| *** zenoway has joined #openstack-dragonflow | 00:12 | |
| *** zenoway has quit IRC | 00:16 | |
| *** zenoway has joined #openstack-dragonflow | 00:48 | |
| *** zenoway has quit IRC | 00:54 | |
| *** zenoway has joined #openstack-dragonflow | 01:03 | |
| *** zenoway has quit IRC | 01:08 | |
| openstackgerrit | heshan proposed openstack/dragonflow: code for local controller reliability https://review.openstack.org/292060 | 01:57 |
|---|---|---|
| *** gongysh has quit IRC | 02:22 | |
| *** gongysh has joined #openstack-dragonflow | 02:22 | |
| *** gongysh has quit IRC | 02:32 | |
| *** gongysh has joined #openstack-dragonflow | 02:32 | |
| *** yamamoto_ has joined #openstack-dragonflow | 02:37 | |
| *** gongysh has quit IRC | 03:01 | |
| *** gongysh has joined #openstack-dragonflow | 03:08 | |
| *** yamamoto_ has quit IRC | 03:20 | |
| *** gongysh has quit IRC | 03:30 | |
| *** gongysh has joined #openstack-dragonflow | 03:32 | |
| *** yamamoto_ has joined #openstack-dragonflow | 04:10 | |
| *** irenab has quit IRC | 04:37 | |
| *** oanson has joined #openstack-dragonflow | 05:03 | |
| openstackgerrit | Omer Anson proposed openstack/dragonflow: use IDLs built-in notify for ovsdb monitor https://review.openstack.org/301691 | 05:11 |
| openstackgerrit | Omer Anson proposed openstack/dragonflow: Allow filtering of tables and columns in OVS DB monitor https://review.openstack.org/302228 | 05:11 |
| *** gongysh has quit IRC | 05:17 | |
| *** gongysh has joined #openstack-dragonflow | 05:19 | |
| *** gampel1 has joined #openstack-dragonflow | 05:35 | |
| *** irenab has joined #openstack-dragonflow | 05:36 | |
| gampel1 | oanson: ) hujie: whats the status with the idl sync_start syn_finish | 05:49 |
| gampel1 | Frank_Duan: ping | 05:50 |
| gampel1 | dingboopt: ping | 05:52 |
| openstackgerrit | Merged openstack/dragonflow: Add control plane test doc https://review.openstack.org/282873 | 06:05 |
| gampel1 | hujie: ping | 06:07 |
| dingboopt | gampel1: pong | 06:09 |
| hujie | I'm back now, we are noon rest just now:) | 06:09 |
| gampel1 | Hi how are you ? hi whats the status off the SG patch | 06:10 |
| *** gongysh has quit IRC | 06:13 | |
| dingboopt | I am busy with other things these days. And I will pick up SG patch soon. Sorry for the delay | 06:13 |
| gampel1 | ok talk with me after the rest | 06:13 |
| gampel1 | sorry | 06:13 |
| hujie | we have waked up:) | 06:13 |
| *** gongysh has joined #openstack-dragonflow | 06:14 | |
| hujie | we can work now, it's 14:15 in Beijing | 06:14 |
| gampel1 | Ok we discussed the sync finish/start | 06:14 |
| gampel1 | and gal have an idea that i think could work | 06:15 |
| gampel1 | If we read all the port form the idl before running the monitor | 06:16 |
| gampel1 | update them all into the application and then send syen_finishe and then start the monitor | 06:17 |
| hujie | I think current patch has to monitor all tables, lack sync_start and sync_finish mechanism and filter patch and tunnel ports, maybe we need more work:) | 06:17 |
| oanson | patch and tunnel ports are filtered. | 06:18 |
| gampel1 | I am not sure i understand , @ | 06:18 |
| gampel1 | omer are you here | 06:18 |
| oanson | Filter for specific tables is done | 06:18 |
| gampel1 | yes it is done in a patch on top | 06:18 |
| *** gongysh has quit IRC | 06:18 | |
| oanson | Patch: https://review.openstack.org/#/c/302228/ | 06:18 |
| hujie | ok I see there are other patched to resolve the problem? | 06:18 |
| oanson | I am working on the sync_start and sync_finish now, as per Gal's suggestion. | 06:19 |
| gampel1 | https://review.openstack.org/#/c/302228/ | 06:19 |
| oanson | I'll have it done today. | 06:19 |
| gampel1 | hujie: I think that if we resolve the sync start/finish we should use the new patches introduce by omer | 06:21 |
| gampel1 | hujie: what do you think ? | 06:21 |
| gampel1 | Frank_Duan: are you here | 06:21 |
| hujie | I have thought about the ideal that Gal suggest before I wrote ovsdb monitor code | 06:22 |
| gampel1 | @Omer will do it, we need to focus on the other features we want to merge before code freeze | 06:23 |
| gampel1 | is Frank_Duan there i do not know whats the status of the SG patch | 06:23 |
| hujie | I think it is a bit complicate, but if we could not find better method we can use it, maybe I could think about the sequence of this method | 06:24 |
| gampel1 | I am not sure i understand we will not loose any messages , the worst case is that we get each port twice | 06:26 |
| hujie | duankebo will be back soon | 06:26 |
| gampel1 | ok do you know the status of teh SG | 06:26 |
| gampel1 | is there someone taking over dingboopt patch we need to merge it today | 06:26 |
| hujie | ok wait a moment please | 06:26 |
| hujie | I'll ask yuanwei | 06:27 |
| Frank_Duan | Hi | 06:27 |
| gampel1 | hi how are you? | 06:27 |
| gampel1 | we have 3 item to discuss | 06:27 |
| hshan | gampel: reliability feature is ready to merge | 06:28 |
| Frank_Duan | Fine, thank you | 06:28 |
| gampel1 | 1) the Ovs Monitor sync start/finishe | 06:28 |
| gampel1 | 2) SG status | 06:28 |
| Frank_Duan | Yuanwei has updated the sg patch yestoday | 06:28 |
| gampel1 | 2) reliability feature | 06:28 |
| gampel1 | 3) | 06:28 |
| gampel1 | https://review.openstack.org/#/c/280538/ | 06:29 |
| gampel1 | it was not updated since | 06:29 |
| gampel1 | Uploaded patch set 15.Mar 29 3:42 PM | 06:29 |
| *** gongysh has joined #openstack-dragonflow | 06:30 | |
| Frank_Duan | Or, my mistake, he is updating. | 06:30 |
| gampel1 | hshan: did you modify the l3 app using the Cookie | 06:31 |
| hshan | yes | 06:31 |
| gampel1 | Ok lets all reveiw it today | 06:32 |
| hshan | ok | 06:32 |
| gampel1 | gsagie, oanson : please review the reliability patches | 06:32 |
| oanson | Will do | 06:32 |
| gsagie | ok | 06:33 |
| gampel1 | Frank_Duan: we discuss with hujie , we think we have a solution for sync start end andmonitoring only what we need | 06:33 |
| hujie | I have a question, when do you read all the existing port info from IDL cache? | 06:33 |
| gampel1 | when do you think we will have a new patch of SG we need to test it a bit before meraging | 06:34 |
| oanson | hujie, I think the DF controller does it. | 06:34 |
| hujie | how do you make sure all the existing ovs port have been in the IDL cache when you read it | 06:34 |
| oanson | There is a call to vswitch_api.sync, which calls idl.run, which reads all the information from the OVS DB | 06:35 |
| hujie | I'm worried when you read IDL cache the data is not the same to ovsdb itself, maybe we will have conclusion after we review your new patch:) | 06:36 |
| gampel1 | we will not loose those event they will be in the queue | 06:36 |
| gampel1 | we read after calling sync on the idl | 06:37 |
| hujie | when you invoke idl.run() you have begined to receive message by your overrided notify() method | 06:37 |
| hujie | I mean the IDL cache update and your receive message by notify() method is working at the same time | 06:39 |
| gsagie | hujie: we are talking about a process that happens before we replace the notify or register for events | 06:40 |
| gsagie | lets say i dont have your code, you are saying that after i run idl.run() the idl is not synced? | 06:40 |
| gsagie | i think its syncing it in a synchronouse matter | 06:41 |
| gsagie | right? | 06:41 |
| oanson | Once the notify function is overriden, no. | 06:42 |
| hujie | ok Omer got it:) | 06:42 |
| oanson | There is a different thread that continuously calls idl.run | 06:42 |
| gsagie | ok, so if you override it, and you call idl.run() dont we get all events? or one event at a time? | 06:42 |
| oanson | hujie, It is possible that we will see the same event twice. | 06:43 |
| hujie | Omer when will you commit your new patch? | 06:43 |
| oanson | gsagie, we get only the differences | 06:43 |
| hujie | twice event is not problem before we could resolve the problem:) | 06:44 |
| oanson | hujie, I will have the patch ready soon. | 06:44 |
| oanson | Could you please explain what you mean? | 06:45 |
| openstackgerrit | Nguyen Hung Phuong proposed openstack/dragonflow: Fix typos in Dragonflow files https://review.openstack.org/302538 | 06:45 |
| hujie | @@ | 06:45 |
| oanson | hujie, in any case, once we see that the OVSDB monitor works well, we can remove the relevant parts from the sync loop in DF controller, and rely only on the OVSDB monitor. | 06:46 |
| hujie | just now you say There is a different thread that continuously calls idl.run | 06:47 |
| hujie | I'll think about it:) | 06:47 |
| hujie | what I worried is the IDL cache update and your receive message by notify() method is working at the same time, It is no longer synchronouse mode after the notify function is overriden | 06:49 |
| hujie | if you have consider this and resolve it in new patch, that 's ok:) | 06:49 |
| gsagie | hujie: all events finally reach the df-controller, so if we are worried about same events we can "filter" them there | 06:50 |
| gsagie | you just need to know if port up or not | 06:50 |
| oanson | hujie: The handling of events is all done in the same thread. That's synchronous. | 06:50 |
| oanson | even if the events come in from different ports. | 06:50 |
| oanson | I can verify the DF controller poll-sync loop to verify it behaves correctly, | 06:51 |
| oanson | but 1) We are working with greenthreads. Things don't happen in parallel. | 06:51 |
| gsagie | I personally think that we can register the OVSDB monitor after we do the sync, we will get new events (regarding new ports that were added) when we finally run the ovsdb monitor thread and call idl.run() again | 06:52 |
| oanson | This is doubly true in python code, due to the GIL. | 06:52 |
| openstackgerrit | Omer Anson proposed openstack/dragonflow: Add sync_started and sync_finished notification events https://review.openstack.org/302540 | 06:52 |
| oanson | ^^^^ This is the patch for sync_started and sync_finished | 06:52 |
| hujie | gsagie: I'm worried about read data from IDL cache is not complete:) | 06:56 |
| hujie | because it is not synchronouse mode when you invoke idl.run | 06:57 |
| hujie | let's review Omer's new patch to see what has happened:) | 06:58 |
| gsagie | okie :) | 06:58 |
| gampel1 | Yuanwei: hi | 06:59 |
| openstackgerrit | yuan wei proposed openstack/dragonflow: This patch implements security group app https://review.openstack.org/280538 | 06:59 |
| openstackgerrit | Omer Anson proposed openstack/dragonflow: Add publisher's table https://review.openstack.org/297167 | 07:00 |
| gampel1 | Yuanwei: will you update this patch as well patch https://review.openstack.org/#/c/299105/ thx | 07:01 |
| gampel1 | hshan: hi can you please see why the CI test are failing | 07:13 |
| hshan | yes, I'm checking that now | 07:16 |
| openstackgerrit | Nguyen Hung Phuong proposed openstack/dragonflow: Fix typos in Dragonflow files https://review.openstack.org/302538 | 07:19 |
| hujie | Ok, I see it, currently you try to read data from IDL cache, I think the data is complete, but is the notify overrided late? or maybe we will lose new online port notify message | 07:31 |
| oanson | The notify is overriden before the read of the IDL cache. | 07:31 |
| oanson | If an interface is added before - it will be in the cache. | 07:32 |
| oanson | If an interface is added after - we will be notified. | 07:32 |
| oanson | If an interface is added in between - we will get the event twice: Once from the cache, and one from the notification. | 07:32 |
| oanson | The third case is very unlikely | 07:32 |
| hujie | ok I see it | 07:33 |
| hujie | good job:) | 07:34 |
| hujie | monitor just one table, you need to modify Neutron code? | 07:34 |
| hujie | why filter patch and tunnel port? | 07:35 |
| hujie | I think we need them | 07:35 |
| yuanwei | OK.I will update patch https://review.openstack.org/#/c/299105/ as soon as I can | 07:36 |
| oanson | I thought we only wanted events on VM ports. | 07:37 |
| hujie | ok please let me explain it:) | 07:38 |
| todin | morning, I am testing dragonflow with dist. DNAT, how would be the br-ex interface of the compute node connected to the physical network interstuctur? | 07:38 |
| hujie | yesterday I said we need to cache ofport info of vm, patch, tunnel port into our own cache, not IDL cache, so we can improve our speed of search, do you agree? | 07:39 |
| hujie | and vlan feature designed by Liuhaixia need the patch port | 07:40 |
| hujie | many of our search operations are the tunnel port search because of remote vm online/offline | 07:42 |
| yuanwei | Please review & test locally patch https://review.openstack.org/280538, thx. We wish we could merge this patch to master this weekend:) | 07:42 |
| gsagie | yuanwei: testing now | 07:45 |
| gampel1 | oanson: ping | 07:47 |
| yuanwei | gsagie: thx:) | 07:49 |
| *** zenoway has joined #openstack-dragonflow | 07:50 | |
| nick-ma | ovsdb local address needs root privilege for access. anyone encounter this problem? https://bugs.launchpad.net/dragonflow/+bug/1567248 | 07:50 |
| openstack | Launchpad bug 1567248 in DragonFlow "No access to ovsdb local address" [High,New] | 07:50 |
| gampel1 | nick-ma: it did not happend to me | 07:52 |
| oanson | pong | 07:53 |
| oanson | nick-ma, on what distribution did this happen? | 07:54 |
| gampel1 | is it new, maybe it is related to the OVS service install | 07:54 |
| nick-ma | oanson: Yes, ubuntu server 14.04 with OVS service patch. | 07:55 |
| nick-ma | oanson: I don't know if it is related with the OVS service patch. I'm testing. | 07:56 |
| oanson | And you say changing it to /var/run/openvswitch/db.sock solved it? | 07:56 |
| oanson | I think it is | 07:56 |
| hujie | oanson: what's your opinion about my reply? | 07:56 |
| oanson | I think the solution is to try and find the 'default' value, as ovsdb-client --help reports it. | 07:56 |
| nick-ma | nop. the ovsdb local address by default is /usr/local/var/. I have to change it to /var/. | 07:56 |
| oanson | nick-ma, I think installing it via the OS as done in the OVS service install patch changes that | 07:57 |
| oanson | hujie, I missed it :). Yes, you are right. | 07:58 |
| oanson | I forgot we need ofport info for patch and tunnel. | 07:58 |
| nick-ma | When I run df-local-controller, it cannot connect to OVSDB because the local address is wrong by default. And when I changed it to /var/, which is reported by ovsdb-client, it shows 'No Access'. Then I rerun controller by sudo, the problem is solved. | 07:59 |
| oanson | Could you please run 'ls -l /var/run/openvswitch/db.sock'? | 07:59 |
| oanson | We may need to update the permissions on that file. | 08:00 |
| nick-ma | srwxr-x--- 1 root root 0 Apr 7 12:22 /var/run/openvswitch/db.sock. | 08:00 |
| nick-ma | definitely, root permission. | 08:00 |
| oanson | Yes. | 08:00 |
| hujie | ok:) | 08:01 |
| oanson | In https://review.openstack.org/#/c/301691/ we go back to use the same connection as the DF controller, which still uses IP. | 08:01 |
| oanson | But it would be better if that went via the unix socket as well. | 08:01 |
| oanson | nick-ma, Could you also run 'id', when logged in as stack? | 08:05 |
| nick-ma | ok. uid=1001(stack) gid=1001(stack) groups=1001(stack),117(libvirtd) | 08:05 |
| gampel1 | hujie: unless there are any objection i am going to merger the IDL patches | 08:09 |
| hujie | ok | 08:11 |
| gampel1 | can you all please review https://review.openstack.org/#/c/302228/2 | 08:12 |
| gampel1 | and https://review.openstack.org/#/c/302540/1 | 08:12 |
| gampel1 | hujie: please review 302228 and 303540 | 08:23 |
| openstackgerrit | Merged openstack/dragonflow: use IDLs built-in notify for ovsdb monitor https://review.openstack.org/301691 | 08:25 |
| gampel1 | gsagie, nick-ma please look at 302228 and 303540 we want to merge them as soon as possible | 08:26 |
| nick-ma | gampel1: ok | 08:28 |
| gampel1 | thx | 08:29 |
| openstackgerrit | Eran Gampel proposed openstack/dragonflow: Enable selective distribution with Pub/Sub https://review.openstack.org/302595 | 08:29 |
| *** yuli_s has joined #openstack-dragonflow | 08:31 | |
| yuli_s | hello all | 08:31 |
| gampel1 | hi | 08:31 |
| gsagie | yuanwei: i dont know if its related to your version but i cant get L3 to work | 08:31 |
| gsagie | i added 2 networks connected them to router and add a VM at each of them and try to ping (default security group with added rules for ping) | 08:32 |
| gsagie | i will try to see if this also happens on master Dragonflow | 08:32 |
| hujie | ok | 08:34 |
| yuanwei | ok, i will also try this locally | 08:35 |
| openstackgerrit | Merged openstack/dragonflow: Fix typos in Dragonflow files https://review.openstack.org/302538 | 08:36 |
| gsagie | btw yuanwei i am counting on your vote for the TC ;) | 08:43 |
| gampel1 | hi please review https://review.openstack.org/#/c/297167/7 | 08:43 |
| yuli_s | i want to talk about this patch: https://review.openstack.org/#/c/298131/3 | 08:44 |
| yuli_s | especially this file: https://review.openstack.org/#/c/298131/3/dragonflow/tests/common/utils.py | 08:44 |
| yuli_s | i added wait_until_true() funtion that executes the neutron.agent.linux.utils.wait_until_true() function | 08:45 |
| yuli_s | I change the default timeout value | 08:45 |
| yuli_s | this was the point when I say that this fn make the code more maintanable | 08:45 |
| gampel1 | you just wrapped there function with yours i do not see the added value | 08:46 |
| oanson | You can give the timeout explicitly | 08:47 |
| oanson | It would make the code of the calling function more clear | 08:47 |
| oanson | I don't think changing a default value warrants writing a function that does nothing but wrap an identical other function. | 08:47 |
| openstackgerrit | Eran Gampel proposed openstack/dragonflow: Send Chassis update on SEND_ALL topic https://review.openstack.org/302601 | 08:47 |
| yuli_s | ok, | 08:47 |
| yuanwei | gsagie: OK, no problem:) | 08:48 |
| yuanwei | besides, in your test, those two VMs are both in the default security group? I tried locally, and found L3 works in my environments... using the default security group(not added rules for ping) | 08:49 |
| gsagie | yes | 08:50 |
| gsagie | sec, checking now with master | 08:50 |
| *** gampel1 has quit IRC | 08:51 | |
| gsagie | you use 'demo' as tenant or 'admin' ? | 08:51 |
| yuanwei | admin | 08:51 |
| gsagie | yeah getting in master the problem as well | 08:55 |
| gsagie | i get "ping: sendto: Network is unreachable" | 08:56 |
| gsagie | you use q-dhcp or the DHCP app? | 08:56 |
| yuli_s | u created a network | 08:56 |
| yuli_s | or u placed vm on public network ? | 08:56 |
| gsagie | i created a network | 08:57 |
| yuli_s | i have just run stack with fresh version | 08:58 |
| yuli_s | i am also getting an error of other kind | 08:58 |
| yuli_s | br-int is not a bridge or a socket | 08:58 |
| yuli_s | br-int is not a bridge or a socket | 08:58 |
| openstackgerrit | Yuli proposed openstack/dragonflow: Fix DHCP fullstack tests https://review.openstack.org/298131 | 08:59 |
| yuli_s | i will reboot and run stack again | 08:59 |
| yuanwei | I used DHCP app | 09:02 |
| *** yuli_s has quit IRC | 09:05 | |
| *** yuli_s has joined #openstack-dragonflow | 09:18 | |
| *** yamamoto_ has quit IRC | 09:40 | |
| *** gongysh has quit IRC | 09:49 | |
| *** gongysh has joined #openstack-dragonflow | 09:50 | |
| gampel | yuanwei: ping | 10:09 |
| gampel | i get allot of error messages when testing the SG patch | 10:10 |
| gampel | DEBUG RyuDFAdapter [-] EventOFPErrorMsg received. | 10:10 |
| gampel | version=0x4, msg_type=0x1, msg_len=0x4c, xid=0xd1afc971 | 10:10 |
| gampel | `-- msg_type: OFPT_ERROR(1) | 10:10 |
| gampel | OFPErrorMsg(type=0x2, code=0x0, data=b'\x04\x0e\x00\x68\xd1\xaf\xc9\x71\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x03\x00\x00\x00\x00\x00\x00\x64\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\x00\x00\x00\x00\x00\x | 10:10 |
| gampel | 01\x00\x12\x80\x00\x00\x04\x00\x00\x00\x19\x80\x00\x0a\x02') | 10:10 |
| gampel | |-- type: OFPET_BAD_ACTION(2) | 10:10 |
| gampel | |-- code: OFPBAC_BAD_TYPE(0) from (pid=14905) error_msg_handler /usr/lib/python2.7/site-packages/ryu/controller/ofp_handler.py:264 | 10:10 |
| gampel | 10:10 | |
| gampel | yuanwei: ping | 10:16 |
| *** gongysh has quit IRC | 10:32 | |
| gampel | Frank_Duan:ping | 10:33 |
| yuanwei | . | 10:38 |
| yuanwei | hi gampel | 10:38 |
| yuanwei | what is your SG configuration? | 10:38 |
| oanson | Hi. Using dragonflow master + sg apps, nova reports that there are no flavors. Has anyone else ran into this? Or is this an issue on my setup? | 10:39 |
| gampel | hi | 10:41 |
| oanson | The following tests fail because of that: dragonflow.tests.fullstack.test_l2_multicast.TestL2Multicast.test_vm_multicast dragonflow.tests.fullstack.test_l2_responder.ArpResponderTest.test_arp_responder dragonflow.tests.fullstack.test_topology.TestTopology.test_topology_create_vm dragonflow.tests.fullstack.test_topology.TestTopology.test_topology_create_vm2 | 10:41 |
| gampel | I am running sg and i am getting Ryu errors when installing the flows | 10:41 |
| *** yamamoto has joined #openstack-dragonflow | 10:42 | |
| gampel | DEBUG RyuDFAdapter [-] EventOFPErrorMsg received. | 10:42 |
| gampel | version=0x4, msg_type=0x1, msg_len=0x4c, xid=0xd1afc999 | 10:42 |
| gampel | `-- msg_type: OFPT_ERROR(1) | 10:42 |
| gampel | OFPErrorMsg(type=0x4, code=0x8, data=b'\x04\x0e\x00\x60\xd1\xaf\xc9\x99\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x4d\x00\x00\x00\x00\x00\x00\x02\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\x00\x00\x00\x00\x00\x | 10:42 |
| gampel | 01\x00\x18\x00\x01\x0e\x04\x00\x00\x00\x3b\x00\x01\xd3\x08') | 10:42 |
| gampel | |-- type: OFPET_BAD_MATCH(4) | 10:42 |
| gampel | |-- code: OFPBMC_BAD_MASK(8) from (pid=14905) error_msg_handler /usr/lib/python2.7/site-packages/ryu/controller/ofp_handler.py:264 | 10:42 |
| gampel | 2016-04-07 13:18:59.481 DEBUG RyuDFAdapter [-] `-- data: version=0x4, msg_type=0xe, msg_len=0x60, xid=0xd1afc999 | 10:42 |
| gampel | `-- msg_type: OFPT_FLOW_MOD(14) from (pid=14905) error_msg_handler /usr/lib/python2.7/site-packages/ryu/controller/ofp_handler.py:271 | 10:42 |
| gampel | 2016-04-07 13:18:59.482 DEBUG RyuDFAdapter [-] EventOFPErrorMsg received. | 10:42 |
| gampel | version=0x4, msg_type=0x1, msg_len=0x4c, xid=0xd1afc99a | 10:42 |
| gampel | `-- msg_type: OFPT_ERROR(1) | 10:42 |
| openstackgerrit | Omer Anson proposed openstack/dragonflow: Allow filtering of tables and columns in OVS DB monitor https://review.openstack.org/302228 | 10:42 |
| *** gongysh has joined #openstack-dragonflow | 10:44 | |
| gampel | yuanwei: Do you get the same error | 10:44 |
| yuanwei | no...Do you create some security group rules? | 10:46 |
| yuanwei | what rules do you create? I will try that in my environment | 10:47 |
| yuanwei | sorry..I just check logs in my environment, and i find this error | 10:48 |
| *** gongysh has quit IRC | 11:02 | |
| *** yuanwei has quit IRC | 11:02 | |
| *** yamamoto has quit IRC | 11:04 | |
| *** yamamoto has joined #openstack-dragonflow | 11:06 | |
| *** yuanwei has joined #openstack-dragonflow | 11:09 | |
| gampel | yuanwei: will you take care of this | 11:17 |
| yuanwei | yes, I'm working on that | 11:18 |
| openstackgerrit | Merged openstack/dragonflow: Fix DHCP fullstack tests https://review.openstack.org/298131 | 11:22 |
| gampel | whats the status of https://review.openstack.org/#/c/292060/ | 11:37 |
| gampel | what about https://review.openstack.org/#/c/299105/ | 11:38 |
| gampel | yuanwei: You could continue to work on this two tomorrow , we can try and merge it tomorrow | 11:39 |
| gampel | It is late your time | 11:39 |
| yuanwei | thx:), not too late, it's OK to me | 11:41 |
| openstackgerrit | Omer Anson proposed openstack/dragonflow: Allow filtering of tables and columns in OVS DB monitor https://review.openstack.org/302228 | 11:49 |
| openstackgerrit | Merged openstack/dragonflow: Enable selective distribution with Pub/Sub https://review.openstack.org/302595 | 12:19 |
| *** yuli_s has quit IRC | 12:21 | |
| oanson | Hi, could you please review https://review.openstack.org/#/c/302228/ ? | 12:41 |
| *** yuanwei has quit IRC | 12:41 | |
| *** yuanwei has joined #openstack-dragonflow | 12:46 | |
| *** Mic22 has quit IRC | 12:55 | |
| *** Mic22 has joined #openstack-dragonflow | 12:59 | |
| openstackgerrit | yuan wei proposed openstack/dragonflow: This patch implements security group app https://review.openstack.org/280538 | 13:04 |
| yuanwei | gampel | 13:10 |
| yuanwei | gampel: I found the cause of that problem. It's because of the lack of thought about IPv6 rules..and I update the patch https://review.openstack.org/280538 to fix this bug | 13:10 |
| *** Mic22 has quit IRC | 13:14 | |
| *** yuanwei has quit IRC | 13:22 | |
| *** yuanwei has joined #openstack-dragonflow | 13:22 | |
| *** oanson has quit IRC | 13:22 | |
| *** yamamoto has quit IRC | 13:40 | |
| openstackgerrit | Merged openstack/dragonflow: Allow filtering of tables and columns in OVS DB monitor https://review.openstack.org/302228 | 13:44 |
| *** irenab has quit IRC | 13:44 | |
| *** yamamoto has joined #openstack-dragonflow | 13:57 | |
| *** yamamoto has quit IRC | 14:03 | |
| *** yamamoto has joined #openstack-dragonflow | 14:29 | |
| openstackgerrit | Eran Gampel proposed openstack/dragonflow: Add sync_started and sync_finished notification events https://review.openstack.org/302540 | 14:58 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/dragonflow: Updated from global requirements https://review.openstack.org/302845 | 15:20 |
| *** yamamoto has quit IRC | 16:06 | |
| *** zenoway has quit IRC | 16:09 | |
| *** zenoway has joined #openstack-dragonflow | 16:10 | |
| *** zenoway has quit IRC | 16:11 | |
| *** zenoway has joined #openstack-dragonflow | 16:11 | |
| *** zenoway has quit IRC | 16:15 | |
| *** zenoway has joined #openstack-dragonflow | 16:16 | |
| *** zenoway has quit IRC | 16:20 | |
| *** oanson has joined #openstack-dragonflow | 17:01 | |
| *** oanson has quit IRC | 17:06 | |
| *** yamamoto has joined #openstack-dragonflow | 17:07 | |
| *** yuanwei has quit IRC | 17:10 | |
| *** yuanwei has joined #openstack-dragonflow | 17:11 | |
| *** oanson has joined #openstack-dragonflow | 17:14 | |
| *** yamamoto has quit IRC | 17:16 | |
| *** hujie has quit IRC | 17:52 | |
| *** hujie has joined #openstack-dragonflow | 17:52 | |
| *** yuanwei has quit IRC | 18:07 | |
| *** yuanwei has joined #openstack-dragonflow | 18:07 | |
| *** zenoway has joined #openstack-dragonflow | 18:12 | |
| *** zenoway has quit IRC | 18:16 | |
| todin | hmm, is rule wants to jump to table 30, but there is no table 30 priority=100,ip,metadata=0x3,nw_src=192.168.42.101 actions=resubmit(,30) | 18:19 |
| *** oanson has quit IRC | 18:24 | |
| *** zenoway has joined #openstack-dragonflow | 18:30 | |
| *** gampel has quit IRC | 18:39 | |
| *** CrayZee has quit IRC | 18:39 | |
| *** saggi has quit IRC | 18:40 | |
| *** zenoway has quit IRC | 23:51 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!