August 2022 - openstack-discuss - lists.openstack.org

Re: ovn-bgp-agent installation issue
by Satish Patel 23 Aug '22

23 Aug '22

Hi Luis, /cc - openstack discuss mailing list Thank you so much for clearing my doubts. I have a counter question. BGP Mode: - In bgp mode where i should run ovn-bgp-agent ? Network node or Compute node? EVPN Mode: - In evpn mode where i should run ovn-bgp-agent? - In my lab i am using your heck "How to use it without BGPVPN" to create vni mapping, do you think because of that i am not seeing vni pushed out to FRR automatically? https://ltomasbo.wordpress.com/2021/06/25/openstack-networking-with-evpn/ - In your demo link I can see you have bridge two different deployment clouds using the same VNI, In this scenario we need a router (or vrf) somewhere correct to bridge to different subnet IPs, correct? or is this L2 stretch? I am going to open a new thread to discuss issues related to expose_tenant_networks=True flag issue. On Tue, Aug 23, 2022 at 3:14 AM Luis Tomas Bolivar <ltomasbo(a)redhat.com> wrote: > Hi Satish! See inline > > On Mon, Aug 22, 2022 at 5:22 PM Satish Patel <satish.txt(a)gmail.com> wrote: > >> Hi Luis, >> >> Welcome back from your vacation, sorry i didn't know you are on vacation. >> Hope you had a wonderful time. >> >> How do i start conversion on opendev thread? is there a mailing list or >> are you saying open thread here - >> https://storyboard.openstack.org/#!/project/x/ovn-bgp-agent >> > > Actually, you did it right (sending it to openstack-discuss), but it seems > at some point we stop adding it on our replies... probably my fault due to > replying on the phone while on vacation. > > For the next issue we can open a different thread and I'll try not to drop > the openstack-discuss list! xD > > It would be great though if you can sent a follow up to the previous > thread with the final outcome (fixed by updating, or adding config X, ..), > so that community can see the project is alive, and other people facing > similar issues can try your solution/config. > >> I have an update for you, after upgrading the OVN/OVS version to the >> latest and that resolved lots of issues. As you mentioned earlier running >> the latest code is very important to get proper functionality. >> > > Awesome! Yeah, it is a new project, so functionality/fixes are being added > regularly, therefore using the latest version is the best approach. The > plan is to soon create a more stable version/release, once we have enough > testing coverage and main functionality covered > >> >> You are saying in BGP mode only FIP will get exposed but not the tenant >> VM ( What is this flag for expose_tenant_network=True ?) >> > > Main idea was: > - BGP mode for exposing VMs/LBs on provider networks, or with FIPs attached > - EVPN mode for tenant networks > > To give you some more context, we started with the BGP mode, and did the > expose_tenant_network to expose also tenant IPs. However, the BGP mode is > lacking an API to decide what to expose. So we moved to the EVPN mode (with > networking-bgpvpn as the API) to expose the tenant networks. > > That said, you are right, and the expose_tenant_network flag is intended > for the BGP mode, to expose all tenant networks (so you need to ensure no > overlapping CIDRs). And, if that is not working (it may be, as there is no > testing coverage for it), it is something we can definitely look at and > fix. So, feel free to open a new thread on openstack-discuss, or bug in > storyboard (whatever works for you better), and I'll try to fix asap. > > >> >> You are saying in EVPN mode only tenant VM ip will get exposed but not >> FIP but in this design what is the use of getting tenant VM exposed but you >> can't expose FIP then how does external people access VMs? ( I am totally >> confused in EVPN design and use case because your tenant VM is already >> talking to each other using geneve tunnel so why do we need EVPN/L2VPN to >> expose them?) >> > > EVPN mode is indeed only for tenant VMs (and loadbalancer). Idea for this > mode is to connect (N/S traffic, as you mention E/W is still using the > normal geneve path) tenant networks between different OpenStack clusters. > EVPN will create a vxlan tunnel connecting them. So, the vni/vxlan id > selected (with networking-bgpvpn) is the vxlan tunnel encap id being used > to connect both tenant networks. > > So, the idea behind the EVPN mode is not to make your VMs in a tenant > network publicly accessible, but being able to access them from a different > tenant network in a different cloud (it actually does not need to be an > OpenStack cloud, anything connected to the same vxlan id. Perhaps this is a > bit more clear in this demo: > https://ltomasbo.wordpress.com/2021/10/01/ovn-bgp-agent-interconnecting-kub… > > >> >> This is current status of my LAB >> >> ovn-bgp-agent in BGP Mode: >> I am successfully able to expose FIP and provider IPs (as you mentioned) >> but when i use expose_tenant_network=True in config then getting error and >> vm tenant ips not getting exposed to bgp so i am not sure what is the use >> case of that flag. >> > > Yep, most probably we broke something here due to lack of upstream testing > for this flag. I'll check asap and try to fix it > >> >> ovn-bgp-agent in EVPN Mode: >> In this mode everything is working and tenant vm also gets exposed but >> when I attach FIP to vm those FIP ip are not getting exposed in BGP (as you >> mentioned in your reply). >> > Yes, EVPN mode is not intended for FIPs or VMs on provider networks. Only > for tenants > > >> But i am seeing one bug here where vni config not getting inserted in FRR >> https://opendev.org/x/ovn-bgp-agent/src/branch/master/ovn_bgp_agent/drivers… >> >> >> Who will trigger that code and when will that code get triggered to >> configure FRR for vrf/vni ? >> > > This is being added by networking-bgpvpn. You can see some details about > the integration into my blogpost ( > https://ltomasbo.wordpress.com/2021/06/25/openstack-networking-with-evpn/) > or in the upstream documentation: > https://opendev.org/x/ovn-bgp-agent/src/branch/master/doc/source/contributo… > > Note though, to make networking-bgpvpn to work with this integration, this > patch is needed (which btw I should include into the documentation): > https://review.opendev.org/c/openstack/networking-bgpvpn/+/803161 > > Cheers, > Luis > > >> >> >> >> >> >> On Mon, Aug 22, 2022 at 5:44 AM Luis Tomas Bolivar <ltomasbo(a)redhat.com> >> wrote: >> >>> Hi Satish, >>> >>> Sorry I was on vacation. Trying to get through my inbox. >>> >>> I'm not sure what is the current status, note there is a difference >>> between EVPN and BGP mode. BGP mode is for FIPs and VMs/LBs on the provider >>> network or with FIPs, while EVPN is for VMs on the tenant networks, without >>> FIPs. Right now you cannot mix them and need to choose either EVPN or BGP >>> mode. I'm planning to give it a try to make it multidriver, but I haven't >>> even started yet. >>> >>> BTW, as you did with the initial email, perhaps it is worth to have >>> conversations on the opendev thread, so that anyone else facing the same >>> issues can get some hints (or even provide feedback), that was one of the >>> main ideas about moving it in there. To be able to have better support. >>> >>> As for the questions: >>> # # on rack-2-host-1 >>> # when i created vm1 which endup on rack-2-host-1 but it doesn't expose >>> the vm ip address. Is that normal behavior? >>> >>> This should be exposed in the node with cr-lrp port, i.e., rack1-host2, >>> and it should be as simple as adding the IP to the lo-2001 dummy device >>> >>> # When I attach a floating ip to vm2 then why does my floating ip >>> address not get exposed in BGP? >>> This is what I mentioned before, you cannot merge BGP and EVPN mode. >>> >>> On Mon, Aug 8, 2022 at 1:41 PM Satish Patel <satish.txt(a)gmail.com> >>> wrote: >>> >>>> Hi Luis, >>>> >>>> Sorry for bugging. I’m almost there now. Do you have any thought of >>>> following question >>>> >>>> Sent from my iPhone >>>> >>>> On Aug 5, 2022, at 1:16 AM, Satish Patel <satish.txt(a)gmail.com> wrote: >>>> >>>> >>>> Good morning Luis, >>>> >>>> Quick question, I have following deployment as per your lab >>>> >>>> rack-1-host-1 (controller) >>>> rack-1-host-2 (compute1 - This is hosting cr-lrp ports, inshort router) >>>> rack-2-host-1 (compute2) >>>> >>>> I have created two vms >>>> >>>> vagrant@rack-1-host-1:~$ nova list >>>> nova CLI is deprecated and will be a removed in a future release >>>> >>>> +--------------------------------------+------+--------+------------+-------------+--------------------------------------+ >>>> | ID | Name | Status | Task State | >>>> Power State | Networks | >>>> >>>> +--------------------------------------+------+--------+------------+-------------+--------------------------------------+ >>>> | aecb4f10-c46f-4551-b112-44e4dc007e88 | vm1 | ACTIVE | - | >>>> Running | private-test=10.0.0.105 | >>>> | ceae14b9-70c2-4dbc-8071-0d64d9a0ca84 | vm2 | ACTIVE | - | >>>> Running | private-test=10.0.0.86, 172.16.1.200 | >>>> >>>> +--------------------------------------+------+--------+------------+-------------+--------------------------------------+ >>>> >>>> # on rack-1-host-2 >>>> when i spun up vm2 which endup on rack-1-host-2 hence it created >>>> vrf-2001 on dummy lo-2001 interface and exposed vm2 ip address >>>> 10.0.0.86/32 >>>> >>>> 96: vrf-2001: <NOARP,MASTER,UP,LOWER_UP> mtu 65575 qdisc noqueue state >>>> UP group default qlen 1000 >>>> link/ether 22:cc:25:b3:7b:96 brd ff:ff:ff:ff:ff:ff >>>> 97: br-2001: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue >>>> master vrf-2001 state UP group default qlen 1000 >>>> link/ether 0a:c3:23:7a:8f:0c brd ff:ff:ff:ff:ff:ff >>>> inet6 fe80::851:67ff:fe64:b2c3/64 scope link >>>> valid_lft forever preferred_lft forever >>>> 98: vxlan-2001: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc >>>> noqueue master br-2001 state UNKNOWN group default qlen 1000 >>>> link/ether 0a:c3:23:7a:8f:0c brd ff:ff:ff:ff:ff:ff >>>> inet6 fe80::8c3:23ff:fe7a:8f0c/64 scope link >>>> valid_lft forever preferred_lft forever >>>> 99: lo-2001: <BROADCAST,NOARP,UP,LOWER_UP> mtu 1500 qdisc noqueue >>>> master vrf-2001 state UNKNOWN group default qlen 1000 >>>> link/ether d6:60:da:91:2e:6d brd ff:ff:ff:ff:ff:ff >>>> inet 10.0.0.86/32 scope global lo-2001 >>>> valid_lft forever preferred_lft forever >>>> inet6 fe80::d460:daff:fe91:2e6d/64 scope link >>>> valid_lft forever preferred_lft forever >>>> >>>> >>>> # on rack-2-host-1 >>>> when i created vm1 which endup on rack-2-host-1 but it doesn't expose >>>> the vm ip address. Is that normal behavior? >>>> >>>> When I attach a floating ip to vm2 then why does my floating ip address >>>> not get exposed in BGP? >>>> >>>> Thank you in advance >>>> >>>> >>>> On Thu, Aug 4, 2022 at 2:30 PM Satish Patel <satish.txt(a)gmail.com> >>>> wrote: >>>> >>>>> Update: Good news, I found what was wrong. >>>>> >>>>> After adding AS it works. In your doc you only added VNI but look like >>>>> it is required to add BGP AS. Or may be your doc is little older and new >>>>> code required AS. >>>>> >>>>> vagrant@rack-1-host-1:~$ sudo ovn-nbctl set logical_switch_port >>>>> c32dcd90-7820-44bd-894f-416e44b36aa0 external_ids:"neutron_bgpvpn\:as"=64999 >>>>> vagrant@rack-1-host-1:~$ sudo ovn-nbctl set logical_switch_port >>>>> f55c1d1e-4b5b-4d8c-b922-3ad4a9700c81 external_ids:"neutron_bgpvpn\:as"=64999 >>>>> >>>>> Now i can see it created VRF and exposed VM tenant ip in lo-2001 >>>>> >>>>> 73: vrf-2001: <NOARP,MASTER,UP,LOWER_UP> mtu 65575 qdisc noqueue state >>>>> UP group default qlen 1000 >>>>> link/ether b2:29:65:3b:ac:db brd ff:ff:ff:ff:ff:ff >>>>> 74: br-2001: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue >>>>> master vrf-2001 state UP group default qlen 1000 >>>>> link/ether d6:0d:61:2d:b7:29 brd ff:ff:ff:ff:ff:ff >>>>> inet6 fe80::6830:47ff:febc:b10b/64 scope link >>>>> valid_lft forever preferred_lft forever >>>>> 75: vxlan-2001: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc >>>>> noqueue master br-2001 state UNKNOWN group default qlen 1000 >>>>> link/ether d6:0d:61:2d:b7:29 brd ff:ff:ff:ff:ff:ff >>>>> inet6 fe80::d40d:61ff:fe2d:b729/64 scope link >>>>> valid_lft forever preferred_lft forever >>>>> 76: lo-2001: <BROADCAST,NOARP,UP,LOWER_UP> mtu 1500 qdisc noqueue >>>>> master vrf-2001 state UNKNOWN group default qlen 1000 >>>>> link/ether fe:a1:a0:87:76:7c brd ff:ff:ff:ff:ff:ff >>>>> inet 10.0.0.83/32 scope global lo-2001 >>>>> valid_lft forever preferred_lft forever >>>>> inet6 fe80::fca1:a0ff:fe87:767c/64 scope link >>>>> >>>>> >>>>> I am continuing doing testing and seeing if I hit any other bug.. >>>>> >>>>> On Thu, Aug 4, 2022 at 11:58 AM Satish Patel <satish.txt(a)gmail.com> >>>>> wrote: >>>>> >>>>>> Luis, >>>>>> >>>>>> I am following your doc, tell me if that doc is outdated or not >>>>>> https://ltomasbo.wordpress.com/2021/06/25/openstack-networking-with-evpn/ >>>>>> >>>>>> I have upgraded pyroute2 to 0.7.2 but still getting the same error, >>>>>> if you look at carefully following logs you will see an agent saying I >>>>>> can't find VNI but it's there, i can see in ovn-nbctl list >>>>>> logical_switch_port. >>>>>> >>>>>> pyroute2 0.7.2 >>>>>> pyroute2.core 0.6.13 >>>>>> pyroute2.ethtool 0.6.13 >>>>>> pyroute2.ipdb 0.6.13 >>>>>> pyroute2.ipset 0.6.13 >>>>>> pyroute2.ndb 0.6.13 >>>>>> pyroute2.nftables 0.6.13 >>>>>> pyroute2.nslink 0.6.13 >>>>>> >>>>>> >>>>>> ovn-bgp-agent logs >>>>>> >>>>>> 2022-08-04 15:52:52.898 396714 DEBUG >>>>>> ovn_bgp_agent.drivers.openstack.utils.ovn [-] Either "neutron_bgpvpn:vni" >>>>>> or "neutron_bgpvpn:as" were not found or have an invalid value in the port >>>>>> f55c1d1e-4b5b-4d8c-b922-3ad4a9700c81 external_ids {'neutron:cidrs': ' >>>>>> 172.16.1.132/24', 'neutron:device_id': >>>>>> '43b2c756-d92c-4fe5-b17b-32d5ab9b1f37', 'neutron:device_owner': >>>>>> 'network:router_gateway', 'neutron:network_name': >>>>>> 'neutron-0d82e6b0-bf9f-484d-85bd-0ba38aab508d', 'neutron:port_name': '', >>>>>> 'neutron:project_id': '', 'neutron:revision_number': '4', >>>>>> 'neutron:security_group_ids': '', 'neutron_bgpvpn:vni': '2001'} >>>>>> get_evpn_info >>>>>> /usr/local/lib/python3.8/dist-packages/ovn_bgp_agent/drivers/openstack/utils/ovn.py:251 >>>>>> >>>>>> 2022-08-04 15:52:52.899 396714 DEBUG >>>>>> ovn_bgp_agent.drivers.openstack.ovn_evpn_driver [-] No EVPN information for >>>>>> CR-LRP Port with IPs ['172.16.1.132/24']. Not exposing it. >>>>>> _expose_ip >>>>>> /usr/local/lib/python3.8/dist-packages/ovn_bgp_agent/drivers/openstack/ovn_evpn_driver.py:220 >>>>>> >>>>>> >>>>>> In OVN i can see vni number >>>>>> >>>>>> _uuid : 3473d0ce-1348-4423-a2dc-6d4df4c06f74 >>>>>> addresses : [router] >>>>>> dhcpv4_options : [] >>>>>> dhcpv6_options : [] >>>>>> dynamic_addresses : [] >>>>>> enabled : true >>>>>> external_ids : {"neutron:cidrs"="172.16.1.132/24", >>>>>> "neutron:device_id"="43b2c756-d92c-4fe5-b17b-32d5ab9b1f37", >>>>>> "neutron:device_owner"="network:router_gateway", >>>>>> "neutron:network_name"=neutron-0d82e6b0-bf9f-484d-85bd-0ba38aab508d, >>>>>> "neutron:port_name"="", "neutron:project_id"="", >>>>>> "neutron:revision_number"="4", "neutron:security_group_ids"="", >>>>>> "neutron_bgpvpn:vni"="2001"} >>>>>> ha_chassis_group : [] >>>>>> name : "f55c1d1e-4b5b-4d8c-b922-3ad4a9700c81" >>>>>> options : {exclude-lb-vips-from-garp="true", >>>>>> nat-addresses=router, router-port=lrp-f55c1d1e-4b5b-4d8c-b922-3ad4a9700c81} >>>>>> parent_name : [] >>>>>> port_security : [] >>>>>> tag : [] >>>>>> tag_request : [] >>>>>> type : router >>>>>> up : true >>>>>> >>>>>> >>>>>> On Thu, Aug 4, 2022 at 11:39 AM Satish Patel <satish.txt(a)gmail.com> >>>>>> wrote: >>>>>> >>>>>>> Hi Luis, >>>>>>> >>>>>>> This is what I have installed on the compute and controller nodes. >>>>>>> >>>>>>> pyroute2 0.6.13 >>>>>>> pyroute2.core 0.6.13 >>>>>>> pyroute2.ethtool 0.6.13 >>>>>>> pyroute2.ipdb 0.6.13 >>>>>>> pyroute2.ipset 0.6.13 >>>>>>> pyroute2.ndb 0.6.13 >>>>>>> pyroute2.nftables 0.6.13 >>>>>>> pyroute2.nslink 0.6.13 >>>>>>> >>>>>>> On Wed, Aug 3, 2022 at 10:10 AM Luis Tomas Bolivar < >>>>>>> ltomasbo(a)redhat.com> wrote: >>>>>>> >>>>>>>> What version of pyroute2 are you using? Maybe it is related to some >>>>>>>> bug in there in an old version (I hit quite a few). You can try to upgrade >>>>>>>> that. Also, as it is on the resync, you can also disable that by setting a >>>>>>>> very long time in there. >>>>>>>> >>>>>>>> >>>>>>>> On Wednesday, August 3, 2022, Satish Patel <satish.txt(a)gmail.com> >>>>>>>> wrote: >>>>>>>> >>>>>>>>> Any thoughts Luis. Thanks >>>>>>>>> >>>>>>>>> Sent from my iPhone >>>>>>>>> >>>>>>>>> On Jul 27, 2022, at 11:53 AM, Satish Patel <satish.txt(a)gmail.com> >>>>>>>>> wrote: >>>>>>>>> >>>>>>>>> >>>>>>>>> Luis, >>>>>>>>> >>>>>>>>> If you look at this logs - >>>>>>>>> https://paste.opendev.org/show/buRbY415guvHFUtSapFK/ >>>>>>>>> >>>>>>>>> I am able to expose the tenant ip when I set "expose_tenant_networks=True" >>>>>>>>> in the ovn-bgp-agent.conf file. But interestingly it exposes ip for the >>>>>>>>> first time but when i delete vm and re-create new vm then its throws >>>>>>>>> following error which you can see full track in above link >>>>>>>>> >>>>>>>>> ERROR oslo_service.periodic_task [-] Error during BGPAgent.sync: KeyError: 'object does not exists' >>>>>>>>> >>>>>>>>> >>>>>>>>> I have just 1x controller and 1x compute machine at present. As >>>>>>>>> you said something is broken when setting up "expose_tenant_networks=True" >>>>>>>>> Let me know if you need more info or logs etc. i will continue to poke and >>>>>>>>> see if anything else I can find. >>>>>>>>> >>>>>>>>> >>>>>>>>> On Wed, Jul 27, 2022 at 10:35 AM Luis Tomas Bolivar < >>>>>>>>> ltomasbo(a)redhat.com> wrote: >>>>>>>>> >>>>>>>>>> Hi Satish,sorry for the delay in replies, I'm on vacations and >>>>>>>>>> have limited connectivity. >>>>>>>>>> >>>>>>>>>> See soon comments/replies inline >>>>>>>>>> >>>>>>>>>> On Wed, Jul 27, 2022 at 4:58 AM Satish Patel < >>>>>>>>>> satish.txt(a)gmail.com> wrote: >>>>>>>>>> >>>>>>>>>>> Luis, >>>>>>>>>>> >>>>>>>>>>> Just checking incase you missed my last email. If you are on >>>>>>>>>>> vacation then ignore it :) Thank you. >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> On Sat, Jul 23, 2022 at 12:27 AM Satish Patel < >>>>>>>>>>> satish.txt(a)gmail.com> wrote: >>>>>>>>>>> >>>>>>>>>>>> Hi Luis, >>>>>>>>>>>> >>>>>>>>>>>> As you suggested, checkout 5 commits back to avoid the >>>>>>>>>>>> Load_Balancer issue that made good progress. >>>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>> Great to hear! >>>>>>>>>> >>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>>> But now i am seeing very odd behavior where floating IP can get >>>>>>>>>>>> exposed to BGP but when i am trying to expose VM Tenant IP then I get a >>>>>>>>>>>> strange error. Here is the full logs output; >>>>>>>>>>>> https://paste.opendev.org/show/buRbY415guvHFUtSapFK/ >>>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>> I was using the evpn mode for the tenant networks, so it has been >>>>>>>>>> a while since I tested the vm tenant ip with plain bgp. Perhaps we broke it >>>>>>>>>> with some of the new features/reshapes. >>>>>>>>>> >>>>>>>>>> Regarding the error, two things: >>>>>>>>>> - Do you have several hosts or just one? The VM IP should be >>>>>>>>>> exposed in the host holding the ovn cr-lrp port (ovn router gateway port >>>>>>>>>> connecting the router to the provider network). >>>>>>>>>> - The error there seems to be related to the re-sync task. I've >>>>>>>>>> hit some issues with pyroute2 in the past, please be sure to use a recent >>>>>>>>>> enough version (minimum supported is 0.6.4) >>>>>>>>>> >>>>>>>>>> >>>>>>>>>>>> >>>>>>>>>>>> On Fri, Jul 22, 2022 at 8:55 AM Satish Patel < >>>>>>>>>>>> satish.txt(a)gmail.com> wrote: >>>>>>>>>>>> >>>>>>>>>>>>> Hi Luis, >>>>>>>>>>>>> >>>>>>>>>>>>> Thank you for reply, >>>>>>>>>>>>> >>>>>>>>>>>>> Let me tell you that your blog is wonderful. I have used your >>>>>>>>>>>>> method to install frr without any docker container. >>>>>>>>>>>>> >>>>>>>>>>>>> What is the workaround here? Can I tell ovn-bgp-agent to not >>>>>>>>>>>>> look for container of frr? >>>>>>>>>>>>> >>>>>>>>>>>> >>>>>>>>>> You just need to point where the frr socket is on the host. >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>>>> I have notice one more thing that it’s trying to run “copy >>>>>>>>>>>>> /tmp/blah running-config” but that command isn’t supported. >>>>>>>>>>>>> >>>>>>>>>>>>> I have tried to run copy command manually on vtysh shell to >>>>>>>>>>>>> see what options are available for copy command but there is only one >>>>>>>>>>>>> command available with copy which is “copy running-config startup-config” >>>>>>>>>>>>> do you think I have wrong version of frr running? I’m running 7.2. >>>>>>>>>>>>> >>>>>>>>>>>> >>>>>>>>>> I think the minimum version I tried was either 7.4 or 7.5. >>>>>>>>>> >>>>>>>>>> >>>>>>>>>>>>> Please let me know if any workaround here. Thank you >>>>>>>>>>>>> >>>>>>>>>>>> >>>>>>>>>> No easy workaround for this, as it is relying on that way to >>>>>>>>>> reconfigure the frr.conf >>>>>>>>>> >>>>>>>>>> Cheers, >>>>>>>>>> Luis >>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>>>> Sent from my iPhone >>>>>>>>>>>>> >>>>>>>>>>>>> On Jul 22, 2022, at 2:41 AM, Luis Tomas Bolivar < >>>>>>>>>>>>> ltomasbo(a)redhat.com> wrote: >>>>>>>>>>>>> >>>>>>>>>>>>> >>>>>>>>>>>>> Hi Satish, >>>>>>>>>>>>> >>>>>>>>>>>>> The one to use should be https://opendev.org/x/ovn-bgp-agent. >>>>>>>>>>>>> The one on my personal github repo was the initial PoC for it. But the >>>>>>>>>>>>> opendev one is the upstream effort to develop it, and is the one being >>>>>>>>>>>>> maintained/updated. >>>>>>>>>>>>> >>>>>>>>>>>>> Looking at your second logs, it seems you are missing FRR (and >>>>>>>>>>>>> its shell, vtysh) in the node. >>>>>>>>>>>>> >>>>>>>>>>>>> Actually, thinking about this: >>>>>>>>>>>>> "Unexpected error while running command. >>>>>>>>>>>>> Command: /usr/bin/vtysh --vty_socket /run/frr/ -c copy >>>>>>>>>>>>> /tmp/tmpiz5s_wvs running-config" >>>>>>>>>>>>> >>>>>>>>>>>>> The ovn-bgp-agent has been developed with "deploying on >>>>>>>>>>>>> containers" in mind, meaning it is assuming there is a frr container >>>>>>>>>>>>> running, and the container running the agent is trying to connect to the >>>>>>>>>>>>> same socket so that it can run the vtysh commands. Perhaps in your case the >>>>>>>>>>>>> frr socket is in a different location than /run/frr/ >>>>>>>>>>>>> >>>>>>>>>>>>> On Fri, Jul 22, 2022 at 6:27 AM Satish Patel < >>>>>>>>>>>>> satish.txt(a)gmail.com> wrote: >>>>>>>>>>>>> >>>>>>>>>>>>>> Folks, >>>>>>>>>>>>>> >>>>>>>>>>>>>> I am trying to create lab of of ovn-bgp-agent using this blog >>>>>>>>>>>>>> https://ltomasbo.wordpress.com/2021/02/04/ovn-bgp-agent-testing-setup/ >>>>>>>>>>>>>> >>>>>>>>>>>>>> >>>>>>>>>>>>>> So far everything went well but I'm stuck at the bgp-agent >>>>>>>>>>>>>> installation and I encounter following error when running bgp-agent. >>>>>>>>>>>>>> Any suggestions? >>>>>>>>>>>>>> >>>>>>>>>>>>>> root@rack-1-host-2:/home/vagrant/bgp-agent# bgp-agent >>>>>>>>>>>>>> 2022-07-22 04:02:39.123 111551 INFO bgp_agent.config [-] >>>>>>>>>>>>>> Logging enabled! >>>>>>>>>>>>>> 2022-07-22 04:02:39.475 111551 CRITICAL bgp-agent [-] >>>>>>>>>>>>>> Unhandled error: AssertionError >>>>>>>>>>>>>> 2022-07-22 04:02:39.475 111551 ERROR bgp-agent Traceback >>>>>>>>>>>>>> (most recent call last): >>>>>>>>>>>>>> 2022-07-22 04:02:39.475 111551 ERROR bgp-agent File >>>>>>>>>>>>>> "/usr/local/bin/bgp-agent", line 10, in <module> >>>>>>>>>>>>>> 2022-07-22 04:02:39.475 111551 ERROR bgp-agent >>>>>>>>>>>>>> sys.exit(start()) >>>>>>>>>>>>>> 2022-07-22 04:02:39.475 111551 ERROR bgp-agent File >>>>>>>>>>>>>> "/usr/local/lib/python3.8/dist-packages/bgp_agent/agent.py", line 76, in >>>>>>>>>>>>>> start >>>>>>>>>>>>>> 2022-07-22 04:02:39.475 111551 ERROR bgp-agent >>>>>>>>>>>>>> bgp_agent_launcher = service.launch(config.CONF, BGPAgent()) >>>>>>>>>>>>>> 2022-07-22 04:02:39.475 111551 ERROR bgp-agent File >>>>>>>>>>>>>> "/usr/local/lib/python3.8/dist-packages/bgp_agent/agent.py", line 44, in >>>>>>>>>>>>>> __init__ >>>>>>>>>>>>>> 2022-07-22 04:02:39.475 111551 ERROR bgp-agent >>>>>>>>>>>>>> self.agent_driver = driver_api.AgentDriverBase.get_instance( >>>>>>>>>>>>>> 2022-07-22 04:02:39.475 111551 ERROR bgp-agent File >>>>>>>>>>>>>> "/usr/local/lib/python3.8/dist-packages/bgp_agent/platform/driver_api.py", >>>>>>>>>>>>>> line 25, in get_instance >>>>>>>>>>>>>> 2022-07-22 04:02:39.475 111551 ERROR bgp-agent >>>>>>>>>>>>>> agent_driver = stevedore_driver.DriverManager( >>>>>>>>>>>>>> 2022-07-22 04:02:39.475 111551 ERROR bgp-agent File >>>>>>>>>>>>>> "/usr/local/lib/python3.8/dist-packages/stevedore/driver.py", line 54, in >>>>>>>>>>>>>> __init__ >>>>>>>>>>>>>> 2022-07-22 04:02:39.475 111551 ERROR bgp-agent >>>>>>>>>>>>>> super(DriverManager, self).__init__( >>>>>>>>>>>>>> 2022-07-22 04:02:39.475 111551 ERROR bgp-agent File >>>>>>>>>>>>>> "/usr/local/lib/python3.8/dist-packages/stevedore/named.py", line 78, in >>>>>>>>>>>>>> __init__ >>>>>>>>>>>>>> 2022-07-22 04:02:39.475 111551 ERROR bgp-agent extensions >>>>>>>>>>>>>> = self._load_plugins(invoke_on_load, >>>>>>>>>>>>>> 2022-07-22 04:02:39.475 111551 ERROR bgp-agent File >>>>>>>>>>>>>> "/usr/local/lib/python3.8/dist-packages/stevedore/extension.py", line 221, >>>>>>>>>>>>>> in _load_plugins >>>>>>>>>>>>>> 2022-07-22 04:02:39.475 111551 ERROR bgp-agent ext = >>>>>>>>>>>>>> self._load_one_plugin(ep, >>>>>>>>>>>>>> 2022-07-22 04:02:39.475 111551 ERROR bgp-agent File >>>>>>>>>>>>>> "/usr/local/lib/python3.8/dist-packages/stevedore/named.py", line 156, in >>>>>>>>>>>>>> _load_one_plugin >>>>>>>>>>>>>> 2022-07-22 04:02:39.475 111551 ERROR bgp-agent return >>>>>>>>>>>>>> super(NamedExtensionManager, self)._load_one_plugin( >>>>>>>>>>>>>> 2022-07-22 04:02:39.475 111551 ERROR bgp-agent File >>>>>>>>>>>>>> "/usr/local/lib/python3.8/dist-packages/stevedore/extension.py", line 257, >>>>>>>>>>>>>> in _load_one_plugin >>>>>>>>>>>>>> 2022-07-22 04:02:39.475 111551 ERROR bgp-agent obj = >>>>>>>>>>>>>> plugin(*invoke_args, **invoke_kwds) >>>>>>>>>>>>>> 2022-07-22 04:02:39.475 111551 ERROR bgp-agent File >>>>>>>>>>>>>> "/usr/local/lib/python3.8/dist-packages/bgp_agent/platform/osp/ovn_bgp_driver.py", >>>>>>>>>>>>>> line 64, in __init__ >>>>>>>>>>>>>> 2022-07-22 04:02:39.475 111551 ERROR bgp-agent >>>>>>>>>>>>>> self._sb_idl = ovn.OvnSbIdl( >>>>>>>>>>>>>> 2022-07-22 04:02:39.475 111551 ERROR bgp-agent File >>>>>>>>>>>>>> "/usr/local/lib/python3.8/dist-packages/bgp_agent/platform/osp/utils/ovn.py", >>>>>>>>>>>>>> line 62, in __init__ >>>>>>>>>>>>>> 2022-07-22 04:02:39.475 111551 ERROR bgp-agent >>>>>>>>>>>>>> super(OvnSbIdl, self).__init__( >>>>>>>>>>>>>> 2022-07-22 04:02:39.475 111551 ERROR bgp-agent File >>>>>>>>>>>>>> "/usr/local/lib/python3.8/dist-packages/bgp_agent/platform/osp/utils/ovn.py", >>>>>>>>>>>>>> line 31, in __init__ >>>>>>>>>>>>>> 2022-07-22 04:02:39.475 111551 ERROR bgp-agent >>>>>>>>>>>>>> super(OvnIdl, self).__init__(remote, schema) >>>>>>>>>>>>>> 2022-07-22 04:02:39.475 111551 ERROR bgp-agent File >>>>>>>>>>>>>> "/usr/local/lib/python3.8/dist-packages/ovs/db/idl.py", line 283, in >>>>>>>>>>>>>> __init__ >>>>>>>>>>>>>> 2022-07-22 04:02:39.475 111551 ERROR bgp-agent schema = >>>>>>>>>>>>>> schema_helper.get_idl_schema() >>>>>>>>>>>>>> 2022-07-22 04:02:39.475 111551 ERROR bgp-agent File >>>>>>>>>>>>>> "/usr/local/lib/python3.8/dist-packages/ovs/db/idl.py", line 2323, in >>>>>>>>>>>>>> get_idl_schema >>>>>>>>>>>>>> 2022-07-22 04:02:39.475 111551 ERROR bgp-agent >>>>>>>>>>>>>> self._keep_table_columns(schema, table, columns)) >>>>>>>>>>>>>> 2022-07-22 04:02:39.475 111551 ERROR bgp-agent File >>>>>>>>>>>>>> "/usr/local/lib/python3.8/dist-packages/ovs/db/idl.py", line 2330, in >>>>>>>>>>>>>> _keep_table_columns >>>>>>>>>>>>>> 2022-07-22 04:02:39.475 111551 ERROR bgp-agent assert >>>>>>>>>>>>>> table_name in schema.tables >>>>>>>>>>>>>> 2022-07-22 04:02:39.475 111551 ERROR bgp-agent AssertionError >>>>>>>>>>>>>> 2022-07-22 04:02:39.475 111551 ERROR bgp-agent >>>>>>>>>>>>>> >>>>>>>>>>>>>> >>>>>>>>>>>>>> >>>>>>>>>>>>>> >>>>>>>>>>>>>> After googling I found one more agent at >>>>>>>>>>>>>> https://opendev.org/x/ovn-bgp-agent and its also throwing an >>>>>>>>>>>>>> error. Which agent should I be using? >>>>>>>>>>>>>> >>>>>>>>>>>>>> root@rack-1-host-2:~# ovn-bgp-agent >>>>>>>>>>>>>> 2022-07-22 04:04:36.780 111761 INFO ovn_bgp_agent.config [-] >>>>>>>>>>>>>> Logging enabled! >>>>>>>>>>>>>> 2022-07-22 04:04:37.247 111761 INFO ovn_bgp_agent.agent [-] >>>>>>>>>>>>>> Service 'BGPAgent' stopped >>>>>>>>>>>>>> 2022-07-22 04:04:37.248 111761 INFO ovn_bgp_agent.agent [-] >>>>>>>>>>>>>> Service 'BGPAgent' starting >>>>>>>>>>>>>> 2022-07-22 04:04:37.248 111761 INFO >>>>>>>>>>>>>> ovn_bgp_agent.drivers.openstack.utils.frr [-] Add VRF leak for VRF >>>>>>>>>>>>>> ovn-bgp-vrf on router bgp 64999 >>>>>>>>>>>>>> 2022-07-22 04:04:37.248 111761 INFO oslo.privsep.daemon [-] >>>>>>>>>>>>>> Running privsep helper: ['sudo', 'privsep-helper', '--privsep_context', >>>>>>>>>>>>>> 'ovn_bgp_agent.privileged.vtysh_cmd', '--privsep_sock_path', >>>>>>>>>>>>>> '/tmp/tmp4cie9eiz/privsep.sock'] >>>>>>>>>>>>>> 2022-07-22 04:04:37.687 111761 INFO oslo.privsep.daemon [-] >>>>>>>>>>>>>> Spawned new privsep daemon via rootwrap >>>>>>>>>>>>>> 2022-07-22 04:04:37.598 111769 INFO oslo.privsep.daemon [-] >>>>>>>>>>>>>> privsep daemon starting >>>>>>>>>>>>>> 2022-07-22 04:04:37.613 111769 INFO oslo.privsep.daemon [-] >>>>>>>>>>>>>> privsep process running with uid/gid: 0/0 >>>>>>>>>>>>>> 2022-07-22 04:04:37.617 111769 INFO oslo.privsep.daemon [-] >>>>>>>>>>>>>> privsep process running with capabilities (eff/prm/inh): >>>>>>>>>>>>>> CAP_NET_ADMIN|CAP_SYS_ADMIN/CAP_NET_ADMIN|CAP_SYS_ADMIN/none >>>>>>>>>>>>>> 2022-07-22 04:04:37.617 111769 INFO oslo.privsep.daemon [-] >>>>>>>>>>>>>> privsep daemon running as pid 111769 >>>>>>>>>>>>>> 2022-07-22 04:04:37.987 111769 ERROR >>>>>>>>>>>>>> ovn_bgp_agent.privileged.vtysh [-] Unable to execute vtysh with >>>>>>>>>>>>>> ['/usr/bin/vtysh', '--vty_socket', '/run/frr/', '-c', 'copy >>>>>>>>>>>>>> /tmp/tmpiz5s_wvs running-config']. Exception: Unexpected error while >>>>>>>>>>>>>> running command. >>>>>>>>>>>>>> Command: /usr/bin/vtysh --vty_socket /run/frr/ -c copy >>>>>>>>>>>>>> /tmp/tmpiz5s_wvs running-config >>>>>>>>>>>>>> Exit code: 1 >>>>>>>>>>>>>> Stdout: '% Unknown command: copy /tmp/tmpiz5s_wvs >>>>>>>>>>>>>> running-config\n' >>>>>>>>>>>>>> Stderr: '' >>>>>>>>>>>>>> Traceback (most recent call last): >>>>>>>>>>>>>> File >>>>>>>>>>>>>> "/usr/local/lib/python3.8/dist-packages/ovn_bgp_agent/privileged/vtysh.py", >>>>>>>>>>>>>> line 30, in run_vtysh_config >>>>>>>>>>>>>> return processutils.execute(*full_args) >>>>>>>>>>>>>> File >>>>>>>>>>>>>> "/usr/local/lib/python3.8/dist-packages/oslo_concurrency/processutils.py", >>>>>>>>>>>>>> line 438, in execute >>>>>>>>>>>>>> raise ProcessExecutionError(exit_code=_returncode, >>>>>>>>>>>>>> oslo_concurrency.processutils.ProcessExecutionError: >>>>>>>>>>>>>> Unexpected error while running command. >>>>>>>>>>>>>> Command: /usr/bin/vtysh --vty_socket /run/frr/ -c copy >>>>>>>>>>>>>> /tmp/tmpiz5s_wvs running-config >>>>>>>>>>>>>> Exit code: 1 >>>>>>>>>>>>>> Stdout: '% Unknown command: copy /tmp/tmpiz5s_wvs >>>>>>>>>>>>>> running-config\n' >>>>>>>>>>>>>> Stderr: '' >>>>>>>>>>>>>> 2022-07-22 04:04:37.990 111761 ERROR oslo_service.service [-] >>>>>>>>>>>>>> Error starting thread.: >>>>>>>>>>>>>> oslo_concurrency.processutils.ProcessExecutionError: Unexpected error while >>>>>>>>>>>>>> running command. >>>>>>>>>>>>>> Command: /usr/bin/vtysh --vty_socket /run/frr/ -c copy >>>>>>>>>>>>>> /tmp/tmpiz5s_wvs running-config >>>>>>>>>>>>>> Exit code: 1 >>>>>>>>>>>>>> Stdout: '% Unknown command: copy /tmp/tmpiz5s_wvs >>>>>>>>>>>>>> running-config\n' >>>>>>>>>>>>>> Stderr: '' >>>>>>>>>>>>>> 2022-07-22 04:04:37.990 111761 ERROR oslo_service.service >>>>>>>>>>>>>> Traceback (most recent call last): >>>>>>>>>>>>>> 2022-07-22 04:04:37.990 111761 ERROR oslo_service.service >>>>>>>>>>>>>> File "/usr/local/lib/python3.8/dist-packages/oslo_service/service.py", line >>>>>>>>>>>>>> 806, in run_service >>>>>>>>>>>>>> 2022-07-22 04:04:37.990 111761 ERROR oslo_service.service >>>>>>>>>>>>>> service.start() >>>>>>>>>>>>>> 2022-07-22 04:04:37.990 111761 ERROR oslo_service.service >>>>>>>>>>>>>> File "/usr/local/lib/python3.8/dist-packages/ovn_bgp_agent/agent.py", line >>>>>>>>>>>>>> 50, in start >>>>>>>>>>>>>> 2022-07-22 04:04:37.990 111761 ERROR oslo_service.service >>>>>>>>>>>>>> self.agent_driver.start() >>>>>>>>>>>>>> 2022-07-22 04:04:37.990 111761 ERROR oslo_service.service >>>>>>>>>>>>>> File >>>>>>>>>>>>>> "/usr/local/lib/python3.8/dist-packages/ovn_bgp_agent/drivers/openstack/ovn_bgp_driver.py", >>>>>>>>>>>>>> line 73, in start >>>>>>>>>>>>>> 2022-07-22 04:04:37.990 111761 ERROR oslo_service.service >>>>>>>>>>>>>> frr.vrf_leak(constants.OVN_BGP_VRF, CONF.bgp_AS, CONF.bgp_router_id) >>>>>>>>>>>>>> 2022-07-22 04:04:37.990 111761 ERROR oslo_service.service >>>>>>>>>>>>>> File >>>>>>>>>>>>>> "/usr/local/lib/python3.8/dist-packages/ovn_bgp_agent/drivers/openstack/utils/frr.py", >>>>>>>>>>>>>> line 110, in vrf_leak >>>>>>>>>>>>>> 2022-07-22 04:04:37.990 111761 ERROR oslo_service.service >>>>>>>>>>>>>> _run_vtysh_config_with_tempfile(vrf_config) >>>>>>>>>>>>>> 2022-07-22 04:04:37.990 111761 ERROR oslo_service.service >>>>>>>>>>>>>> File >>>>>>>>>>>>>> "/usr/local/lib/python3.8/dist-packages/ovn_bgp_agent/drivers/openstack/utils/frr.py", >>>>>>>>>>>>>> line 93, in _run_vtysh_config_with_tempfile >>>>>>>>>>>>>> 2022-07-22 04:04:37.990 111761 ERROR oslo_service.service >>>>>>>>>>>>>> ovn_bgp_agent.privileged.vtysh.run_vtysh_config(f.name) >>>>>>>>>>>>>> 2022-07-22 04:04:37.990 111761 ERROR oslo_service.service >>>>>>>>>>>>>> File "/usr/local/lib/python3.8/dist-packages/oslo_privsep/priv_context.py", >>>>>>>>>>>>>> line 271, in _wrap >>>>>>>>>>>>>> 2022-07-22 04:04:37.990 111761 ERROR oslo_service.service >>>>>>>>>>>>>> return self.channel.remote_call(name, args, kwargs, >>>>>>>>>>>>>> 2022-07-22 04:04:37.990 111761 ERROR oslo_service.service >>>>>>>>>>>>>> File "/usr/local/lib/python3.8/dist-packages/oslo_privsep/daemon.py", line >>>>>>>>>>>>>> 215, in remote_call >>>>>>>>>>>>>> 2022-07-22 04:04:37.990 111761 ERROR oslo_service.service >>>>>>>>>>>>>> raise exc_type(*result[2]) >>>>>>>>>>>>>> 2022-07-22 04:04:37.990 111761 ERROR oslo_service.service >>>>>>>>>>>>>> oslo_concurrency.processutils.ProcessExecutionError: Unexpected error while >>>>>>>>>>>>>> running command. >>>>>>>>>>>>>> 2022-07-22 04:04:37.990 111761 ERROR oslo_service.service >>>>>>>>>>>>>> Command: /usr/bin/vtysh --vty_socket /run/frr/ -c copy /tmp/tmpiz5s_wvs >>>>>>>>>>>>>> running-config >>>>>>>>>>>>>> 2022-07-22 04:04:37.990 111761 ERROR oslo_service.service >>>>>>>>>>>>>> Exit code: 1 >>>>>>>>>>>>>> 2022-07-22 04:04:37.990 111761 ERROR oslo_service.service >>>>>>>>>>>>>> Stdout: '% Unknown command: copy /tmp/tmpiz5s_wvs running-config\n' >>>>>>>>>>>>>> 2022-07-22 04:04:37.990 111761 ERROR oslo_service.service >>>>>>>>>>>>>> Stderr: '' >>>>>>>>>>>>>> 2022-07-22 04:04:37.990 111761 ERROR oslo_service.service >>>>>>>>>>>>>> 2022-07-22 04:04:37.993 111761 INFO ovn_bgp_agent.agent [-] >>>>>>>>>>>>>> Service 'BGPAgent' stopping >>>>>>>>>>>>>> 2022-07-22 04:04:37.994 111761 INFO ovn_bgp_agent.agent [-] >>>>>>>>>>>>>> Service 'BGPAgent' stopped >>>>>>>>>>>>>> >>>>>>>>>>>>> >>>>>>>>>>>>> >>>>>>>>>>>>> -- >>>>>>>>>>>>> LUIS TOMÁS BOLÍVAR >>>>>>>>>>>>> Principal Software Engineer >>>>>>>>>>>>> Red Hat >>>>>>>>>>>>> Madrid, Spain >>>>>>>>>>>>> ltomasbo(a)redhat.com >>>>>>>>>>>>> >>>>>>>>>>>>> >>>>>>>>>>>>> >>>>>>>>>> >>>>>>>>>> -- >>>>>>>>>> LUIS TOMÁS BOLÍVAR >>>>>>>>>> Principal Software Engineer >>>>>>>>>> Red Hat >>>>>>>>>> Madrid, Spain >>>>>>>>>> ltomasbo(a)redhat.com >>>>>>>>>> >>>>>>>>>> >>>>>>>>> >>>>>>>> >>>>>>>> -- >>>>>>>> LUIS TOMÁS BOLÍVAR >>>>>>>> Principal Software Engineer >>>>>>>> Red Hat >>>>>>>> Madrid, Spain >>>>>>>> ltomasbo(a)redhat.com >>>>>>>> >>>>>>>> >>>>>>>> >>> >>> -- >>> LUIS TOMÁS BOLÍVAR >>> Principal Software Engineer >>> Red Hat >>> Madrid, Spain >>> ltomasbo(a)redhat.com >>> >>> >> > > -- > LUIS TOMÁS BOLÍVAR > Principal Software Engineer > Red Hat > Madrid, Spain > ltomasbo(a)redhat.com > >

2 1

[ironic][xena] problems updating redfish_password for existing node
by Wade Albright 23 Aug '22

23 Aug '22

Hi, I'm hitting a problem when trying to update the redfish_password for an existing node. I'm curious to know if anyone else has encountered this problem. I'm not sure if I'm just doing something wrong or if there is a bug. Or if the problem is unique to my setup. I have a node already added into ironic with all the driver details set, and things are working fine. I am able to run deployments. Now I need to change the redfish password on the host. So I update the password for redfish access on the host, then use an 'openstack baremetal node set <node> --driver-info redfish_password=<newpass>' command to set the new redfish_password. Once this has been done, deployment no longer works. I see redfish authentication errors in the logs and the operation fails. I waited a bit to see if there might just be a delay in updating the password, but after awhile it still didn't work. I restarted the conductor, and after that things work fine again. So it seems like the password is cached or something. Is there a way to force the password to update? I even tried removing the redfish credentials and re-adding them, but that didn't work either. Only a conductor restart seems to make the new password work. We are running Xena, using rpm installation on Oracle Linux 8.5. Thanks in advance for any help with this issue.

3 21

RE: [glance][cinder] Glance cinder backend or image cache ?
by Simon Dodsley 23 Aug '22

23 Aug '22

> Hello ! > > We're planning a new cluster without Ceph storage. > We will use an iSCSI pureStorage array for all storage except object > storage. > > PureStorage provide only a cinder driver. > So we will need to user boot on volume nova instances. > > My question is about glance. > PureStorage speak about Image Cache since Liberty: > https://support.purestorage.com/Solutions/OpenStack/z_Legacy_OpenStack_Refe… > > But I wonder if using the cinder backend of glance is a better, simpler, > transparent option ? The Image Cache you referenced is not specific to Pure. It is a common Cinder feature that many backend can leverage. You can use both the image cache and cinder as a glance store as well but they are redundant. It is best to pick one or the other. It depends on the images format you are using. For example if you are using QCOW2 as the format the glance cinder backend cannot optimize it, but the image cache can optimize. Simon

1 0

[TripleO] Adding Undercloud and overcloud container images to Foreman
by Ruslanas Gžibovskis 23 Aug '22

23 Aug '22

Hi all, Have anyone tried adding a container images registry from quay.io to the foreman installation? So far what I managed to achieve is - adding each image one by one, not a registry itself. I add it over Content > Products > (Added Product docker-registries) > Repositories > New Repository > Name > Type (select docker) > Upstream URL: quay.io > Upstream repository (tripleoussuri or tripleowallaby or any other) fails, it works only if I specify full image: tripleoussuri/centos-binary-heat-engine So my question, is there any way to copy whole repo using Foreman? Maybe I just go in incorrect place? Thanks. Regards -- Ruslanas Gžibovskis +370 6030 7030

1 0

[glance][cinder] Glance cinder backend or image cache ?
by Gilles Mocellin 23 Aug '22

23 Aug '22

Hello ! We're planning a new cluster without Ceph storage. We will use an iSCSI pureStorage array for all storage except object storage. PureStorage provide only a cinder driver. So we will need to user boot on volume nova instances. My question is about glance. PureStorage speak about Image Cache since Liberty: https://support.purestorage.com/Solutions/OpenStack/z_Legacy_OpenStack_Refe… But I wonder if using the cinder backend of glance is a better, simpler, transparent option ?

1 0

[openstack-helm] No Meeting Tomorrow
by Gage Hugo 22 Aug '22

22 Aug '22

Hey team, Since there is nothing on the agenda for tomorrow's meeting, it has been cancelled. We will plan on meeting next week at the usual time.

1 0

[edge] OpenStack updates for an Akraino presentation
by Ildiko Vancsa 22 Aug '22

22 Aug '22

Hi OpenStack community, I’m reaching out to you as I would like to ask for a little help and pointers for a presentation I’m building about OpenStack. I was recently asked by the Akraino community to give a presentation about recently added features and enhancements and further updates in OpenStack. Akraino is a project that is primarily focused on edge computing use cases and building blocks of end-to-end edge infrastructures. To make my presentation relevant and most informative for them, I would like to highlight any recent development or ongoing work item that is, or can be, relevant for edge use cases. If there are features in your project that you think would be good to highlight or if your project team is working on relevant work items in the current release cycle, I would really appreciate if you could share pointers to any information about them! Thanks and Best Regards, Ildikó ——— Ildikó Váncsa Senior Manager, Community & Ecosystem Open Infrastructure Foundation

1 0

[openstack-ansible] [yoga] utility_container failure
by Father Vlasie 22 Aug '22

22 Aug '22

Hello everyone, I have happily progressed to the second step of running the playbooks, namely "openstack-ansible setup-infrastructure.yml" Everything looks good except for just one error which is mystifying me: ---------------- TASK [Get list of repo packages] ********************************************************************************************************************************************************** fatal: [infra1_utility_container-5ec32cb5]: FAILED! => {"changed": false, "content": "", "elapsed": 30, "msg": "Status code was -1 and not [200]: Request failed: <urlopen error timed out>", "redirected": false, "status": -1, "url": "http://192.168.3.9:8181/constraints/upper_constraints_cached.txt"} ---------------- 192.168.3.9 is the IP listed in user_variables.yml under haproxy_keepalived_internal_vip_cidr Any help or pointers would be very much appreciated! Thank you, Father Vlasie

2 9

evacuate vm
by fereshteh loghmani 22 Aug '22

22 Aug '22

hello i use kolla ansiable with wallaby version. when i evacuate server some of my vm with centos OS have error about xfs and for solve this issue i change server to rescue mode and from rescue i use command xfs_repair -l and my server will be ok and boot successfully. could you help me about this issue that why some of my vm had xfs error ?

1 0

Kolla-ansible Xena installation Error
by KK CHN 22 Aug '22

22 Aug '22

I am following the documentation https://docs.openstack.org/project-deploy-guide/kolla-ansible/xena/quicksta… on VM with Debian 10 distro But I am getting the errors as follows . Here is my /etc/hosts whats wrong here ?? ############################################## (venv) cloud@Xena:~$ cat /etc/hosts 127.0.0.1 localhost #127.0.1.1 Xena #10.184.48.94 localhost Xena # The following lines are desirable for IPv6 capable hosts ::1 localhost ip6-localhost ip6-loopback ff02::1 ip6-allnodes ff02::2 ip6-allrouters # BEGIN ANSIBLE GENERATED HOSTS 10.184.48.94 Xena # END ANSIBLE GENERATED HOSTS (venv) cloud@Xena:~$ ################################################### ERROR fatal: [localhost]: FAILED! => {"changed": false, "module_stderr": "sudo: a password is required\n", "module_stdout": "", "msg": "MODULE FAILURE\nSee stdout/stderr for the exact error", "rc": 1} I am trying this installation on a Virtual Machine on Debian 10 OS. (venv) cloud@Xena:~$ kolla-ansible -i ./all-in-one bootstrap-servers Bootstrapping servers : ansible-playbook -e @/etc/kolla/globals.yml -e @/etc/kolla/passwords.yml -e CONFIG_DIR=/etc/kolla -e kolla_action=bootstrap-servers /home/cloud/venv/share/kolla-ansible/ansible/kolla-host.yml --inventory ./all-in-one [WARNING]: Invalid characters were found in group names but not replaced, use -vvvv to see details PLAY [Gather facts for all hosts] *************************************************************************************************************************************** TASK [Gather facts] ***************************************************************************************************************************************************** ok: [localhost] TASK [Group hosts to determine when using --limit] ********************************************************************************************************************** ok: [localhost] [WARNING]: Could not match supplied host pattern, ignoring: all_using_limit_True PLAY [Gather facts for all hosts (if using --limit)] ******************************************************************************************************************** skipping: no hosts matched PLAY [Apply role baremetal] ********************************************************************************************************************************************* TASK [baremetal : include_tasks] **************************************************************************************************************************************** included: /home/cloud/venv/share/kolla-ansible/ansible/roles/baremetal/tasks/bootstrap-servers.yml for localhost TASK [baremetal : Ensure localhost in /etc/hosts] *********************************************************************************************************************** fatal: [localhost]: FAILED! => {"changed": false, "module_stderr": "sudo: a password is required\n", "module_stdout": "", "msg": "MODULE FAILURE\nSee stdout/stderr for the exact error", "rc": 1} PLAY RECAP ************************************************************************************************************************************************************** localhost : ok=3 changed=0 unreachable=0 failed=1 skipped=0 rescued=0 ignored=0 Command failed ansible-playbook -e @/etc/kolla/globals.yml -e @/etc/kolla/passwords.yml -e CONFIG_DIR=/etc/kolla -e kolla_action=bootstrap-servers /home/cloud/venv/share/kolla-ansible/ansible/kolla-host.yml --inventory ./all-in-one (venv) cloud@Xena:~$ Any hints welcome. Thank you all krish

2 1