Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2021-08-31
14:33:55 sean-k-mooney but this is lower on my list of priorities
14:35:35 gibi ack
14:52:56 opendevreview Merged openstack/nova master: Remove module level caching https://review.opendev.org/c/openstack/nova/+/806394
15:04:52 spatel is anyone running DPDK on Intel X710 nic, i would like to buy so looking for some good feedback ?
15:10:11 spatel currently i am using Intel® 82599 10 Gigabit Ethernet but it has limitation of rx/tx queue which is max=2 per VF
15:50:29 gibi fyi nova weekly meeting starts in 10 minutes here in the channel
15:51:34 sean-k-mooney spatel: x710 used ot be the primaly nic that intel used for dpdk development so its fully supported but its a few years old now and they have proably moveed tothe 800 seriese they lauched last year
15:52:10 sean-k-mooney spatel: this is the dpdk docs for that nic https://doc.dpdk.org/guides/nics/i40e.html
15:52:14 spatel do you know how many rx/tx queue does x710 support ?
15:52:32 spatel x710 is older so may be cheaper for me to get it up and running with dpdk
15:52:56 sean-k-mooney x710 is much much newer then 82599
15:53:10 sean-k-mooney but i know the queue count is actul configurable in the driver
15:53:22 sean-k-mooney im not sure what its set too by default for VFs
15:53:34 spatel 82599 has hard limit for 2 rx/tx queue per VF
15:53:47 sean-k-mooney ah Reserved number of Queues per VF (default 4)
15:53:57 spatel Look like x710 has 16 per VF (default 4) but we can change it so that is good
15:54:00 sean-k-mooney https://doc.dpdk.org/guides/nics/i40e.html#runtime-config-options
15:54:31 spatel nic look like x710 is good choice for me in low cost
15:55:20 sean-k-mooney the 72x revision are nic if you need 25G or think will upgrade at some point
15:55:41 sean-k-mooney they are not much more expensive and also are slightly more power effeinct
15:56:20 spatel 10G is more then enough, my whole requirement is to run DPDK and remove SRIOV dependency which doesn't support security-group and bonding
15:56:33 spatel hmm
15:57:29 spatel let me see what i can get because all depend on budget but good to know
15:57:48 spatel sean-k-mooney i have successfully deploy SRIOV with OVN and its working great
15:58:01 sean-k-mooney good to know
15:58:11 sean-k-mooney do not use it for PF passhtouhg however
15:58:21 sean-k-mooney we need to remove that at some point
15:58:39 spatel remove it from where? NOVA?
15:59:36 sean-k-mooney ovn
15:59:52 spatel I have question for you how does OVN provide redundancy for network node? currently my compute nodes are network node so lets say i have 10 compute nodes then how OVN schedule gateway chassis?
15:59:52 sean-k-mooney it allows vnic_type=direct-physical but does not manage it properly
16:00:06 bauzas 8 seconds before the meeting ;)
16:00:13 bauzas so, wait a bit for the answer :p
16:00:20 spatel :)
16:00:21 sean-k-mooney that proably better asked in neutron too
16:00:34 spatel thank you!
16:00:34 bauzas gibi: you're late
16:01:25 opendevmeet The meeting name has been set to 'nova'
16:01:25 opendevmeet Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:01:25 opendevmeet Meeting started Tue Aug 31 16:01:25 2021 UTC and is due to finish in 60 minutes. The chair is gibi. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:01:25 gibi #startmeeting nova
16:01:53 gibi I was just reading the scrollback about the nics
16:01:55 gibi :)
16:01:56 gibi o/
16:01:59 gmann o/
16:02:05 elodilles o/
16:02:15 dansmith o/
16:02:45 gibi #topic Bugs (stuck/critical)
16:02:55 gibi we have one bug marked critical
16:02:59 gibi https://bugs.launchpad.net/nova/+bug/1940555 around SQLAlchemy URL handling and the fix is going through the gate: https://review.opendev.org/c/openstack/nova/+/805663
16:03:15 bauzas \o
16:03:24 gibi it is bounced since I wrote the agenda but I requeued it
16:03:42 sean-k-mooney ack
16:03:53 gibi #link 17 new untriaged bugs (+6 since the last meeting): #link https://bugs.launchpad.net/nova/+bugs?search=Search&field.status=New
16:04:09 gibi I had no time to check the untriaged bugs in the last couple of days
16:04:22 gibi we have 0 bugs marked with xena-rc-potential tag #link https://bugs.launchpad.net/nova/+bugs?field.tag=xena-rc-potential
16:04:27 gibi please start marking release critical bugs with xena-rc-potential tag
16:04:53 bauzas tbc, those need to be regression bugs
16:04:59 gibi yepp
16:05:12 bauzas ie. something that worked before Xena but now has issues with this release
16:05:37 bauzas we only have 2 weeks before RC1 in order to merge bugfixes that *aren't* regressions
16:05:49 bauzas or those would be fixed in the next Yoga release
16:05:59 gibi and backported :)
16:06:09 sean-k-mooney i dont think we currenlty have any critical bugs that would need the rc flag
16:06:13 bauzas so, pings are appreciated besides classic feature review requests
16:06:30 bauzas we have an interesting race condition issue with placement
16:06:41 gibi yepp I saved that for the gate issue topic
16:06:51 gibi lets go there
16:06:52 bauzas this isn't only impacting the gate, right?
16:07:20 gibi it is a new race so so far we only have infor from the gate about it
16:07:23 bauzas but yeah, let's discuss this bug on the gate section
16:07:28 gibi #topic Gate status
16:07:32 gibi Nova gate bugs #link https://bugs.launchpad.net/nova/+bugs?field.tag=gate-failure
16:07:41 gibi We had the allocation deletion conflict bug #link https://bugs.launchpad.net/nova/+bug/1836754 that hopefully resolved when we landed #link https://review.opendev.org/c/openstack/nova/+/688802 yesterday
16:08:05 gibi We have a project creation race in allocation update bug as well #link https://storyboard.openstack.org/#!/story/2009159 that is affecting the those gate jobs that are booting more than on VM at the same time. Octavia is heavily affected. I've pushed a reproducer #link https://review.opendev.org/c/openstack/placement/+/806730 but I don't know how to fix the actual bug.
16:08:40 gibi this is the one bauzas mentioned above ^^
16:09:30 bauzas if the race isn't occurring a lot of times, we should maybe just call the transaction again
16:09:37 gibi in short if two VM for the same project is scheduled at the same time then the PUT /allocations call could cause a race in the project creation in placement
16:10:16 gibi bauzas: that was my first idea, but looking the code I'm not sure about the amount of surgery needed for this
16:10:33 dansmith it must be the *first* two vms right?
16:10:34 gibi if somebody has deeper sqlAlchemy knowledge than me then help is appreciated
16:10:45 bauzas dansmith: it *should* be yeah
16:10:47 gibi dansmith: first two VMs for a project yes
16:10:55 sean-k-mooney gibi: sorry im a bit confusted by the term project createion in a placment context
16:10:59 dansmith that definitely seems uncool
16:11:15 bauzas I guess we could hit the problem often with multicreate
16:11:30 bauzas ie. server create --max 2
16:11:30 gibi sean-k-mooney: placement stores the project_id and user_id of each consumer so that usages can be aggregated per project
16:11:38 sean-k-mooney why does placement need to be project wawre beyond storigng the project/user id in the allocation?
16:11:54 bauzas sean-k-mooney: because we use it from an API query
16:12:01 dansmith sean-k-mooney: it's relational so it has a table for it I think
16:12:07 sean-k-mooney yes im aware it there for usage
16:12:11 bauzas you can filter per project
16:12:17 gibi https://docs.openstack.org/api-ref/placement/?expanded=list-usages-detail#list-usages
16:12:19 sean-k-mooney but im not sure what resouce in the db we would need to create that would race
16:12:21 bauzas this was my question earlier this EU morning time :)
16:12:34 gibi sean-k-mooney: it is the row in the projects table
16:12:35 dansmith sean-k-mooney: the first time that entry goes into the table for the project?
16:12:43 gibi yes ^^
16:12:48 bauzas correct

Earlier   Later