| *** seroyer has quit IRC | 00:00 | |
| *** jwcroppe has joined #openstack-powervm | 00:03 | |
| *** jwcroppe has quit IRC | 00:10 | |
| *** clbush__ has joined #openstack-powervm | 00:14 | |
| *** thorst has quit IRC | 00:32 | |
| *** thorst has joined #openstack-powervm | 00:33 | |
| *** thorst has quit IRC | 00:42 | |
| *** seroyer has joined #openstack-powervm | 00:47 | |
| *** seroyer has quit IRC | 00:54 | |
| *** clbush__ has quit IRC | 01:14 | |
| *** jwcroppe has joined #openstack-powervm | 01:16 | |
| *** thorst has joined #openstack-powervm | 01:21 | |
| *** thorst has quit IRC | 01:21 | |
| *** thorst has joined #openstack-powervm | 01:22 | |
| *** thorst has quit IRC | 01:30 | |
| *** thorst has joined #openstack-powervm | 01:47 | |
| *** thorst has quit IRC | 01:48 | |
| *** esberglu has quit IRC | 02:35 | |
| *** thorst has joined #openstack-powervm | 02:48 | |
| *** thorst has quit IRC | 02:57 | |
| *** tjakobs has joined #openstack-powervm | 03:51 | |
| *** thorst has joined #openstack-powervm | 03:55 | |
| *** thorst has quit IRC | 04:02 | |
| *** shyama has joined #openstack-powervm | 04:03 | |
| *** thorst has joined #openstack-powervm | 05:01 | |
| *** thorst has quit IRC | 05:08 | |
| *** tjakobs has quit IRC | 05:18 | |
| *** thorst has joined #openstack-powervm | 06:05 | |
| *** thorst has quit IRC | 06:12 | |
| *** thorst has joined #openstack-powervm | 07:10 | |
| *** thorst has quit IRC | 07:16 | |
| *** k0da has joined #openstack-powervm | 07:44 | |
| *** viclarson has joined #openstack-powervm | 07:51 | |
| *** thorst has joined #openstack-powervm | 08:15 | |
| viclarson | так зафига здесь две сетевухи? | 08:18 |
|---|---|---|
| *** thorst has quit IRC | 08:22 | |
| *** jwcroppe has quit IRC | 09:06 | |
| *** thorst has joined #openstack-powervm | 09:20 | |
| *** thorst has quit IRC | 09:27 | |
| *** viclarson has quit IRC | 10:03 | |
| *** thorst has joined #openstack-powervm | 10:25 | |
| *** thorst has quit IRC | 10:32 | |
| *** shyama has quit IRC | 10:43 | |
| *** shyama has joined #openstack-powervm | 11:05 | |
| *** viclarson has joined #openstack-powervm | 11:21 | |
| *** seroyer has joined #openstack-powervm | 11:28 | |
| *** thorst has joined #openstack-powervm | 11:40 | |
| *** jwcroppe has joined #openstack-powervm | 11:42 | |
| viclarson | <viclarson> padmin@novalink:~$ pvmctl vea list -i lpar_id=6 | 11:57 |
| viclarson | <viclarson> Virtual Ethernet Adapters | 11:57 |
| viclarson | <viclarson> +------+------------+------+--------------+------+-------+--------------+ | 11:57 |
| viclarson | <viclarson> | PVID | VSwitch | LPAR | MAC | Slot | Trunk | Tagged VLANs | | 11:57 |
| viclarson | <viclarson> +------+------------+------+--------------+------+-------+--------------+ | 11:57 |
| viclarson | <viclarson> | 604 | ETHERNET0 | 6 | FA163EA10B74 | 2 | False | | | 11:57 |
| viclarson | <viclarson> | 4094 | MGMTSWITCH | 6 | 5661137C0F52 | 3 | False | | | 11:57 |
| viclarson | <viclarson> +------+------------+------+--------------+------+-------+--------------+ | 11:57 |
| viclarson | <viclarson> why we need two interfaces at client vm? | 11:57 |
| viclarson | <viclarson> root@ubuntu:~# cat /etc/network/interfaces | 11:57 |
| viclarson | <viclarson> # Injected by Nova on instance boot | 11:57 |
| viclarson | <viclarson> # | 11:57 |
| viclarson | <viclarson> # This file describes the network interfaces available on your system | 11:57 |
| viclarson | <viclarson> # and how to activate them. For more information, see interfaces(5). | 11:57 |
| viclarson | <viclarson> # The loopback network interface | 11:57 |
| viclarson | <viclarson> auto lo | 11:57 |
| viclarson | <viclarson> iface lo inet loopback | 11:57 |
| viclarson | <viclarson> auto eth1 | 11:57 |
| viclarson | <viclarson> iface eth1 inet static | 11:58 |
| viclarson | <viclarson> hwaddress ether 56:61:13:7c:0f:52 | 11:58 |
| viclarson | <viclarson> address None | 11:58 |
| viclarson | <viclarson> netmask None | 11:58 |
| viclarson | <viclarson> broadcast None | 11:58 |
| viclarson | <viclarson> iface eth1 inet6 static | 11:58 |
| viclarson | <viclarson> hwaddress ether 56:61:13:7c:0f:52 | 11:58 |
| viclarson | <viclarson> address fe80::5461:13ff:fe7c:0f52 | 11:58 |
| viclarson | <viclarson> netmask 64 | 11:58 |
| viclarson | <viclarson> why eth0 is not configured? | 11:58 |
| *** k0da has quit IRC | 12:03 | |
| *** edmondsw has joined #openstack-powervm | 12:18 | |
| *** kylek3h has joined #openstack-powervm | 12:18 | |
| *** seroyer has quit IRC | 12:24 | |
| *** k0da has joined #openstack-powervm | 12:33 | |
| thorst | viclarson: are you using cloudinit or metadata? | 12:36 |
| *** jwcroppe has quit IRC | 12:39 | |
| *** burgerk has joined #openstack-powervm | 12:41 | |
| viclarson | we need cloudinit and metadata | 12:45 |
| viclarson | default way cloudinit obtain init script from metadata service | 12:46 |
| viclarson | but it doesnt have to be | 12:48 |
| viclarson | if you asking about murano we d obtain config for murano config and apply it in every way | 12:51 |
| viclarson | if you asking about murano we d obtain config for murano agent and apply it in every way | 12:51 |
| *** tblakes has joined #openstack-powervm | 12:53 | |
| thorst | viclarson: OK - do you see a cd-drive mounted on the system? We have some issues currently with the mixed IPv4 and IPv6 configs | 12:55 |
| *** seroyer has joined #openstack-powervm | 12:55 | |
| viclarson | root@ubuntu:~# mount -l | 12:56 |
| viclarson | proc on /proc type proc (rw,noexec,nosuid,nodev) | 12:56 |
| viclarson | sysfs on /sys type sysfs (rw,noexec,nosuid,nodev) | 12:56 |
| viclarson | none on /sys/fs/cgroup type tmpfs (rw) | 12:56 |
| viclarson | none on /sys/fs/fuse/connections type fusectl (rw) | 12:56 |
| viclarson | none on /sys/kernel/debug type debugfs (rw) | 12:56 |
| viclarson | none on /sys/kernel/security type securityfs (rw) | 12:56 |
| viclarson | udev on /dev type devtmpfs (rw,mode=0755) | 12:56 |
| viclarson | devpts on /dev/pts type devpts (rw,noexec,nosuid,gid=5,mode=0620) | 12:56 |
| viclarson | tmpfs on /run type tmpfs (rw,noexec,nosuid,size=10%,mode=0755) | 12:56 |
| viclarson | none on /run/lock type tmpfs (rw,noexec,nosuid,nodev,size=5242880) | 12:56 |
| viclarson | none on /run/shm type tmpfs (rw,nosuid,nodev) | 12:56 |
| viclarson | none on /run/user type tmpfs (rw,noexec,nosuid,nodev,size=104857600,mode=0755) | 12:57 |
| viclarson | none on /sys/fs/pstore type pstore (rw) | 12:57 |
| viclarson | systemd on /sys/fs/cgroup/systemd type cgroup (rw,noexec,nosuid,nodev,none,name=systemd) | 12:57 |
| viclarson | *-cdrom | 12:57 |
| viclarson | description: SCSI CD-ROM | 12:57 |
| viclarson | product: VOPTA | 12:57 |
| viclarson | vendor: AIX | 12:57 |
| viclarson | physical id: 0.2.0 | 12:57 |
| viclarson | bus info: scsi@1:0.2.0 | 12:57 |
| viclarson | logical name: /dev/cdrom | 12:57 |
| viclarson | logical name: /dev/sr0 | 12:57 |
| viclarson | capabilities: removable audio | 12:57 |
| viclarson | configuration: ansiversion=4 status=ready | 12:57 |
| viclarson | *-medium | 12:57 |
| viclarson | physical id: 0 | 12:57 |
| viclarson | logical name: /dev/cdrom | 12:57 |
| viclarson | this? | 12:57 |
| thorst | yeah...but look through here. This is applicable to your environment (it says PowerVC - but they use these drivers as well) | 12:58 |
| thorst | https://www.ibm.com/developerworks/community/wikis/home?lang=en_us#!/wiki/Wc1c29d23e0fd_4346_b509_f1c00a2099f0/page/Troubleshooting%20when%20a%20VM%20using%20cloud-init%20does%20not%20ping%20or%20boot | 12:58 |
| thorst | Section 1.2.2 is what I bet is wrong. I'd like to wait for kriskend to join IRC as she and I have some patches around this that haven't gotten into nova yet... | 12:59 |
| *** apearson has joined #openstack-powervm | 12:59 | |
| *** apearson_ has joined #openstack-powervm | 13:01 | |
| seroyer | thorst, are you refering to the interfaces.template? | 13:02 |
| thorst | seroyer: yeah | 13:02 |
| thorst | I need to find that link.. | 13:02 |
| thorst | efried adreznec seroyer apearson esberglu: https://review.openstack.org/#/c/381772/ | 13:04 |
| efried | thorst, any significant changes from the etherpad? | 13:04 |
| *** apearson has quit IRC | 13:04 | |
| seroyer | thorst, this the link you are looking for? https://review.openstack.org/#/c/264157/5/nova/virt/interfaces.template | 13:04 |
| thorst | efried: yeah, please re-review | 13:04 |
| efried | on it. | 13:04 |
| thorst | seroyer: yeah...thanks. viclarson - recommend you patch the link that seroyer sent. We need to get that into main nova... | 13:05 |
| efried | thorst, is our CI providing non-gating votes for nova changes? | 13:06 |
| thorst | efried: yep :-) | 13:06 |
| thorst | well wait | 13:06 |
| efried | Okay, so line 32-35? | 13:06 |
| thorst | its running against the nova change sets. They don't want us voting. | 13:06 |
| thorst | even if non-gating | 13:07 |
| efried | Okay, so actually silent. | 13:07 |
| thorst | that came from the last mid cycle | 13:07 |
| viclarson | i dont understand all or nothing | 13:07 |
| viclarson | we have problem now | 13:07 |
| thorst | yeah, I think we publish the logs, but we don't have a link to them | 13:07 |
| viclarson | vm cant access controllee because mac filtering at wswitch | 13:07 |
| viclarson | is it possible to switch off mac filtering? | 13:08 |
| thorst | seroyer: I don't think so? | 13:08 |
| viclarson | http://www.ibm.com/support/knowledgecenter/SSB27U_6.3.0/com.ibm.zvm.v630.hcpb7/setfvs.htm in this manner? | 13:08 |
| viclarson | or using pvmctl | 13:09 |
| viclarson | i dont have access to hmc | 13:09 |
| viclarson | only novalink and not-fully funtion viosees | 13:09 |
| *** mdrabe has joined #openstack-powervm | 13:09 | |
| thorst | viclarson: you shouldn't need access to HMC...just novalink. But I think its more a question whether or not PowerVM can turn off mac filtering...if it can, then you should be able to tweak it | 13:10 |
| thorst | but I'm not sure we support it... | 13:10 |
| seroyer | A virtual Ethernet adapter can only send as self (single MAC address) and can only receive as self (no promiscuous support). A trunk adapter can send as any MAC, but will receive only frames whose destination MAC does not exist on the same VLAN/vSwitch pair. But you can only have a single active trunk on a VLAN at a time, so you can’t use SEA and another trunk at the same time. | 13:12 |
| thorst | so net is, unless the VM has a dedicated card you can't send. The client VM (workload - the thing you deployed) shouldn't need more. But the controller, the thing running the DHCP agent, L3, etc..., sounds like it needs a dedicated card. | 13:13 |
| viclarson | can dedicated card be attached to lpar by hmc? | 13:18 |
| viclarson | or by pvmctl? | 13:18 |
| viclarson | or i need to disassemble server? | 13:19 |
| thorst | pvmctl can do it | 13:21 |
| thorst | as long as you have a card that is plugged in, wired properly, etc... | 13:21 |
| thorst | you still need one on your VIOS to host the Shared Ethernet for the VMs | 13:21 |
| thorst | but this becomes a 'special' (pet) VM | 13:21 |
| efried | thorst, https://review.openstack.org/#/c/380762/ - is there some procedural reason the change set needs to be a cherry-pick? | 13:22 |
| thorst | efried: I got yelled at once from the neutron guys for not cherry picking | 13:22 |
| thorst | so I think so | 13:22 |
| efried | okay, I'll see what I can do. gerrit wouldn't let me cherry-pick, so I'll have to do it manually, and I'm not positive the end result will actually look any different once it gets up to gerrit. | 13:23 |
| apearson_ | viclarson - you would need to detach the slot from an existing lpar and attach it to another: pvmctl ioslot list/attach/detach | 13:23 |
| viclarson | why? | 13:23 |
| thorst | viclarson: who is that why to? Not sure as we had a couple convos in there. | 13:30 |
| efried | thorst, I see the difference: the cherry-pick winds up with the same change ID. It's done. | 13:32 |
| viclarson | asked apearson: why reattach slot from existing lpar to another? | 13:32 |
| thorst | viclarson: well, you only have to detach from an existing lpar if it is assigned to one currently | 13:33 |
| thorst | if its not assigned to a LPAR yet, just attach to VM | 13:33 |
| viclarson | detach what?) | 13:34 |
| viclarson | existing dedicated card? | 13:35 |
| thorst | viclarson: do you have multiple VIOSes or a single one? | 13:35 |
| viclarson | two | 13:35 |
| thorst | OK, cool. So each VIOS probably has a network card attached to it | 13:35 |
| thorst | and probably a FC card that hosts an SSP or something like that. | 13:35 |
| apearson_ | viclarson - so you asked about how to pass through a dedicated NIC to your lpar. You can get a list of the physical PCI slots on the system with 'pvmctl ioslot list'. You can then assign the slot directly (pass through) to the lpar with 'pvmctl ioslot attach'. however, more than likely it's already attached to an existing lpar - so you'll have to detach it from said lpar first. If it's not currently assigned to anything, then you can s | 13:35 |
| apearson_ | kip the detach step. | 13:35 |
| thorst | you need to attach a network card (because we allow direct attach of the card to the LPAR) to your controller. | 13:36 |
| thorst | so pvmctl ioslot list - you will see all the cards | 13:36 |
| thorst | you then need to find one you can attach to your lpar. Without cutting from your VIOSes | 13:36 |
| viclarson | padmin@novalink:~$ pvmctl ioslot list | 13:37 |
| viclarson | I/O Slots | 13:37 |
| viclarson | +--------------------------+------------------+-------------------------------------------------------+ | 13:37 |
| viclarson | | Loc Code (DRC Name) | LPAR | Description | | 13:37 |
| viclarson | +--------------------------+------------------+-------------------------------------------------------+ | 13:37 |
| viclarson | | U78C9.001.WZS0FY6-P1-C10 | vios1 | 1 Gigabit Ethernet (UTP) 4 Port Adapter PCIE-4x/Short | | 13:37 |
| viclarson | | U78C9.001.WZS0FY6-P1-C11 | | Empty slot | | 13:37 |
| viclarson | | U78C9.001.WZS0FY6-P1-C12 | vios2 | 1 Gigabit Ethernet (UTP) 4 Port Adapter PCIE-4x/Short | | 13:37 |
| viclarson | | U78C9.001.WZS0FY6-P1-C14 | vios1 | SAS RAID Controller, PCIe2, Dual-port 6Gb | | 13:37 |
| viclarson | | U78C9.001.WZS0FY6-P1-C15 | vios2 | SAS RAID Controller, PCIe2, Dual-port 6Gb | | 13:37 |
| viclarson | | U78C9.001.WZS0FY6-P1-C2 | vios1 | PCIe2 4-port (10Gb FCoE & 1GbE) SR&RJ45 Adapter | | 13:37 |
| viclarson | | U78C9.001.WZS0FY6-P1-C3 | vios1 | PCIe2 16Gb 2-Port Fibre Channel Adapter | | 13:37 |
| viclarson | | U78C9.001.WZS0FY6-P1-C4 | | Empty slot | | 13:37 |
| viclarson | | U78C9.001.WZS0FY6-P1-C5 | | Empty slot | | 13:37 |
| viclarson | | U78C9.001.WZS0FY6-P1-C6 | vios2 | PCIe2 16Gb 2-Port Fibre Channel Adapter | | 13:37 |
| viclarson | | U78C9.001.WZS0FY6-P1-C7 | vios2 | PCIe2 4-port (10Gb FCoE & 1GbE) SR&RJ45 Adapter | | 13:37 |
| viclarson | | U78C9.001.WZS0FY6-P1-C8 | | Empty slot | | 13:37 |
| viclarson | | U78C9.001.WZS0FY6-P1-C9 | | Empty slot | | 13:37 |
| viclarson | | U78C9.001.WZS0FY6-P1-T2 | novalink_2147CBW | Universal Serial Bus UHC Spec | | 13:37 |
| viclarson | +--------------------------+------------------+-------------------------------------------------------+ | 13:37 |
| viclarson | i like it U78C9.001.WZS0FY6-P1-C7 | 13:37 |
| thorst | viclarson: That may work...just make sure that isn't being used to host your SEA on your VIOS. | 13:39 |
| viclarson | d i attach it to controller vm? | 13:39 |
| apearson_ | viclarson - I assume P1-C7 is indeed an Ethernet adapter and not a FC adapter (sometimes it's hard to tell from the description). But yeah, make sure that device isn't in use on vios2 first. | 13:39 |
| thorst | apearson_: pvmctl won't let viclarson detach if its in use/ | 13:39 |
| viclarson | do you prefer this? U78C9.001.WZS0FY6-P1-C6 | 13:40 |
| thorst | viclarson: Not Fibre Channel...that's storage | 13:40 |
| thorst | C7, C12, C10, C2...those are all Ethernet adapters (I think) | 13:41 |
| thorst | at least two of those are in use on your vios | 13:41 |
| thorst | you should only detach what is not in use. | 13:41 |
| viclarson | and attach to controller? | 13:41 |
| thorst | viclarson: yes, I think so. Based on what I think you're doing :-) | 13:42 |
| thorst | and you need to of course make sure that card is plugged in, connected to a switch, etc... | 13:42 |
| apearson_ | viclarson - you can check if it's in use with: pvmctl sea list That will give you the 'enX' values. Then you can query the location codes with something like: viosvrcmd --id 3 -c "lsdev -field name physloc" | grep ent | 13:42 |
| viclarson | lets get novalink down second time | 13:46 |
| *** burgerk has quit IRC | 13:46 | |
| -openstackstatus- NOTICE: The Gerrit service on review.openstack.org is being restarted to address performance degradation and should return momentarily | 14:01 | |
| *** esberglu has joined #openstack-powervm | 14:03 | |
| *** burgerk has joined #openstack-powervm | 14:13 | |
| viclarson | padmin@novalink:~$ pvmctl ioslot attach --drc-names U78C9.001.WZS0FY6-P1-C12 -p id=4 | 14:14 |
| viclarson | [PVME01050402-0224] Partition communications not active on partition with ID 4 . Unable to run remote command. | 14:14 |
| *** tjakobs has joined #openstack-powervm | 14:14 | |
| apearson_ | viclarson - you need 'rmc' to be active on that partition. You need to have the 'rsct' packages installed to enable this (pvmctl lpar list -d name rmc_state). If you don't have the rsct packages installed, you'll need to power off the lpar first, attach the slot, then power it back on. | 14:15 |
| *** dwayne has quit IRC | 14:16 | |
| viclarson | target os is ubuntu | 14:16 |
| viclarson | is rsct package available for ubuntu? | 14:16 |
| viclarson | http://www.ibm.com/support/knowledgecenter/linuxonibm/liabu/liabuinstallras.htm | 14:19 |
| *** jwcroppe has joined #openstack-powervm | 14:21 | |
| apearson_ | viclarson - here's the repo: deb http://ppa.launchpad.net/ibmpackages/rsct/ubuntu xenial main | 14:23 |
| apearson_ | viclarson - that URL is out of date...it doesn't indicate that you can now get them directly from the PPA repo | 14:25 |
| viclarson | ty | 14:25 |
| thorst | apearson_: is there a marker debian that can be installed? | 14:25 |
| apearson_ | thorst - what is a marker debian? Not familiar with that terminology.. | 14:26 |
| viclarson | padmin@novalink:~$ viosvrcmd --id 3 -c "lsdev -field name physloc" | grep ent | 14:27 |
| viclarson | ent0 U78C9.001.WZS0FY6-P1-C7-T1 | 14:27 |
| viclarson | ent1 U78C9.001.WZS0FY6-P1-C7-T2 | 14:27 |
| viclarson | ent2 U78C9.001.WZS0FY6-P1-C7-T3 | 14:27 |
| viclarson | ent3 U78C9.001.WZS0FY6-P1-C7-T4 | 14:27 |
| viclarson | ent4 U78C9.001.WZS0FY6-P1-C12-T1 | 14:27 |
| viclarson | ent5 U78C9.001.WZS0FY6-P1-C12-T2 | 14:27 |
| viclarson | ent6 U78C9.001.WZS0FY6-P1-C12-T3 | 14:27 |
| viclarson | ent7 U78C9.001.WZS0FY6-P1-C12-T4 | 14:27 |
| viclarson | ent8 U8286.42A.2147CBW-V3-C3-T1 | 14:27 |
| viclarson | ent9 U8286.42A.2147CBW-V3-C6-T1 | 14:27 |
| viclarson | ent10 | 14:27 |
| viclarson | ent11 U8286.42A.2147CBW-V3-C7-T1 | 14:27 |
| viclarson | padmin@novalink:~$ viosvrcmd --id 2 -c "lsdev -field name physloc" | grep ent | 14:27 |
| viclarson | ent0 U78C9.001.WZS0FY6-P1-C10-T1 | 14:27 |
| viclarson | ent1 U78C9.001.WZS0FY6-P1-C10-T2 | 14:27 |
| viclarson | ent2 U78C9.001.WZS0FY6-P1-C10-T3 | 14:27 |
| viclarson | ent3 U78C9.001.WZS0FY6-P1-C10-T4 | 14:27 |
| viclarson | ent4 U78C9.001.WZS0FY6-P1-C2-T1 | 14:27 |
| viclarson | ent5 U78C9.001.WZS0FY6-P1-C2-T2 | 14:27 |
| viclarson | ent6 U78C9.001.WZS0FY6-P1-C2-T3 | 14:27 |
| viclarson | ent7 U78C9.001.WZS0FY6-P1-C2-T4 | 14:27 |
| viclarson | ent8 U8286.42A.2147CBW-V2-C3-T1 | 14:27 |
| viclarson | ent9 U8286.42A.2147CBW-V2-C6-T1 | 14:27 |
| viclarson | ent10 | 14:27 |
| viclarson | ent11 U8286.42A.2147CBW-V2-C7-T1 | 14:28 |
| viclarson | padmin@novalink:~$ pvmctl sea list | 14:28 |
| viclarson | Network Bridges | 14:28 |
| viclarson | +------+-----------+-------+----------+-------+--------+------------+ | 14:28 |
| viclarson | | PVID | VSwitch | VIOS | Back Dev | SEA | Ld Grp | Addl VLANs | | 14:28 |
| viclarson | +------+-----------+-------+----------+-------+--------+------------+ | 14:28 |
| viclarson | | 604 | ETHERNET0 | vios1 | ent6 | ent10 | ent8 | | | 14:28 |
| viclarson | | | | vios2 | ent2 | ent10 | ent8 | | | 14:28 |
| viclarson | +------+-----------+-------+----------+-------+--------+------------+ | 14:28 |
| viclarson | for any | 14:28 |
| viclarson | padmin@novalink:~$ pvmctl ioslot attach --drc-names U78C9.001.WZS0FY6-P1-C10 -p id=4 | 14:28 |
| thorst | apearson_: I guess, is there a single package someone can install to get all the necessary debians? | 14:28 |
| viclarson | [PVME010501A6-0402] I/O slot 2104001E is in use and cannot be removed. | 14:28 |
| viclarson | padmin@novalink:~$ pvmctl ioslot attach --drc-names U78C9.001.WZS0FY6-P1-C7 -p id=4 | 14:28 |
| viclarson | [PVME010501A6-0402] I/O slot 21010010 is in use and cannot be removed. | 14:28 |
| viclarson | padmin@novalink:~$ pvmctl ioslot attach --drc-names U78C9.001.WZS0FY6-P1-C6 -p id=4 | 14:28 |
| viclarson | [PVME010501A6-0402] I/O slot 21010018 is in use and cannot be removed. | 14:28 |
| viclarson | can be one port be detached? | 14:28 |
| viclarson | T1, T2, ... | 14:28 |
| thorst | viclarson: can we use pastebin for those big dumps? | 14:28 |
| viclarson | can | 14:28 |
| apearson_ | viclarson - ok, so you are using P1-C7 for your SEA (ent2 is the backing device which is P1-C7-T3). It appears then that P1-C12 is not in use... | 14:30 |
| apearson_ | thorst - rsct.core.utils I think will make sure that rsct.core is also installed | 14:31 |
| viclarson | apearson_ the same | 14:32 |
| viclarson | padmin@novalink:~$ pvmctl ioslot attach --drc-names U78C9.001.WZS0FY6-P1-C12 -p id=4 | 14:32 |
| viclarson | [PVME010501A6-0402] I/O slot 21020014 is in use and cannot be removed. | 14:32 |
| viclarson | or d be detached before? | 14:32 |
| apearson_ | the attach is trying a detach first automatically...(that's actually a bug :) ). But it means that the device is in fact configured in the OS and you need to unconfigure it. To do so... | 14:33 |
| apearson_ | viosvrcmd --id 3 -c "rmdev -dev ent4" Then repeat for ent5, ent6, and ent7 (the 4 parts on C12 | 14:33 |
| apearson_ | that tells the OS to unconfigure the ports...the detach should work after that. | 14:34 |
| openstackgerrit | Eric Fried proposed openstack/networking-powervm: [WIP] Refactor, consolidate, and clean up agents https://review.openstack.org/378000 | 14:40 |
| *** apearson_ has quit IRC | 14:43 | |
| *** apearson has joined #openstack-powervm | 14:44 | |
| *** burgerk_ has joined #openstack-powervm | 14:46 | |
| *** burgerk has quit IRC | 14:50 | |
| *** apearson has quit IRC | 15:02 | |
| *** apearson has joined #openstack-powervm | 15:03 | |
| *** dwayne has joined #openstack-powervm | 15:11 | |
| viclarson | apearson_ attached succesfully, but unfortunately it seems like card not connected to switch (NO-CARRIER) | 15:11 |
| *** tblakes has quit IRC | 15:15 | |
| efried | apearson ^^ | 15:17 |
| viclarson | jfyi nothing to do i think | 15:18 |
| *** tblakes has joined #openstack-powervm | 15:19 | |
| apearson | viclarson - bummer... So your other option would be to: 1) Tear down the redundant side of the SEA (so your vios1 would be the only one providing bridging for the ethernet). Or 2) Get that card patched in. Or 3) Try running the controller node on a separate system... | 15:19 |
| apearson | You could try the C10 card from vios1...see if that was patched...it doedsn't seem to be in use. But my guess is probably not. | 15:20 |
| viclarson | 2) 3) i dont have cables and separate systems, only first option, but i'm not ready yet | 15:26 |
| viclarson | i believe c10 is not patched | 15:27 |
| viclarson | apearson i think two ports of the same network adapter is used by vios itself. is idea of terminating second vios is good enough? | 15:36 |
| apearson | Yes - you could shutdown that second VIOS and use it's adapter. Make sure that you have redundant I/O for your running lpars though... that is, that the vscsi storage fro your lpars is flowing through vios1 (pvmctl scsi list) | 15:37 |
| viclarson | will try) | 15:40 |
| openstackgerrit | Matt Rabe proposed openstack/nova-powervm: Change get_vea_slot to get_vnet_slot to account for vNIC. https://review.openstack.org/381935 | 15:41 |
| *** shyama has quit IRC | 15:41 | |
| openstackgerrit | Matt Rabe proposed openstack/nova-powervm: Change get_vea_slot to get_vnet_slot to account for vNIC. https://review.openstack.org/381935 | 15:44 |
| *** mdrabe has quit IRC | 15:57 | |
| *** mdrabe has joined #openstack-powervm | 16:03 | |
| openstackgerrit | Matt Rabe proposed openstack/nova-powervm: Change get_vea_slot to get_vnet_slot to account for vNIC. https://review.openstack.org/381935 | 16:07 |
| *** k0da has quit IRC | 16:13 | |
| *** viclarson has quit IRC | 16:26 | |
| *** openstackgerrit has quit IRC | 16:26 | |
| *** openstackgerrit has joined #openstack-powervm | 16:27 | |
| *** openstackgerrit has quit IRC | 16:28 | |
| *** openstackgerrit has joined #openstack-powervm | 16:29 | |
| *** openstackgerrit has quit IRC | 16:30 | |
| *** openstackgerrit has joined #openstack-powervm | 16:30 | |
| *** shyama has joined #openstack-powervm | 16:31 | |
| *** mdrabe has quit IRC | 16:33 | |
| *** mdrabe has joined #openstack-powervm | 16:33 | |
| *** viclarson has joined #openstack-powervm | 16:35 | |
| *** tblakes has quit IRC | 17:12 | |
| *** seroyer has quit IRC | 17:17 | |
| *** seroyer has joined #openstack-powervm | 17:18 | |
| thorst | mdrabe: I just W+1'd that vnet slot change. You mind proposing it back to newton? | 17:52 |
| thorst | efried: https://review.openstack.org/#/c/380761/2 -> Is that function that goes into stable/newton but NOT master? | 17:53 |
| efried | Correct - see my review comment. | 17:53 |
| efried | on PS1 | 17:53 |
| efried | thorst ^^ | 17:54 |
| thorst | so, I don't think that's proper | 17:54 |
| thorst | we need it in master and backported to newton | 17:54 |
| thorst | and the other needs to be rebased on top of it. | 17:54 |
| thorst | :-/ | 17:54 |
| thorst | everything needs to go in master first unless truly independent for that branch (like specific version updates that we've done in the past) | 17:55 |
| openstackgerrit | Merged openstack/nova-powervm: Change get_vea_slot to get_vnet_slot to account for vNIC. https://review.openstack.org/381935 | 17:55 |
| efried | thorst, say wha? | 18:06 |
| efried | I mean, I hear your words. But that's crazy, yo. | 18:06 |
| openstackgerrit | Eric Fried proposed openstack/networking-powervm: [WIP] Refactor, consolidate, and clean up agents https://review.openstack.org/378000 | 18:07 |
| *** viclarson has quit IRC | 18:07 | |
| openstackgerrit | Eric Fried proposed openstack/networking-powervm: Time out port update waiting for vif plug https://review.openstack.org/382026 | 18:09 |
| efried | But okay, crazy is my middle name. ^^ thorst | 18:10 |
| efried | gonna make 378000 a mess. | 18:10 |
| *** tblakes_ has joined #openstack-powervm | 18:11 | |
| *** tblakes__ has joined #openstack-powervm | 18:11 | |
| *** tblakes__ is now known as tblakes | 18:15 | |
| *** tblakes_ has quit IRC | 18:16 | |
| mdrabe | efried, thorst: cherry-picking the get_vnet_slot change back to stable/newton conflicts cuz the VIF plug/unplug UT never made in stable/newton: https://github.com/openstack/nova-powervm/commit/6452f460aa4e688cf49eb7e933ee3bc8a5fc5a9b. Should that UT be backported? | 18:18 |
| efried | Not really bug-worthy. | 18:18 |
| efried | But you're not the first to run into that merge conflict. | 18:18 |
| efried | Was kinda hoping it would be a one-and-done. | 18:18 |
| mdrabe | Yea... it's a weird case | 18:19 |
| efried | But since it appears it's going to continue being a PITA, might as well. | 18:19 |
| efried | You gonna do that? | 18:19 |
| mdrabe | Yea I can do that | 18:19 |
| mdrabe | Here's a question thogh | 18:19 |
| efried | Thanks. Need a bug and all. | 18:19 |
| efried | Actually, yeah, you might as well fold it into your current cherry-pick | 18:19 |
| mdrabe | though*, should it be backported per the original commit or per my cherry pick? | 18:19 |
| mdrabe | ok | 18:19 |
| efried | up to you. | 18:19 |
| efried | Oh, wait, the process police will probably have a fit. | 18:19 |
| efried | (thorst) | 18:20 |
| efried | So better do it by the book. | 18:20 |
| efried | And in order. | 18:20 |
| mdrabe | Well the problem with backporting per the original commit is, then a bug report would be needed | 18:21 |
| mdrabe | Unless I associate the backport with the bug I opened... | 18:22 |
| efried | mdrabe, Sure, open a bug. "Cherry-picks and rebases are a pain in the ass." | 18:34 |
| mdrabe | Sure I'll mark it perpetually opened | 18:35 |
| mdrabe | efried: I just backported the UT with the cherry-pick | 18:36 |
| mdrabe | https://review.openstack.org/#/c/382034/ | 18:36 |
| efried | mdrabe, You're either going to have to rebase it on top of https://review.openstack.org/#/c/381836/ or rip out test_plug_raises (and then have another merge conflict when that guy goes in). | 18:47 |
| efried | mdrabe, oh, there's a change in that method - so yeah, rebase it on top of https://review.openstack.org/#/c/381836/ | 18:49 |
| efried | You can do that from gerrit, probably. | 18:49 |
| efried | Punch rebase, check the box that says change parent, and enter 381836 | 18:49 |
| mdrabe | efried: yeap somebody's gonna have to rebase | 18:49 |
| mdrabe | your's isn't merged yet? | 18:49 |
| efried | Not the newton one. | 18:50 |
| mdrabe | My master one is merged | 18:50 |
| mdrabe | So I think you'll have to rebase your master one | 18:50 |
| efried | No, my master one was merged before you got started, so you were already okay there. | 18:50 |
| mdrabe | Oh, maybe there's another review it was in conflict with | 18:51 |
| mdrabe | Could swear it was in conflict with something. Anyway for stable/newton one of us should get merged first, preferrably yours | 18:51 |
| efried | Your newton one is in conflict with my newton one. | 18:51 |
| efried | I'm agreeing with that, but you don't have to wait to rebase yours on mine. You can do it from gerrit. | 18:52 |
| efried | Punch rebase, check the box that says change parent, and enter 381836 | 18:52 |
| efried | Although I guess that'll probably give you a merge conflict. | 18:52 |
| mdrabe | Yea it's a merge conflict | 18:53 |
| mdrabe | I'd have to workspace it | 18:53 |
| mdrabe | Best to wait for merge goshforbid you push up a new commit for that review | 18:53 |
| mdrabe | namsayin' | 18:54 |
| *** thorst has quit IRC | 18:54 | |
| *** thorst has joined #openstack-powervm | 18:55 | |
| openstackgerrit | Eric Fried proposed openstack/nova-powervm: Send custom vif plug/unplug event https://review.openstack.org/373532 | 18:55 |
| thorst | lol | 18:56 |
| efried | laugh it up thorst ball. This is all your fault anyway. | 18:57 |
| efried | Prefer sitting in meetings to doing our code reviews. | 18:57 |
| thorst | o, preference isn't it. | 18:58 |
| thorst | I want to cooooode | 18:58 |
| thorst | speaking of which... | 18:58 |
| thorst | seroyer: is it time for me to do another live migration? | 18:58 |
| thorst | I have 4 VMs ready | 18:58 |
| seroyer | thorst, Yes! | 18:58 |
| thorst | eek...ok | 18:59 |
| seroyer | Really ready to get this work item off of my queue. :-) | 18:59 |
| thorst | you and me both... | 19:01 |
| thorst | except I'll have a pile of UT to fix | 19:01 |
| seroyer | This is my UT. :-D | 19:01 |
| thorst | seroyer: I don't think you understand the meaning of UT | 19:01 |
| thorst | ;-) | 19:01 |
| seroyer | Not having Mock as an option makes it hard. And LPM is super non-trivial to decompose into small tests. | 19:03 |
| thorst | true dat | 19:03 |
| thorst | seroyer: its kicked off | 19:08 |
| thorst | have the IPs and what not? | 19:08 |
| efried | I used to love sitting in a presentation by Java developers telling me I should do test-driven development on the AIX kernel. | 19:09 |
| seroyer | Did test driven development in hypervisor. Worked really well for my part of the hypervisor. Fully automated. Harder to do when you need to touch hardware though. | 19:10 |
| thorst | seroyer: done | 19:11 |
| openstackgerrit | Eric Fried proposed openstack/networking-powervm: [WIP] Refactor, consolidate, and clean up agents https://review.openstack.org/378000 | 19:19 |
| *** k0da has joined #openstack-powervm | 19:22 | |
| *** shyama has quit IRC | 19:35 | |
| openstackgerrit | Merged openstack/nova-powervm: Send custom vif plug/unplug event https://review.openstack.org/373532 | 19:51 |
| *** miltonm has quit IRC | 20:40 | |
| *** miltonm has joined #openstack-powervm | 20:56 | |
| *** thorst has quit IRC | 21:10 | |
| openstackgerrit | Eric Fried proposed openstack/networking-powervm: [WIP] Refactor, consolidate, and clean up agents https://review.openstack.org/378000 | 21:19 |
| *** thorst has joined #openstack-powervm | 21:19 | |
| *** thorst has quit IRC | 21:24 | |
| *** kylek3h has quit IRC | 21:29 | |
| *** burgerk__ has joined #openstack-powervm | 21:29 | |
| *** kylek3h has joined #openstack-powervm | 21:29 | |
| *** burgerk_ has quit IRC | 21:32 | |
| *** burgerk__ has quit IRC | 21:33 | |
| *** kylek3h has quit IRC | 21:34 | |
| *** jwcroppe has quit IRC | 21:40 | |
| *** jwcroppe has joined #openstack-powervm | 21:40 | |
| *** jwcroppe has quit IRC | 21:40 | |
| *** thorst has joined #openstack-powervm | 21:55 | |
| *** thorst has quit IRC | 22:00 | |
| *** mdrabe has quit IRC | 22:07 | |
| *** edmondsw has quit IRC | 22:10 | |
| openstackgerrit | Eric Fried proposed openstack/networking-powervm: Refactor, consolidate, and clean up agents https://review.openstack.org/378000 | 22:13 |
| *** apearson has quit IRC | 22:14 | |
| *** k0da has quit IRC | 22:15 | |
| *** tblakes has quit IRC | 22:17 | |
| *** tjakobs has quit IRC | 22:28 | |
| *** thorst has joined #openstack-powervm | 23:49 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!