sridhar_ram | s3wong: thanks a lot! | 00:02 |
---|---|---|
sridhar_ram | bobh_: current main issue is monitoring failure in new tosca template.. | 00:03 |
openstackgerrit | Merged openstack/tacker: Fix monitoring policy in new sample tosca templates https://review.openstack.org/299657 | 00:03 |
bobh_ | sridhar_ram: ok - need to get my test env back online so I can try it | 00:04 |
sridhar_ram | bobh_: looking at the traceback and the related code.. this is something might take sometime to fix.. please confirm | 00:04 |
sridhar_ram | bobh_: sounds good.. another minor fix based on test, review when you get a chance .. https://review.openstack.org/299676 | 00:08 |
sridhar_ram | this is the one from Sripriya | 00:08 |
bobh_ | sridhar_ram: will do | 00:09 |
*** uck has quit IRC | 00:11 | |
*** sridhar_ram has quit IRC | 00:18 | |
*** lhcheng has quit IRC | 00:25 | |
*** sridhar_ram has joined #tacker | 00:33 | |
*** vishnoianil has quit IRC | 00:37 | |
*** vishnoianil has joined #tacker | 00:39 | |
openstackgerrit | Merged openstack/tacker: Revert multiple mgmtdriver check for old templates https://review.openstack.org/299676 | 00:40 |
*** sripriya has quit IRC | 00:46 | |
*** vishnoianil has quit IRC | 00:54 | |
sridhar_ram | bobh_: quick check.. if the issue seems localized and a possible simple fix exists. Otherwise we will take it up in a follow on maintenance release | 00:57 |
bobh_ | sridhar_ram: ok - just got my devstack running and the templates loaded | 00:58 |
bobh_ | sridhar_ram: need a little while to debug | 00:58 |
sridhar_ram | bobh_: sounds good.. will stand by. | 00:58 |
*** sripriya has joined #tacker | 01:15 | |
*** sripriya_ has joined #tacker | 01:17 | |
sridhar_ram | bobh_: I'm heading home.. will check up as soon as I can. | 01:18 |
sridhar_ram | *check back | 01:18 |
*** lhcheng has joined #tacker | 01:19 | |
*** sripriya has quit IRC | 01:20 | |
*** sridhar_ram has quit IRC | 01:25 | |
*** lhcheng_ has joined #tacker | 01:27 | |
*** lhcheng has quit IRC | 01:29 | |
*** vishwanathj has joined #tacker | 01:30 | |
*** sridhar_ram_away has joined #tacker | 01:31 | |
*** sridhar_ram_away is now known as sridhar_ram_ | 01:32 | |
bobh_ | sridhar_ram_: I think I found the problem - working on the fix | 01:41 |
sridhar_ram_ | bobh_: cool | 01:42 |
*** s3wong has quit IRC | 01:44 | |
*** prashantD_ has quit IRC | 01:48 | |
bobh_ | sridhar_ram_: so this ends up being a rabbit hole.... | 02:03 |
sridhar_ram_ | bobh_: ah, really ? | 02:10 |
sridhar_ram_ | bobh_: we need high degree of comfort to absorb it at this stage | 02:11 |
sridhar_ram_ | bobh_: perhaps we can release-note it for now and fix it in 0.3.1 release | 02:12 |
*** dkushwaha has joined #tacker | 02:13 | |
*** dkushwaha has quit IRC | 02:17 | |
vishwanathj | bobh_ curious to know the root cuase | 02:18 |
*** tbh has joined #tacker | 02:19 | |
vishwanathj | bobh_ if you are looking for a quick fix, consider changing the https://github.com/openstack/tacker/blob/master/tacker/vm/tosca/lib/tacker_nfv_defs.yaml#L29 to "type: map" | 02:20 |
bobh_ | vishwanathj: thanks - that fixed the rabbit hole - I think | 02:35 |
vishwanathj | bobh_ are you planning to push a patch set? | 02:35 |
bobh_ | vishwanathj: hoping to - testing now | 02:36 |
vishwanathj | bobh_ sridhar_ram as part of your patchset upload, can you push another simple change in heat.py https://github.com/openstack/tacker/blob/master/tacker/vm/infra_drivers/heat/heat.py#L104 ..... should be "if not len(device_template_dict['name']) and 'metadata' in vnfd_dict:"......the issue is blocking if you have more than one TOSCA template uploaded as there is no way to select the VNFD at VNF deploy time..... | 02:38 |
vishwanathj | the name for the template shows up as blank currently | 02:40 |
bobh_ | vishwanathj: hat if name isn't in the dict? won't we get a key error? | 02:42 |
*** lhcheng has joined #tacker | 02:42 | |
bobh_ | vishwanathj: if ('name' not in device_template_dict or not len(device_template_dict['name'])) and 'metadata' in vnfd_dict: | 02:43 |
vishwanathj | the problem is name is always present at that point in the code with an empty value..... | 02:43 |
*** Janki has joined #tacker | 02:43 | |
bobh_ | ok | 02:43 |
vishwanathj | its never the case that name is not present as a key | 02:43 |
bobh_ | I'll put it in my patchset | 02:43 |
vishwanathj | I tested it this afternoon | 02:43 |
*** Janki has quit IRC | 02:44 | |
*** lhcheng_ has quit IRC | 02:45 | |
openstackgerrit | Sripriya Seetharam proposed openstack/tacker: Deprecate infra driver nova for Mitaka https://review.openstack.org/299724 | 02:49 |
*** dkushwaha has joined #tacker | 03:00 | |
* sridhar_ram_ wandering back to irc | 03:05 | |
sridhar_ram_ | bobh_: vishwanathj: looks you both found a way out of the rabbit hole. Is a patchset on the way ? | 03:07 |
bobh_ | sridhar_ram_: running tox now | 03:07 |
sridhar_ram_ | bobh_: cool | 03:08 |
sridhar_ram_ | sripriya_: ^^ | 03:08 |
sripriya_ | sridhar_ram: seems like a quick fix for now? | 03:08 |
bobh_ | sripriya_: it's a combination of a real fix and a quickfix | 03:09 |
bobh_ | sripriya_: some existing technical debt that needs to be cleaned up when the existing yaml format is removed | 03:09 |
sripriya_ | bobh_: i see, ok | 03:09 |
bobh_ | sripriya_: and some enhancements to the format when we figure out why tosca-parser is complaining | 03:10 |
sripriya_ | bobh_: ack | 03:10 |
dkushwaha | sridhar_ram_, hi. Please review the patch-set https://review.openstack.org/#/c/291032/ | 03:37 |
*** lhcheng has quit IRC | 03:40 | |
*** uck has joined #tacker | 03:51 | |
*** uck has quit IRC | 03:52 | |
openstackgerrit | Bob Haddleton proposed openstack/tacker: Fix monitoring of TOSCA VNFDs https://review.openstack.org/299737 | 03:54 |
*** bobh_ has quit IRC | 03:55 | |
*** janki has joined #tacker | 03:56 | |
*** dkushwaha_ has joined #tacker | 04:01 | |
*** dkushwaha_ has quit IRC | 04:06 | |
*** sridhar_ram has joined #tacker | 04:25 | |
*** sridhar_ram1 has joined #tacker | 04:28 | |
*** sridhar_ram has quit IRC | 04:30 | |
openstackgerrit | Nguyen Hung Phuong proposed openstack/tacker: Fix typos in Tacker files https://review.openstack.org/299749 | 04:38 |
sridhar_ram1 | tbh: ping | 04:42 |
*** tbh has quit IRC | 04:48 | |
*** vishwanathj has quit IRC | 04:53 | |
*** janki has quit IRC | 05:02 | |
*** dkushwaha has quit IRC | 05:05 | |
*** dkushwaha has joined #tacker | 05:06 | |
*** Shahid_ has joined #tacker | 05:25 | |
*** sridhar_ram1 has quit IRC | 05:26 | |
*** Shahid_ has quit IRC | 06:00 | |
*** sridhar_ram_ has quit IRC | 06:04 | |
dkushwaha | sripriya_, hi | 06:09 |
*** s3wong has joined #tacker | 06:14 | |
*** gongysh has quit IRC | 07:06 | |
*** zeih has joined #tacker | 07:13 | |
*** gongysh has joined #tacker | 07:23 | |
*** mbound has joined #tacker | 07:59 | |
*** Shahid_ has joined #tacker | 08:19 | |
dkushwaha | gongysh, hi | 08:26 |
*** mbound has quit IRC | 08:38 | |
*** zeih has quit IRC | 08:51 | |
*** tbh has joined #tacker | 09:11 | |
*** zeih has joined #tacker | 09:16 | |
*** tbh has quit IRC | 09:50 | |
*** tbh has joined #tacker | 09:51 | |
*** Shahid_ has quit IRC | 09:56 | |
*** Shahid_ has joined #tacker | 10:08 | |
*** zeih has quit IRC | 10:19 | |
*** zeih has joined #tacker | 10:22 | |
*** tbh has quit IRC | 10:30 | |
*** zeih has quit IRC | 10:33 | |
*** Shahid_ has quit IRC | 10:42 | |
*** mbound has joined #tacker | 10:50 | |
*** mbound has quit IRC | 11:13 | |
*** mbound has joined #tacker | 11:13 | |
*** mbound has quit IRC | 11:15 | |
*** dhellmann has quit IRC | 11:22 | |
*** dhellmann_ has joined #tacker | 11:22 | |
*** Shahid_ has joined #tacker | 11:25 | |
*** dhellmann_ is now known as dhellmann | 11:26 | |
openstackgerrit | dharmendra kushwaha proposed openstack/tacker: Use python package instead of sshpass to execute remote commands. https://review.openstack.org/269424 | 11:44 |
*** zeih has joined #tacker | 12:29 | |
*** mbound has joined #tacker | 12:48 | |
*** mbound has quit IRC | 12:49 | |
*** mbound has joined #tacker | 12:49 | |
*** mbound has quit IRC | 12:51 | |
*** bobh_ has joined #tacker | 13:21 | |
*** zeih has quit IRC | 13:26 | |
*** uck has joined #tacker | 14:17 | |
*** dkushwaha_ has joined #tacker | 14:20 | |
*** gongysh has quit IRC | 14:25 | |
*** Janki has joined #tacker | 14:36 | |
*** Janki has quit IRC | 14:37 | |
*** gongysh has joined #tacker | 14:42 | |
*** dkushwaha_ has quit IRC | 14:52 | |
*** gongysh has quit IRC | 14:56 | |
*** Shahid_ has quit IRC | 15:16 | |
*** zeih has joined #tacker | 15:25 | |
*** Janki has joined #tacker | 15:31 | |
*** Janki has quit IRC | 15:32 | |
*** uck has quit IRC | 15:39 | |
*** dandruta has joined #tacker | 15:46 | |
*** sridhar_ram has joined #tacker | 16:10 | |
*** mbound has joined #tacker | 16:31 | |
*** mbound has quit IRC | 16:33 | |
*** prashantD has joined #tacker | 16:39 | |
*** vishnoianil has joined #tacker | 16:52 | |
*** s3wong has joined #tacker | 16:58 | |
*** mbound has joined #tacker | 17:19 | |
*** dkushwaha_ has joined #tacker | 17:27 | |
sripriya_ | dkushwaha_: hello | 17:28 |
*** zeih has quit IRC | 17:29 | |
*** sridhar_ram is now known as sridhar_ram_afk | 17:37 | |
*** lhcheng has joined #tacker | 17:44 | |
*** prashantD has quit IRC | 17:45 | |
*** dandruta has quit IRC | 17:55 | |
*** sridhar_ram_afk is now known as sridhar_ram | 17:55 | |
*** prashantD has joined #tacker | 17:59 | |
dkushwaha_ | sripriya_, hello | 18:00 |
*** lhcheng has quit IRC | 18:01 | |
*** lhcheng has joined #tacker | 18:03 | |
*** lhcheng has quit IRC | 18:03 | |
*** lhcheng has joined #tacker | 18:04 | |
*** lhcheng has quit IRC | 18:07 | |
sripriya_ | dkushwaha_: i just responded to your question on testing openwrt for the python pkg patch | 18:08 |
sripriya_ | dkuhshwaha_: let me know if you need any help | 18:09 |
dkushwaha_ | sripriya_, Thanks for your consideration. I just test regarding the same, but getting some different issue | 18:13 |
dkushwaha_ | sripriya_, I am looking into that | 18:13 |
sripriya_ | dkushwaha_: cool | 18:13 |
*** openstackgerrit has quit IRC | 19:03 | |
*** openstackgerrit has joined #tacker | 19:04 | |
*** lhcheng has joined #tacker | 19:13 | |
*** sridhar_ram1 has joined #tacker | 20:14 | |
dkushwaha_ | sripriya_, hi | 20:14 |
*** sridhar_ram has quit IRC | 20:15 | |
sripriya_ | dkushwaha_: hello | 20:19 |
dkushwaha_ | sripriya_, in https://review.openstack.org/#/c/269424/11/tacker/vm/mgmt_drivers/openwrt/openwrt.py . While executing remote command if an error occurred, do we need to raise an exception(i.e. MgmtDriverException)? | 20:19 |
sripriya_ | dkushwaha_: correct | 20:20 |
dkushwaha_ | sripriya_, My doubt is, in case of sshpass, it does not generate any exception. So is't good if we also maintain the same behavior? Please suggest. | 20:24 |
dkushwaha_ | sripriya_, in that case we just log error message. | 20:24 |
sripriya_ | dkushwaha_: can we check the stderr_ and log a warning message? | 20:25 |
dkushwaha_ | sripriya_, yes. | 20:26 |
sripriya_ | dkushwaha_: we do not want to fail the vnf create if config failed, | 20:26 |
dkushwaha_ | sripriya_, let me explain the scenario... | 20:27 |
*** sridhar_ram1 has quit IRC | 20:29 | |
sripriya_ | dkushwaha_: i see that you raise an exception if it fails, this will exit the vnf create as a failed operation | 20:29 |
dkushwaha_ | sripriya_, in vnf-create when openwrt goes up, cmd_executer goes to fire command, but at that time it seems network services not started properly, so cmd_executer failed with error EHOSTUNREACH. And in that case currently we are generating MgmtDriverException. Due to which vnf create failed. | 20:35 |
sripriya_ | dkushwaha_: did you try this with the old template or the new tosca template? | 20:36 |
dkushwaha_ | sripriya_, with devstack/samples/config-openwrt-with-firewall.yaml | 20:37 |
sripriya_ | dkushwaha_: and vnfd template with openwrt.yaml? | 20:37 |
dkushwaha_ | sripriya_, yes | 20:38 |
sripriya_ | dkushwaha_: in your case, looks like openwrt came up after a really long boot time which exceeded the default boot wait time of 30 seconds | 20:41 |
sripriya_ | dkushwaha_: IMO, you should attempt to connect to instance until a timeout before failing the command | 20:41 |
sripriya_ | dkushwaha_: see https://github.com/openstack/tacker/blob/0fdd5a471759491d5114cc77125fd5aea4159a62/tacker/vm/plugin.py#L50 | 20:43 |
sripriya_ | dkushwaha_: my suggestion would be attempt to config for a certain timeout if EHOSTUNREACH is returned before giving up | 20:44 |
dkushwaha_ | sripriya_, ok, fine. | 20:45 |
sripriya_ | dkushwaha_: also lets not fail vnf-create if config failed | 20:45 |
sripriya_ | dkushwaha_: however for vnf-update it is all about config update, we will have to handle that scenario differently | 20:46 |
dkushwaha_ | sripriya_, right. | 20:46 |
sripriya_ | dkushwaha_: once mgmt driver raises exception it will be good to catch the exception in plugin file and take action appropriately | 20:48 |
dkushwaha_ | sripriya_, yes sure, it will be a good approach. | 20:49 |
sripriya_ | dkushwaha_: i understand sridhar_ram had put this bug under release must fix, but it is not under must fix now and will be going to master and not for mitaka | 20:55 |
sripriya_ | dkushwaha_: so you can take your time :-) | 20:55 |
dkushwaha_ | sripriya_, :-) | 20:57 |
*** dkushwaha_ has quit IRC | 21:48 | |
*** santoshk has joined #tacker | 21:57 | |
*** amotoki has joined #tacker | 22:01 | |
*** bobh_ has quit IRC | 22:02 | |
*** sridhar_ram has joined #tacker | 22:27 | |
openstackgerrit | Santosh Kodicherla proposed openstack/tacker: Adding vnf testcase with monitoring with new tosca template https://review.openstack.org/300248 | 23:13 |
*** amotoki has quit IRC | 23:13 | |
*** bobh_ has joined #tacker | 23:15 | |
*** bobh_ has quit IRC | 23:17 | |
*** santoshk has quit IRC | 23:19 | |
*** sridhar_ram has quit IRC | 23:34 | |
*** bobh_ has joined #tacker | 23:38 | |
*** bobh_ has quit IRC | 23:38 | |
*** lhcheng has quit IRC | 23:58 | |
*** s3wong has quit IRC | 23:58 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!