quit IRC | 22:59 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!