| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2018-11-13 | |||
| 15:39:42 | logan- | mriedem: yep! | |
| 16:17:20 | spatel | sean-k-mooney: morning | |
| 16:17:28 | spatel | are you in summit? | |
| 16:17:50 | spatel | if no then i need 2 min from you.. | |
| 16:18:03 | sean-k-mooney | spatel: no im not im afraid | |
| 16:18:12 | spatel | :) | |
| 16:18:32 | sean-k-mooney | how can i help | |
| 16:18:40 | spatel | i am testing influx + grafana for openstack dashboard... | |
| 16:19:16 | spatel | currently i am using telegraf and it looks amazing | |
| 16:19:27 | spatel | now i am confused what i should use collectd or telegraf ? | |
| 16:19:39 | spatel | need you honest opinion | |
| 16:19:43 | sean-k-mooney | ah ok thats a little out of my are of expertiese but i know its a solution many use. | |
| 16:20:15 | sean-k-mooney | well telegraf is much newer, collectd has more support but it is also a legacy solution | |
| 16:20:19 | spatel | This is first time i am going to use so want to make sure i won't hit the wall moving forward | |
| 16:20:29 | sean-k-mooney | it really depends on what you intend to use it for | |
| 16:21:06 | spatel | people saying telegraf has tags feature which will be very handy which collected doesn't support yet | |
| 16:21:47 | spatel | also telegraf doesn't have libvird support | |
| 16:22:10 | spatel | libvirtd* | |
| 16:22:30 | sean-k-mooney | spatel: well the 2 are not mutally exclcive | |
| 16:22:51 | sean-k-mooney | ralonsoh: you have worked on yardstick can you provide advice to spatel | |
| 16:22:52 | spatel | otherwise I think i will give it a try both and see... | |
| 16:23:12 | openstackgerrit | Dirk Mueller proposed openstack/nova master: Handle alternative UEFI boot loader locations on SLES https://review.openstack.org/348394 | |
| 16:23:20 | sean-k-mooney | i would proably default to collectd jsut becase im slightly more familure with it | |
| 16:23:34 | spatel | make sense! | |
| 16:23:55 | spatel | also telegraf use go lang which is new for me too :( | |
| 16:24:08 | spatel | i can handle python easily | |
| 16:24:15 | sean-k-mooney | mnaser: what solution does vexhost user for its monitoring? is it internal or do you use collectd, telegraf, nagios or zabex | |
| 16:24:32 | sean-k-mooney | spatel: collectd is all or mostly c | |
| 16:24:39 | spatel | mnaser: is in Openstack summit :) i was him on youtube.. | |
| 16:25:13 | spatel | but collected can use plugin written in pythn | |
| 16:25:20 | sean-k-mooney | spatel: yes he is. there will be a lot of people at teh summit that can provide input on this topic. | |
| 16:25:45 | sean-k-mooney | spatel: collectd can use plugins written in python java and several other languges | |
| 16:26:04 | spatel | lets me try both and see who is handy.. | |
| 16:26:16 | spatel | worth exploring them | |
| 16:26:25 | sean-k-mooney | spatel: one of the advantages of it being around and used for so long is there are lots of language bindings | |
| 16:26:48 | spatel | hmm | |
| 16:26:52 | spatel | Point | |
| 16:27:16 | spatel | collectd is little fast because its written in C | |
| 16:28:42 | sean-k-mooney | spatel: well i hope that helped a little. enjoy the summit and try not to let two many vendors sell you things you dont need. | |
| 16:28:42 | spatel | I have 165 compute node in my cloud :) running 207 VMs | |
| 16:28:56 | ralonsoh | spatel, in https://github.com/opnfv/yardstick/blob/master/yardstick/dispatcher/influxdb.py you have an example of how to store the information in influxdb | |
| 16:29:00 | sean-k-mooney | hehe nice | |
| 16:29:05 | spatel | sean-k-mooney: i am in office :( not lucky! | |
| 16:29:30 | ralonsoh | spatel, you also need to retrieve this information from collectd. You can use rabbitmq to retrieve collectd information | |
| 16:29:34 | openstackgerrit | Dirk Mueller proposed openstack/nova master: Handle alternative UEFI boot loader locations on SLES https://review.openstack.org/348394 | |
| 16:29:38 | spatel | thanks ralonsoh! reading... | |
| 16:30:33 | spatel | You can use rabbitmq to retrieve collectd information - this is interesting.. | |
| 16:30:41 | sean-k-mooney | ralonsoh: thanks i rely on you and emma when it comes to monitoring to fill it the gaps as its not someting i normally dable in. | |
| 16:31:06 | ralonsoh | spatel, here you have an example of how to handle the collectd info: https://github.com/opnfv/yardstick/blob/master/yardstick/network_services/nfvi/resource.py | |
| 16:31:29 | ralonsoh | sean-k-mooney, well, I'm not contributing any more to Yardstick (for now) | |
| 16:32:02 | sean-k-mooney | ralonsoh: true but you have much more uptoday knowladge of where things stood then i. | |
| 16:33:10 | ralonsoh | spatel: collectd dumps the information in rabbitmq. How to create a mq consumer: https://github.com/opnfv/yardstick/blob/master/yardstick/network_services/nfvi/resource.py#L111 | |
| 16:33:40 | sean-k-mooney | ralonsoh: collectd can also use the netwroking pluging to dump the data directly into infuxdb too | |
| 16:33:51 | spatel | you are saying instead of sending info directly to influx use rabbitmq for transport? | |
| 16:33:55 | sean-k-mooney | so you dont need to have rabbit if you dont need it | |
| 16:34:16 | ralonsoh | spatel, you can dump the information to influxdb too | |
| 16:34:50 | spatel | yes! trying to understand what will be the advantage using rabbitmq | |
| 16:35:53 | sean-k-mooney | spatel: there are pros and cons. you are trading scaleableity for grater message que load | |
| 16:36:09 | spatel | hmm! | |
| 16:36:28 | sean-k-mooney | if you go the rabbit route i would personally have a different rabbitmq server to not impact your openstack performance | |
| 16:36:41 | sean-k-mooney | but i would start with going direct to infuxdb | |
| 16:37:04 | spatel | ++ for direct to influxdb | |
| 16:37:17 | spatel | I am new so i don't want to make it extra complex.. :) | |
| 16:37:34 | spatel | i will go with rabbitmq when my cloud will be 1000 compute node | |
| 16:38:12 | spatel | Thanks!! folks it was very helpful.. | |
| 16:38:13 | sean-k-mooney | ralonsoh: do you know how mature the baromater installer is. it supports deploying collect+infux+graphan correct in containers | |
| 16:38:43 | sean-k-mooney | spatel: glad we could help. its always better to start simple :) | |
| 16:38:51 | spatel | yup! | |
| 16:39:07 | ralonsoh | sean-k-mooney, Emma was working actively in this project. I don't even know where can I download it! | |
| 16:40:47 | sean-k-mooney | its amazing how in big companiens 2 people can be sitting 3 meters from each other and be working on totoally different things. no worries | |
| 16:41:38 | sean-k-mooney | spatel: by the way you are using openstack ansible. correct its proably worth poping over to there irc channel and asking them if they have any supported monitoring solutions and what the recommend too | |
| 16:43:42 | spatel | I did ask and they do have some stuff but not very documented and don't know if folks using them or not https://github.com/openstack/openstack-ansible-ops/blob/master/cluster_metrics/playbook-influx-db.yml | |
| 16:44:24 | spatel | Let me poke around and will see how it goes... | |
| 17:55:54 | openstackgerrit | Merged openstack/nova master: Make _instances_cores_ram_count() be smart about cells https://review.openstack.org/569055 | |
| 21:38:40 | openstackgerrit | Merged openstack/nova master: Fix a help string in nova-manage https://review.openstack.org/616723 | |
| 21:51:06 | openstackgerrit | Dirk Mueller proposed openstack/nova master: Handle alternative UEFI boot loader locations on SLES https://review.openstack.org/348394 | |
| 22:07:25 | openstackgerrit | Merged openstack/nova master: Update compute API.get() mocks in test_server_metadata https://review.openstack.org/615341 | |
| 22:07:32 | openstackgerrit | Merged openstack/nova master: Update compute API.get() stubs in test_serversV21 https://review.openstack.org/615342 | |
| 22:07:40 | openstackgerrit | Merged openstack/nova master: Update compute API.get() stubs in test_server_actions https://review.openstack.org/615343 | |
| #openstack-nova - 2018-11-14 | |||
| 03:17:22 | openstackgerrit | yenai proposed openstack/nova master: Rollback detaching when reserving volume is failed https://review.openstack.org/617861 | |
| 06:58:11 | sean-k-mooney | the summit starts in 1 hour right? at least the keynote should? | |
| 06:58:58 | sean-k-mooney | or it could have stared yesterday... | |
| 07:17:14 | tetsuro | sean-k-mooney: The summit started yesterday. The keynote of Day 2 starts in 1 hour. | |
| 07:17:39 | sean-k-mooney | tetsuro: hehe yes i started watching the youtube vids | |
| 07:18:08 | sean-k-mooney | tetsuro: are you at the summit this time | |
| 07:18:24 | tetsuro | Yes I am, but I didn't see you. | |
| 07:18:43 | tetsuro | you aren't this time? | |
| 07:19:14 | sean-k-mooney | no i choose not to travel this time as i have done quite alot of travel, at least for me already this year | |
| 07:19:45 | tetsuro | I see. I hope I can see you next Denver :) | |
| 07:19:55 | tetsuro | Anyway, gotta leave the hotel | |
| 07:20:09 | sean-k-mooney | yes same, im hoping to be in denver. enjoy | |
| 07:20:21 | tetsuro | bye. | |
| 07:41:31 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Add DownCellFixture https://review.openstack.org/614810 | |
| 07:59:25 | openstackgerrit | Dan Smith proposed openstack/nova master: Add CellsV2 FAQ about API design decisions https://review.openstack.org/617898 | |
| 08:08:14 | mriedem | dansmith: this is pretty simple, something we talked about at the ptg to fix a few issues in scheduling to a specific node https://review.openstack.org/#/c/603585/ | |
| 08:16:59 | dansmith | mriedem: okay I +Wd the spec, but I wish it included more about the actual nova side that actually solves a problem instead of just a placement-side API that _could_ be used to solve a problem | |
| 08:18:48 | sean-k-mooney | dansmith: the usecase we discussed at the ptg was related to numa affinity right | |
| 08:19:29 | dansmith | sean-k-mooney: no | |
| 08:19:41 | dansmith | sean-k-mooney: boot or live migration with a target host | |
| 08:20:14 | sean-k-mooney | dansmith: the in_tree paramater? | |