*** Sukhdev has quit IRC | 00:03 | |
*** Sukhdev has joined #openstack-ironic | 00:03 | |
*** Sukhdev has quit IRC | 00:08 | |
*** mtanino has quit IRC | 00:30 | |
*** naohirot has joined #openstack-ironic | 00:33 | |
*** ijw has joined #openstack-ironic | 00:36 | |
*** r-daneel_ has joined #openstack-ironic | 00:45 | |
*** r-daneel has quit IRC | 00:45 | |
*** saripurigopi has joined #openstack-ironic | 00:48 | |
*** smoriya has joined #openstack-ironic | 00:49 | |
*** saripurigopi has quit IRC | 00:55 | |
*** saripurigopi has joined #openstack-ironic | 01:02 | |
*** mitchjameson has quit IRC | 01:08 | |
*** r-daneel_ has quit IRC | 01:46 | |
*** achanda has joined #openstack-ironic | 01:46 | |
*** mitchjameson has joined #openstack-ironic | 01:53 | |
*** yonglihe has quit IRC | 01:59 | |
*** yonglihe has joined #openstack-ironic | 01:59 | |
*** achanda has quit IRC | 02:02 | |
*** Marga_ has quit IRC | 02:21 | |
*** saripurigopi has quit IRC | 02:24 | |
*** rloo has quit IRC | 02:27 | |
*** ramineni has joined #openstack-ironic | 02:38 | |
*** Marga_ has joined #openstack-ironic | 02:39 | |
*** Marga_ has quit IRC | 02:43 | |
*** Marga_ has joined #openstack-ironic | 02:43 | |
*** Marga_ has quit IRC | 02:45 | |
*** Marga_ has joined #openstack-ironic | 02:45 | |
*** Marga_ has quit IRC | 02:46 | |
*** Marga_ has joined #openstack-ironic | 02:50 | |
*** Marga_ has quit IRC | 02:52 | |
*** yonglihe has quit IRC | 02:59 | |
openstackgerrit | Yongli He proposed openstack/bifrost: local install with -K if needed https://review.openstack.org/187841 | 03:08 |
---|---|---|
*** yonglihe has joined #openstack-ironic | 03:11 | |
*** spandhe has quit IRC | 03:15 | |
openstackgerrit | Tan Lin proposed openstack/ironic: Lower capability value https://review.openstack.org/190900 | 03:30 |
openstackgerrit | Merged openstack/ironic: remove unneeded sqlalchemy-migrate requirement https://review.openstack.org/190696 | 03:35 |
*** dguerri` is now known as dguerri | 03:44 | |
*** dguerri is now known as dguerri` | 03:54 | |
*** saripurigopi has joined #openstack-ironic | 03:55 | |
*** mitchjameson has quit IRC | 03:58 | |
*** yog_ has joined #openstack-ironic | 04:01 | |
*** Marga_ has joined #openstack-ironic | 04:09 | |
*** Marga_ has quit IRC | 04:09 | |
*** Marga_ has joined #openstack-ironic | 04:10 | |
*** Sukhdev has joined #openstack-ironic | 04:19 | |
*** yuikotakada has joined #openstack-ironic | 04:22 | |
yuikotakada | TGIF, Ironic | 04:22 |
openstackgerrit | Anusha Ramineni proposed openstack/ironic-specs: Add Zapping support to iLO drivers https://review.openstack.org/145404 | 04:23 |
*** Marga_ has quit IRC | 04:25 | |
openstackgerrit | Gopi Krishna S proposed openstack/ironic: Address follow-up comments on ucs drivers https://review.openstack.org/190913 | 04:29 |
*** rameshg87 has joined #openstack-ironic | 04:34 | |
rameshg87 | good morning ironic | 04:34 |
yuikotakada | rameshg87: good morning | 04:34 |
rameshg87 | yuikotakada: o/ | 04:35 |
saripurigopi | good morning rameshg87,yuikotakada | 04:41 |
yuikotakada | saripurigopi: g'morning | 04:41 |
rameshg87 | saripurigopi: o/ | 04:44 |
*** zz_natorious is now known as natorious | 04:52 | |
openstackgerrit | Merged openstack/ironic: Add additional logging around cleaning https://review.openstack.org/190728 | 04:53 |
*** chlong is now known as chlong_mtg | 04:59 | |
openstackgerrit | Tan Lin proposed openstack/ironic: Set the capability value to lowercase https://review.openstack.org/190900 | 04:59 |
*** natorious is now known as zz_natorious | 05:00 | |
*** bkero has quit IRC | 05:10 | |
*** bkero has joined #openstack-ironic | 05:10 | |
*** smoriya has quit IRC | 05:28 | |
*** gridinv_ has quit IRC | 05:31 | |
*** mitchjameson has joined #openstack-ironic | 05:43 | |
*** zz_naotok has quit IRC | 05:54 | |
*** yuikotakada has quit IRC | 05:56 | |
*** vishwana_ has quit IRC | 05:58 | |
*** vishwanathj has joined #openstack-ironic | 06:00 | |
chlong_mtg | ping | 06:04 |
chlong_mtg | Can I ask an ironic question here? | 06:04 |
*** yuikotakada has joined #openstack-ironic | 06:04 | |
*** radek_ has joined #openstack-ironic | 06:05 | |
chlong_mtg | ironic question: I'm having problems registering newly created VM's in ironic cli for use with my overcloud to scale out. Any ideas? I'm following the instructions here: https://www.rdoproject.org/TripleO-CLI#Post-Deployment | 06:06 |
chlong_mtg | I'm always receiving this error: InvalidParameterValue: SSH connection cannot be established: Failed to establish SSH connection to host 192.168.122.1 | 06:06 |
*** chlong_mtg is now known as chlong | 06:06 | |
*** amotoki has joined #openstack-ironic | 06:16 | |
*** Sukhdev has quit IRC | 06:19 | |
*** Haomeng has joined #openstack-ironic | 06:27 | |
openstackgerrit | Tan Lin proposed openstack/ironic: Remove unnecessary validation in PXE https://review.openstack.org/190952 | 06:38 |
*** lazy_prince has joined #openstack-ironic | 06:41 | |
*** Guest____ has joined #openstack-ironic | 06:45 | |
*** amotoki has quit IRC | 06:47 | |
lazy_prince | o/ | 06:48 |
*** amotoki has joined #openstack-ironic | 06:50 | |
*** Nisha has joined #openstack-ironic | 06:50 | |
*** Guest____ is now known as jcoufal | 06:51 | |
*** jcoufal is now known as Guest76410 | 06:51 | |
*** Guest76410 has quit IRC | 06:53 | |
*** jcoufal_ has joined #openstack-ironic | 06:53 | |
*** mitchjameson has quit IRC | 06:54 | |
*** dtantsur|afk is now known as dtantsur | 06:54 | |
dtantsur | morning Ironic | 06:54 |
dtantsur | TGIF! | 06:54 |
yuikotakada | lazy_prince: o/ | 06:55 |
yuikotakada | dtantsur: TGIF! | 06:55 |
dtantsur | :) | 06:55 |
lazy_prince | :) | 06:55 |
*** romainh has joined #openstack-ironic | 06:56 | |
rameshg87 | chlong: may be you should check with dtantsur about rdo thing. he might be able to point out whom to check with. | 06:57 |
rameshg87 | dtantsur: lazy_prince: morning o/ | 06:57 |
chlong | rameshg87, OK | 06:57 |
dtantsur | rameshg87, o/ | 06:57 |
chlong | dtantsur, ping | 06:58 |
dtantsur | chlong, hi! | 06:58 |
chlong | ironic question: I'm having problems registering newly created VM's in ironic cli for use with my overcloud to scale out. Any ideas? I'm following the instructions here: https://www.rdoproject.org/TripleO-CLI#Post-Deployment | 06:58 |
chlong | dtantsur, who should I be speaking to about this ? | 06:58 |
dtantsur | chlong, let's continue on #rdo, I probably can help you | 06:59 |
*** derekh has joined #openstack-ironic | 06:59 | |
*** ijw has quit IRC | 07:03 | |
yuikotakada | dtantsur: IMO, we Ironic-inspector should set feature freeze date same as other projects in order to recognize schedule. Currently there is just rough release date, right? WDYT? | 07:04 |
openstackgerrit | Ramakrishnan G proposed openstack/ironic: IPA: Do a soft power off at the end of deployment https://review.openstack.org/185667 | 07:04 |
dtantsur | yuikotakada, for 2.1.0? depending on what you mean by "other projects" :) everyone is doing their own thing now, e.g. Ironic is likely to have shorted FF | 07:05 |
yuikotakada | dtantsur, yeah, for v2.1.0 | 07:05 |
yuikotakada | "other projects" means like nova or ironic and so on | 07:05 |
dtantsur | yuikotakada, we can just follow Ironic, as it's the closest "other project" for us :) nova might have a different schedule | 07:06 |
*** naohirot has quit IRC | 07:06 | |
yuikotakada | oh | 07:06 |
yuikotakada | are we following to ironic? That's right, I forgot it :p | 07:07 |
dtantsur | I guess we do. Let's see what we agree on for Ironic actually (if we actually agree on something) :D | 07:07 |
yuikotakada | :D | 07:09 |
*** athomas has joined #openstack-ironic | 07:11 | |
*** Haomeng|2 has joined #openstack-ironic | 07:14 | |
*** Haomeng has quit IRC | 07:16 | |
openstackgerrit | Ramakrishnan G proposed openstack/ironic: Add boot interface in Ironic https://review.openstack.org/166512 | 07:21 |
*** rameshg87 is now known as rameshg87-lunch | 07:22 | |
*** ifarkas has joined #openstack-ironic | 07:28 | |
*** Nisha has quit IRC | 07:51 | |
*** dtantsur is now known as dtantsur|bbl | 07:51 | |
*** dlpartain has joined #openstack-ironic | 07:52 | |
openstackgerrit | Merged stackforge/proliantutils: Add iSCSI boot support in RIS https://review.openstack.org/187162 | 07:56 |
*** ijw has joined #openstack-ironic | 08:04 | |
*** Nisha has joined #openstack-ironic | 08:05 | |
openstackgerrit | Ramakrishnan G proposed openstack/ironic: IPA: Do a soft power off at the end of deployment https://review.openstack.org/185667 | 08:07 |
*** ijw_ has joined #openstack-ironic | 08:08 | |
*** dlpartain has left #openstack-ironic | 08:10 | |
*** ijw has quit IRC | 08:10 | |
*** rameshg87-lunch is now known as rameshg87 | 08:11 | |
*** ijw_ has quit IRC | 08:13 | |
*** erelwell has joined #openstack-ironic | 08:13 | |
*** chlong is now known as chlong_afk | 08:16 | |
*** lucasagomes has joined #openstack-ironic | 08:16 | |
*** yuikotakada has quit IRC | 08:17 | |
openstackgerrit | Merged openstack/ironic: Remove unnecessary validation in PXE https://review.openstack.org/190952 | 08:18 |
*** subscope has joined #openstack-ironic | 08:20 | |
*** jistr has joined #openstack-ironic | 08:20 | |
*** yuikotakada has joined #openstack-ironic | 08:21 | |
*** stendulker has joined #openstack-ironic | 08:27 | |
*** ndipanov has joined #openstack-ironic | 08:31 | |
*** dmitchell has joined #openstack-ironic | 08:35 | |
*** e0ne has joined #openstack-ironic | 08:38 | |
*** MattMan has quit IRC | 08:41 | |
openstackgerrit | Gopi Krishna S proposed openstack/ironic: Address follow-up comments on ucs drivers https://review.openstack.org/190913 | 08:48 |
*** NobodyCam has quit IRC | 08:49 | |
*** NobodyCam has joined #openstack-ironic | 08:50 | |
*** lucasagomes has quit IRC | 08:52 | |
*** ifarkas has quit IRC | 08:52 | |
*** ifarkas has joined #openstack-ironic | 08:53 | |
*** dmitchell__ has joined #openstack-ironic | 08:54 | |
*** lucasagomes has joined #openstack-ironic | 08:56 | |
*** e0ne has quit IRC | 08:57 | |
*** dmitchell has quit IRC | 08:58 | |
openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Wake-On-Lan Power interface https://review.openstack.org/179078 | 09:02 |
*** chlong_afk has quit IRC | 09:03 | |
*** ndipanov has quit IRC | 09:04 | |
*** dguerri` is now known as dguerri | 09:05 | |
*** ijw has joined #openstack-ironic | 09:08 | |
*** pelix has joined #openstack-ironic | 09:09 | |
*** ijw has quit IRC | 09:13 | |
stendulker | Hi | 09:14 |
stendulker | For dhclient request, dnsmasq dhcp is responding with the IP, but that IP is not getting set to the interface. | 09:16 |
stendulker | Node perpetually requests for the IP and dnsmasq pertually send the IP address to it. Without the node setting the received IP. | 09:17 |
stendulker | Any idea what could be the issue here? | 09:17 |
stendulker | I face this issue with CoreOS and openSUSE linux | 09:18 |
openstackgerrit | Ramakrishnan G proposed openstack/ironic: Refactor method to add or update capability string https://review.openstack.org/162451 | 09:18 |
stendulker | user images of CoreOS and openSUSE linux | 09:18 |
openstackgerrit | Gopi Krishna S proposed openstack/ironic: Address follow-up comments on ucs drivers https://review.openstack.org/190913 | 09:20 |
*** Marga_ has joined #openstack-ironic | 09:34 | |
openstackgerrit | Nisha Agarwal proposed openstack/ironic-specs: Capabilities should accept values as dictionary https://review.openstack.org/182934 | 09:37 |
*** ifarkas has quit IRC | 09:38 | |
*** ifarkas has joined #openstack-ironic | 09:39 | |
openstackgerrit | Gopi Krishna S proposed openstack/ironic: Address follow-up comments on ucs drivers https://review.openstack.org/190913 | 09:43 |
*** ndipanov has joined #openstack-ironic | 09:44 | |
*** mgoddard1 has joined #openstack-ironic | 09:46 | |
*** mgoddard has quit IRC | 09:47 | |
*** ndipanov has quit IRC | 09:49 | |
*** ndipanov has joined #openstack-ironic | 09:49 | |
*** Nisha has quit IRC | 09:50 | |
*** romainh has quit IRC | 09:50 | |
*** mgoddard1 has quit IRC | 09:50 | |
rameshg87 | lucasagomes: hi | 09:51 |
lucasagomes | rameshg87, hello there! | 09:51 |
rameshg87 | lucasagomes: a quick question | 09:51 |
*** toabctl has quit IRC | 09:52 | |
rameshg87 | lucasagomes: I think it's better if we make the raid stuff not to wait for zapping | 09:52 |
rameshg87 | lucasagomes: because we have seen that raid can be used in cleaning as well | 09:52 |
*** toabctl has joined #openstack-ironic | 09:52 | |
rameshg87 | lucasagomes: since zapping/cleaning is just a difference of priority values, can we get the raid stuff merged before zapping ? | 09:53 |
rameshg87 | lucasagomes: raid generic stuff has now 2x+2s and inband raid had 4 +2s (in previous patch set) | 09:53 |
rameshg87 | wdyt ? | 09:53 |
lucasagomes | rameshg87, /me thinks | 09:57 |
lazy_prince | stendulker: did you check the dhclient log files..? | 09:57 |
lucasagomes | rameshg87, so how one triggers raid? Via cleaning? Or passthru? | 09:58 |
rameshg87 | lucasagomes: through zapping or cleaning | 09:58 |
rameshg87 | lucasagomes: the create_configuration will be decorated and the priority value will be configurable | 09:59 |
rameshg87 | lucasagomes: if it's non-zero it can be done during cleaning | 09:59 |
rameshg87 | lucasagomes: so that means it will work with the existing framework itself | 09:59 |
lucasagomes | rameshg87, ack yeah sounds good | 10:01 |
lucasagomes | if that was the decision to have on zapping and/or cleaning | 10:01 |
lucasagomes | it seems fine | 10:01 |
rameshg87 | yeah | 10:01 |
* rameshg87 goes and removes gerrit dependency of raid spec | 10:01 | |
rameshg87 | lucasagomes: have mentioned it in spec | 10:02 |
rameshg87 | https://review.openstack.org/#/c/173214/11/specs/liberty/ironic-generic-raid-interface.rst L17-24 | 10:02 |
* lucasagomes reviews it | 10:03 | |
*** e0ne has joined #openstack-ironic | 10:04 | |
*** e0ne is now known as e0ne_ | 10:04 | |
openstackgerrit | Ramakrishnan G proposed openstack/ironic-specs: Add generic RAID configuration spec for liberty https://review.openstack.org/173214 | 10:05 |
stendulker | lazy_prince: what is path of dhclient logs | 10:05 |
*** e0ne_ is now known as e0ne | 10:06 | |
openstackgerrit | Ramakrishnan G proposed openstack/ironic-specs: Add inband RAID configuration spec for liberty https://review.openstack.org/173218 | 10:06 |
stendulker | lazy_prince: Did not nee any thing under /var/log | 10:06 |
*** ijw has joined #openstack-ironic | 10:06 | |
lazy_prince | it depends on distro but it should be in /var/log/dhcpcd | 10:06 |
lazy_prince | you could also try running dhclient in verbose mode to see if it reports any error.. | 10:07 |
*** romainh has joined #openstack-ironic | 10:10 | |
*** alexpilotti has joined #openstack-ironic | 10:10 | |
stendulker | lazy_prince: Running on opensuse. Did not find dhclient binary on the node. Probably n/w service is issuing the dhclient request | 10:11 |
*** ijw has quit IRC | 10:12 | |
lazy_prince | stendulker: can you see if you have dhcpcd or something similar..? | 10:13 |
lazy_prince | on SLES, I could see there is dhcpcd... | 10:13 |
lucasagomes | rameshg87, question... so it was made possible to run on cleaning so we could have software raid? | 10:15 |
lucasagomes | or that spec is specific for hardware raid? | 10:15 |
stendulker | lazy_prince: dhcpd binary also do not exist here | 10:15 |
rameshg87 | lucasagomes: JayF first pointed out that it can be run on cleaning with software raid | 10:16 |
lazy_prince | dhcpd or dhcpcd..? | 10:16 |
lucasagomes | rameshg87, yeah... it indeed seems possible | 10:16 |
rameshg87 | lucasagomes: but then we generally came to an agreement that it can left out as operator's choice to make it available as cleaning or zapping | 10:16 |
lucasagomes | anyway... just a question, it's out of the scope of the spec anyway | 10:16 |
lucasagomes | rameshg87, seems sane | 10:16 |
lucasagomes | yeah | 10:16 |
lazy_prince | stendulker: dhcpd or dhcpcd..? | 10:17 |
rameshg87 | lucasagomes: if someone wants to make sure the node is set everytime with same raid configuration before it's available | 10:17 |
rameshg87 | they could make it as part of cleaning | 10:17 |
rameshg87 | in hp hardware at least, the tenant can modify raid configuration with inband utilities | 10:17 |
lazy_prince | stendulker: also check if the log files exists in /var/lib.. | 10:18 |
lucasagomes | rameshg87, I see | 10:18 |
lucasagomes | rameshg87, +a | 10:19 |
lucasagomes | thanks happy hacking | 10:19 |
stendulker | lazy_prince: no dhcpd or dhcpcd | 10:19 |
rameshg87 | \o/ | 10:20 |
rameshg87 | thanks :) | 10:20 |
stendulker | no releant logs in /var/lib as well | 10:20 |
stendulker | lazy_prince: all i could see on server side the mac address beig given IP but no ack from the node for that IP | 10:20 |
lazy_prince | stendulker: aha.. so then which utility is requesting IP address.. | 10:20 |
*** mgoddard has joined #openstack-ironic | 10:21 | |
stendulker | lazy_prince: 03:05:26.652789 IP 0.0.0.0.bootpc > 255.255.255.255.bootps: BOOTP/DHCP, Request from 9c:b6:54:af:93:3c (oui Unknown), length 312 | 10:21 |
stendulker | 03:05:26.653016 IP 10.10.1.197.bootps > 10.10.1.190.bootpc: BOOTP/DHCP, Reply, length 326 | 10:21 |
rameshg87 | lucasagomes: and there is an inband one too. please have a look at it when you get some time | 10:21 |
saripurigopi | I'm installing devstack on ubuntu14.1 , it is failing with 2015-07-23 09:39:56.624 | ++ die 251 'PXE boot file /usr/lib/syslinux/pxelinux.0 not found.' | 10:22 |
rameshg87 | lucasagomes: I guess you are last spec core to not +2 it yet :D | 10:22 |
*** athomas has quit IRC | 10:23 | |
lucasagomes | rameshg87, ack | 10:25 |
lazy_prince | stendulker: here.. | 10:25 |
lazy_prince | for me the tcpdump shows as follows: | 10:25 |
lazy_prince | 06:27:43.576668 IP 0.0.0.0.bootpc > 255.255.255.255.bootps: BOOTP/DHCP, Request from 00:50:56:98:1b:d2 (oui Unknown), length 548 | 10:25 |
lazy_prince | 06:27:43.577277 IP 192.168.0.51.bootps > 255.255.255.255.bootpc: BOOTP/DHCP, Reply, length 344 | 10:25 |
lazy_prince | in your case the bootpc 2nd line has a ip address.. however in my case it 255.255.255.255 | 10:26 |
stendulker | lazy_prince: in your case it looks like broadcasting | 10:26 |
*** romainh has quit IRC | 10:27 | |
lazy_prince | thats how it should be as the client still does not have an ip address | 10:27 |
lazy_prince | once IP address is received, then it should only be renewal | 10:28 |
stendulker | stendulker: yes, but for ubuntu deploy ramdisk it does get IP through DHCP , though I have not captured tcpdump of tht | 10:29 |
*** athomas has joined #openstack-ironic | 10:29 | |
stendulker | lazy_prince: Your thing makes sense. But its dnsmasq thing, tht should take care of the protocol.. | 10:29 |
lazy_prince | stendulker: these lines that i have posted are for ubuntu deploy images.. :) | 10:30 |
stendulker | lazy_prince: ok :) | 10:30 |
openstackgerrit | Yuiko Takada proposed stackforge/ironic-discoverd: Use keystonemiddleware auth credentials https://review.openstack.org/170782 | 10:30 |
stendulker | lazy_prince: this is suse whole disk image | 10:30 |
stendulker | lazy_pricne: being deployed on to the BM | 10:30 |
lazy_prince | stendulker: are you sure, your system has to get 10.10.1.190 IP address..? it could also be that some other is renewing the lease.. just a thought.. | 10:31 |
openstackgerrit | Merged openstack/ironic-specs: Add generic RAID configuration spec for liberty https://review.openstack.org/173214 | 10:31 |
*** yuikotakada has quit IRC | 10:32 | |
lazy_prince | the other thing to check is if the image has driver for your NIC and that someone else is not using the same IP address for something else.. | 10:32 |
stendulker | lazy_prince: That part I had confirmed tht no other dhcp is responding to this mac | 10:32 |
rameshg87 | ifarkas: ^^^^^ | 10:33 |
stendulker | lazy_prince: I will try to deploy again and chk what tcpdump shows for ubuntu deploy ramdisk | 10:33 |
*** subscope has quit IRC | 10:33 | |
*** yuanying has quit IRC | 10:34 | |
*** rameshg87 is now known as rameshg87-away | 10:34 | |
ifarkas | rameshg87-away, thanks, will review it | 10:41 |
*** subscope has joined #openstack-ironic | 10:50 | |
devananda | http://www.uefi.org/sites/default/files/resources/UEFI_Plugfest_May_2015_HTTP_Boot_Redfish_Samer_El-Haj_ver1.2.pdf | 10:50 |
devananda | also, good morning | 10:50 |
stendulker | lazy_prince: For ubuntu deploy ramdisk also I get similar kind of entries. | 10:51 |
stendulker | lazy_prince: 03:56:16.455182 IP 0.0.0.0.bootpc > 255.255.255.255.bootps: BOOTP/DHCP, Request from 9c:b6:54:af:93:3c (oui Unknown), length 300 | 10:51 |
stendulker | lazy_prince: 03:56:16.456255 IP 10.10.1.197.bootps > 10.10.1.194.bootpc: BOOTP/DHCP, Reply, length 326 | 10:51 |
*** ramineni has quit IRC | 10:52 | |
stendulker | lazy_prince: But ubuntu does accept 10.10.1.194 as its IP and moves forward with deploy | 10:52 |
* lazy_prince out of ideas now.. | 10:53 | |
stendulker | lazy_prince: thank you for your help :) | 10:54 |
lazy_prince | stendulker: I was not able to help.. but anyways, i will accept your ty.. :P | 10:55 |
*** rameshg87-away is now known as rameshg87 | 10:56 | |
rameshg87 | ifarkas: it just got merged (again) :) | 10:56 |
*** dtantsur|bbl is now known as dtantsur | 10:57 | |
ifarkas | rameshg87, hah, I misread it | 11:00 |
ifarkas | rameshg87, but that is excellent news!!! | 11:00 |
rameshg87 | :) | 11:01 |
*** stendulker has quit IRC | 11:02 | |
*** ijw has joined #openstack-ironic | 11:06 | |
*** romainh has joined #openstack-ironic | 11:09 | |
*** ijw has quit IRC | 11:12 | |
yuriyz | Hi Ironic | 11:22 |
*** yog_ has quit IRC | 11:22 | |
*** saripurigopi has quit IRC | 11:22 | |
yuriyz | lucasagomes please look at https://review.openstack.org/#/c/174785/14/ironic/drivers/modules/deploy_utils.py , validation for capabilities in lowercase | 11:23 |
lucasagomes | yuriyz, good afternoon | 11:23 |
lucasagomes | yuriyz, I will do in a bit, I gotta go out collect something in the POST and grab some lunch | 11:23 |
lucasagomes | will review it right after I get back | 11:24 |
*** lucasagomes is now known as lucas-hungry | 11:24 | |
*** ndipanov has quit IRC | 11:33 | |
dtantsur | morning, devananda, yuriyz, and everyone I missed during my language lessons :) | 11:33 |
openstackgerrit | Merged openstack/ironic-lib: Updated from global requirements https://review.openstack.org/189900 | 11:34 |
dtantsur | any specs-core up for reviewing the ENROLL spec https://review.openstack.org/#/c/179151 please? | 11:34 |
openstackgerrit | Merged openstack/bifrost: Correct CSV processing in inventory https://review.openstack.org/190781 | 11:35 |
*** saripurigopi has joined #openstack-ironic | 11:37 | |
*** e0ne is now known as e0ne_ | 11:38 | |
openstackgerrit | Gopi Krishna S proposed openstack/ironic: Address follow-up comments on ucs drivers https://review.openstack.org/190913 | 11:39 |
*** krtaylor has joined #openstack-ironic | 11:42 | |
*** saripurigopi has quit IRC | 11:42 | |
openstackgerrit | Merged openstack/bifrost: Change pip install method and add six to requirements https://review.openstack.org/190768 | 11:44 |
openstackgerrit | Merged stackforge/ironic-discoverd: Use keystonemiddleware auth credentials https://review.openstack.org/170782 | 11:48 |
*** ndipanov has joined #openstack-ironic | 11:48 | |
*** e0ne_ has quit IRC | 11:49 | |
rameshg87 | dtantsur: hi, do you want someone else also to take a look at https://review.openstack.org/#/c/189241/ | 11:57 |
rameshg87 | 2 x +2s already :) | 11:58 |
openstackgerrit | Ramakrishnan G proposed openstack/ironic: Add driver interface for RAID configuration https://review.openstack.org/155230 | 11:58 |
openstackgerrit | Ramakrishnan G proposed openstack/ironic: Refactor method to add or update capability string https://review.openstack.org/162451 | 11:58 |
dtantsur | rameshg87, well, I don't known IPA really well... | 11:58 |
*** zhenguo has quit IRC | 11:58 | |
rameshg87 | dtantsur: okay. will wait for someone's to have a look at it. np. | 11:59 |
sambetts | morning dtantsur | 12:02 |
dtantsur | sambetts, morning! | 12:02 |
sambetts | UGT morning ironic :) | 12:02 |
openstackgerrit | John Trowbridge proposed stackforge/ironic-discoverd: Rename edeploy plugin https://review.openstack.org/189944 | 12:02 |
openstackgerrit | John Trowbridge proposed stackforge/ironic-discoverd: Fix edeploy plugin puts too much data in Ironic extra column https://review.openstack.org/189427 | 12:02 |
sambetts | (it totally afternoon for me right now) | 12:02 |
rameshg87 | sambetts: o/ | 12:03 |
sambetts | o/ rameshg87 | 12:03 |
rameshg87 | sambetts: that's what ugt was for right ? :) | 12:03 |
sambetts | rameshg87: haha yup | 12:03 |
rameshg87 | it's morning whenever we begin work :) | 12:03 |
dtantsur | ++ once I had morning at 5pm | 12:04 |
dtantsur | (after the summit, maybe you remember_ | 12:04 |
*** tiagogomes_ has joined #openstack-ironic | 12:05 | |
TheJulia | good morning everyone | 12:06 |
dtantsur | TheJulia, morning | 12:06 |
sambetts | Morning TheJulia | 12:06 |
*** ijw has joined #openstack-ironic | 12:06 | |
*** dprince has joined #openstack-ironic | 12:06 | |
rameshg87 | dtantsur: :) | 12:07 |
* rameshg87 goes home | 12:07 | |
*** rameshg87 has quit IRC | 12:08 | |
*** ndipanov has quit IRC | 12:09 | |
tiagogomes_ | Hi, I am getting an "No valid host was found. There are not enough hosts available" error when trying to provision with Nova. My baremetal flavor is http://paste.openstack.org/show/285134/ , and the hypervisor details are this http://paste.openstack.org/show/285135/ . Anyone has any clue about this? | 12:10 |
dtantsur | tiagogomes_, hey, here's a guide I once wrote on this error: https://repos.fedorapeople.org/repos/openstack-m/docs/master/troubleshooting/troubleshooting-overcloud.html#no-valid-host-found-error | 12:11 |
*** ijw has quit IRC | 12:12 | |
openstackgerrit | Sam Betts proposed stackforge/ironic-discoverd: Provide hook into process when a node isn't found https://review.openstack.org/190124 | 12:14 |
sambetts | dtantsur: | 12:14 |
sambetts | dtantsur: ^ I moved the find node info stuff out like we discussed, which got rid of the complexity issue :) | 12:15 |
dtantsur | cool! | 12:15 |
tiagogomes_ | thanks for the link dtantsur, I'll have a look | 12:15 |
tiagogomes_ | dtantsur, "Make sure that enough nodes are in available state according to ironic node-list". I can't see available in any column. Should it be on the "Provisioning State" column? | 12:22 |
dtantsur | tiagogomes_, yep. If you have older ironic and/or client, it may look like None (or NOSTATE?) | 12:23 |
*** subscope has quit IRC | 12:23 | |
tiagogomes_ | dtantsur, I am using Kilo version. It displays None | 12:23 |
*** ndipanov has joined #openstack-ironic | 12:23 | |
dtantsur | tiagogomes_, which client version? | 12:23 |
tiagogomes_ | dtantsur, 0.4.1 | 12:24 |
dtantsur | tiagogomes_, ok, so it's fine, this guide is for RDO, where we have 0.6 (at least) already | 12:24 |
dtantsur | None is fine anyway | 12:24 |
*** trown|outttypeww is now known as trown | 12:30 | |
lazy_prince | jroll: ironic neutron plugin refactored as a ML2 extension driver -> https://github.com/kumarom/ironic-neutron-plugin.git | 12:31 |
v92 | nice | 12:34 |
openstackgerrit | Merged openstack/ironic: Add vendor-passthru to attach and boot an ISO https://review.openstack.org/188317 | 12:35 |
tiagogomes_ | mmm, everything looks fine but I still get the "No valid host was found" error | 12:35 |
dtantsur | hmmm | 12:36 |
*** jjohnson2 has joined #openstack-ironic | 12:37 | |
dtantsur | tiagogomes_, could you paste your `ironic node-show UUID`? | 12:37 |
*** subscope has joined #openstack-ironic | 12:38 | |
tiagogomes_ | dtantsur, http://paste.openstack.org/show/285138/ | 12:38 |
*** MattMan has joined #openstack-ironic | 12:38 | |
tiagogomes_ | dtantsur, I have instance_info set, because I tried Ironic without Nova before. Could be because of that? | 12:39 |
dtantsur | tiagogomes_, no, but I see your node properties != your flavor. If you use exact filters (we in RDO don't IIRC), that may be the problem | 12:40 |
dtantsur | tiagogomes_, what if you make your flavor = your nodes (ram and disk-wise)? | 12:40 |
tiagogomes_ | dtantsur that's what I had before, and it didn't work | 12:41 |
dtantsur | the same error? | 12:41 |
tiagogomes_ | yep | 12:41 |
dtantsur | weird. anyway before trying further, I suggest bringing them in sync | 12:42 |
dtantsur | tiagogomes_, any exceptions in Nova or Ironic conductor logs? | 12:42 |
*** yog_ has joined #openstack-ironic | 12:43 | |
tiagogomes_ | dtantsur, http://paste.openstack.org/show/285139/ | 12:45 |
devananda | dtantsur: http://www.gnu.org/software/libtool/manual/html_node/Libtool-versioning.html | 12:45 |
tiagogomes_ | Ironic conductor doesn't show any error | 12:46 |
devananda | fwiw, just having coffee with mordred and discussing api versioning and whether vendor_passthru changes should bump it (answer: no) | 12:46 |
dtantsur | devananda, one of the reasons I was aleays avoiding libtool. As they say "Note that this can cause problems because libtool’s compatibility requirements are actually stricter than is necessary." | 12:47 |
dtantsur | devananda, I don't disagree, I just point it as source of inconsistency in the whole approach | 12:47 |
*** mordred has joined #openstack-ironic | 12:47 | |
devananda | dtantsur: ironic never claimed to express a contract around, or support in any way, what is in /vendor_passthru/ | 12:48 |
*** trown is now known as trown|brb | 12:48 | |
devananda | though the semantic of how one discovers that -- that's our contract | 12:48 |
devananda | I was reviewing lucas-hungry's change to add File support to vendor responses | 12:48 |
devananda | didn't know WSME could do that :) | 12:49 |
dtantsur | devananda, I still don't understand while exposing a list of strings via API (= my feature detection proposal) is worse than making hard work on every API-touching patch feature gating it's changes (= Nova approach) | 12:49 |
dtantsur | devananda, wow, haven't see that patch | 12:49 |
devananda | 189716 | 12:49 |
dtantsur | oh amazing! I wonder what output format is, but it's anyway amazing | 12:50 |
devananda | wtf? I dont have stable branch rights on ironic now | 12:51 |
dtantsur | O__O | 12:51 |
* devananda pokes infra team | 12:55 | |
devananda | dtantsur: http://www.gnu.org/software/libtool/manual/libtool.html#Updating-version-info | 12:59 |
devananda | this is a different way to express the min/max/current that our API expresses today | 13:00 |
*** e0ne has joined #openstack-ironic | 13:01 | |
*** krtaylor has quit IRC | 13:02 | |
devananda | dtantsur: so, what I think you want is for the client to express "does the REST API support version X?" whereas our implementatoin, today, only allows the client to express "plesae give me version X" | 13:03 |
*** lucas-hungry is now known as lucasagomes | 13:03 | |
devananda | dtantsur: and I agree that a client who expects version X does not care about the existence of new interfaces -- so there is no benefit to that client if our REST API hides them | 13:04 |
lucasagomes | devananda, morning tiagogomes_ | 13:04 |
lucasagomes | devananda, yeah, we need that for the iPXE dynamic config :-) | 13:05 |
lucasagomes | so we are going to generate the conf files and iPXE will download it directly from our api | 13:05 |
lucasagomes | slick | 13:05 |
*** trown|brb is now known as trown | 13:05 | |
*** ijw has joined #openstack-ironic | 13:06 | |
BadCub | morning folks | 13:07 |
*** [1]cdearborn has joined #openstack-ironic | 13:07 | |
lucasagomes | dtantsur, the output format is the raw content of the file | 13:08 |
lucasagomes | if you do a wget <api url>/vendor_passthru/method it will download a file | 13:08 |
lucasagomes | BadCub, good ugt morning | 13:09 |
BadCub | mornin lucasagomes and happy friday! | 13:09 |
devananda | dtantsur: we need to increment the api version number when adding an interface (rest api endpoint, etc) but we don't need to hide it when the requested version was less than that | 13:10 |
devananda | dtantsur: so that a client who depends on that interface is ble t odetect that older servers do not support it | 13:10 |
openstackgerrit | Vladyslav Drok proposed openstack/ironic: Remove tools/flakes.py https://review.openstack.org/191061 | 13:10 |
lucasagomes | BadCub, TGIF :-) happy friday there too | 13:10 |
dtantsur | devananda, I still prefer string-based detection (I was even thinking about using spec names as feature names), but sounds good as well | 13:11 |
dtantsur | devananda, what about other API changes? e.g. we feature-gated new provision verbs, which I think was a mistake. wdyt? | 13:11 |
dtantsur | BadCub, morning, happy Friday :) | 13:11 |
*** ijw has quit IRC | 13:11 | |
devananda | dtantsur: string based? spec names as api version? huh? | 13:11 |
dtantsur | devananda, it's still about potential API like $ ironic feature-supported <feature-name> | 13:12 |
dtantsur | but well, I'm ok with versions, unless we go insane while hiding new stuff from users :) | 13:12 |
devananda | dtantsur: omg no | 13:12 |
*** achanda has joined #openstack-ironic | 13:13 | |
dtantsur | well, we did hide from users thing like new management verbs. I think lucasagomes tries to introduce microversion for new argument to existing API. I don't think it should happen any more, wdyt, devananda? | 13:13 |
* lucasagomes reads | 13:14 | |
lucasagomes | oh, yeah cause that's a new feature in the api right? | 13:14 |
lucasagomes | be able to fetch a subset of attributes on the resources | 13:14 |
lucasagomes | (for ref, the patch is this one https://review.openstack.org/#/c/183491/) | 13:15 |
mordred | the thing is - I think that's talking about different things | 13:15 |
dtantsur | I meant to say, bumping version is good, but not hiding this feaure, right? | 13:15 |
mordred | ironic feature-supported is a thing implemented by the client program and client library | 13:15 |
devananda | dtantsur: so for S/NOSTATE/ACTIVE/, we had to have a backwards-capable wrapper | 13:15 |
mordred | it can answer that question | 13:15 |
lucasagomes | dtantsur, I iwll read the scrollback, I dunno what it meas by hiding | 13:15 |
mordred | dtantsur: I, btw, agree with you that hiding the feature is not useful, and I think I've convinced devananda of that too :) | 13:16 |
dtantsur | lucasagomes, this: https://github.com/openstack/ironic/blob/master/ironic/api/controllers/v1/node.py#L58-L69 | 13:16 |
dtantsur | lucasagomes, or this: https://github.com/openstack/ironic/blob/master/ironic/api/controllers/v1/node.py#L79-L87 | 13:16 |
dtantsur | mordred, good to hear :) | 13:16 |
lucasagomes | dtantsur, oh right | 13:17 |
mordred | dtantsur: I think the question the client interaction is asking for is "I am programmed against API version 4, can you (the server) support that for me please" | 13:17 |
lucasagomes | I thought it was done because someone might want to check like schema | 13:17 |
mordred | to which the server can respond either "yes" or "no" | 13:17 |
devananda | mordred: hiding new endpoints - yes, I agree, not useful | 13:17 |
* lucasagomes thinks we should have schemas in our api | 13:17 | |
mordred | supporting API version 4 does not say anything about theoretical existence of things that are not part of vesrion 4 | 13:17 |
devananda | and in order to support version 4, sometimes the server needs to wrap / translate / transform some data so that it looks the same as it did in version 4 | 13:18 |
mordred | if the client, on the other hand requests version 4 and then uses version 5 - it _is_ a bug in the client program | 13:18 |
mordred | but I do not believe it's useful for the server to enforce that bug | 13:18 |
dtantsur | devananda, only when change is breaking IMO | 13:18 |
mordred | it _might_ be useful for the client library to detect and warn the user about such a behavior | 13:18 |
mordred | since it means the program may be accidentally incompatible with a different cloud | 13:19 |
mordred | but now we're getting in to some esoteric problems | 13:19 |
dtantsur | devananda, I don't agree https://github.com/openstack/ironic/blob/master/ironic/api/controllers/v1/node.py#L58-L69 is sane, unless we account for insane clients (which I've seen tbh, but they're hard to account for) | 13:19 |
[1]cdearborn | Morning Ironic! | 13:20 |
TheJulia | good morning | 13:20 |
dtantsur | [1]cdearborn, morning! | 13:20 |
devananda | dtantsur: looking at list of features we've bumped api version for so far -- | 13:21 |
devananda | # v1.2: Renamed NOSTATE ("None") to AVAILABLE ("available") | 13:21 |
devananda | # v1.3: Add node.driver_internal_info | 13:21 |
devananda | # v1.4: Add MANAGEABLE state | 13:21 |
devananda | # v1.5: Add logical node names | 13:21 |
devananda | # v1.6: Add INSPECT* states | 13:21 |
devananda | two of those added a new resource: driver_internal_info and "name" | 13:21 |
devananda | two added new verbs -- which affects both the allowable values in a POST request, and the set of possible returned values for provision-state | 13:22 |
devananda | entirely new semantics in the client became possible -- but the old semantics were still valid, too | 13:22 |
devananda | only 1.2 was incompatible in the strictest sense | 13:22 |
dtantsur | right | 13:22 |
BadCub | morning and happy friday dtantsur devananda mordred | 13:23 |
*** krtaylor has joined #openstack-ironic | 13:26 | |
NobodyCam | Good morning Ironicers and..... | 13:26 |
NobodyCam | TGIF! | 13:26 |
dtantsur | NobodyCam, hey, TGIF! | 13:26 |
[1]cdearborn | morning NobodyCam, +25! | 13:26 |
NobodyCam | morning dtantsur and [1]cdearborn :) | 13:27 |
mordred | morning BadCub ! | 13:27 |
NobodyCam | how goes this Friday | 13:27 |
NobodyCam | morning mordred :) | 13:27 |
devananda | dtantsur: so, in summary, I dont think we need to hide compatible changes any longer, but we should be bumping the current version on any api-affecting change | 13:27 |
mordred | morning NobodyCam ! | 13:27 |
NobodyCam | morning devananda | 13:27 |
NobodyCam | :) | 13:27 |
[1]cdearborn | I officially have a lab connection to my desk, so I don't have to hike to the gym any more to do git review. Life Is Good! | 13:28 |
dtantsur | devananda, ++ agreed | 13:28 |
dtantsur | devananda, can/should we move this discussion to the ML for visibility once we have at least rough consensus now? | 13:28 |
devananda | dtantsur: and if we add a small incompatible change that we CAN wrap, then we dont need to bump the minimum supported | 13:28 |
devananda | for example - changing the default state for newly created nodes to ENROLL or MANAGEABLE | 13:29 |
devananda | that behavior change MUST be guarded with a new version | 13:29 |
*** achanda_ has joined #openstack-ironic | 13:29 | |
dtantsur | definitely (it's in the spec IIRC) | 13:29 |
devananda | but we CAN support the old behavior still, if a client requests version < 1.7 (for instance) | 13:29 |
*** achanda has quit IRC | 13:31 | |
dtantsur | devananda, speaking of which, any chances you have a look at the ENROLL spec (179151)? It's a hard change, I think it's fair to drag you into reviewing it :) | 13:32 |
devananda | dtantsur: not yet. will look in a minute | 13:32 |
dtantsur | thnx! | 13:32 |
*** bradjones has quit IRC | 13:35 | |
*** bradjones has joined #openstack-ironic | 13:38 | |
*** bradjones has quit IRC | 13:38 | |
*** bradjones has joined #openstack-ironic | 13:38 | |
*** derekh is now known as derekh_afk | 13:39 | |
openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Refactor node's and driver's vendor passthru to a common place https://review.openstack.org/190121 | 13:50 |
openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Allow vendor methods to serve static files https://review.openstack.org/189716 | 13:50 |
openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Change return value of [driver_]vendor_passthru to dict https://review.openstack.org/190120 | 13:50 |
* lucasagomes fixes the py34 problem | 13:50 | |
*** krtaylor has quit IRC | 13:51 | |
*** dmitchell has joined #openstack-ironic | 13:52 | |
NobodyCam | morning lucasagomes | 13:52 |
lucasagomes | btw I installed py34 now, but when I run the tests a get loads of "object is not iterable" | 13:52 |
lucasagomes | I know the cause and that it's a mock bug | 13:52 |
lucasagomes | but it kinda sucks :-( | 13:52 |
lucasagomes | NobodyCam, morning | 13:52 |
devananda | dtantsur: 179151 +2'd -- I'm good with approving it, but also happy to give others a couple more days | 13:52 |
devananda | lucasagomes: see my response o nthe patch fixing py34 on fedora? | 13:53 |
lucasagomes | devananda, we have a patch for that? /me looks | 13:53 |
lucasagomes | I haven't | 13:53 |
* lucasagomes finds it | 13:53 | |
lucasagomes | https://review.openstack.org/#/c/190835/ | 13:53 |
devananda | yes | 13:54 |
*** dmitchell__ has quit IRC | 13:54 | |
lucasagomes | devananda, right, what version pf py34 you have there? I think it may affect other OSs later | 13:54 |
lucasagomes | cause it's a bug in mock | 13:54 |
lucasagomes | fedora just happens to have a newer version | 13:54 |
lucasagomes | AFAUI, I will read more but I might be wrong | 13:54 |
lucasagomes | devananda, but that's a super valid point, I would like to have some fedora gating | 13:55 |
NobodyCam | lucasagomes: when you have a few minute could I by chance draw on your RH experence for a biFrost review | 13:56 |
NobodyCam | https://review.openstack.org/#/c/185294 | 13:56 |
lucasagomes | NobodyCam, ack | 13:56 |
NobodyCam | awesome TY | 13:56 |
TheJulia | ty lucasagomes | 13:57 |
devananda | lucasagomes: ++ on fedora jobs, but I think they'll need to run as third-party CI :-/ | 13:57 |
lucasagomes | devananda, http://packages.ubuntu.com/trusty/python3 seems that's 3.4.0 that we run on gate | 13:57 |
devananda | yup | 13:57 |
lucasagomes | The following fails on python 3.4.2, 3.4.3 and 3.5.0a2 (downloaded from python.org and compiled on Ubuntu 14.04). | 13:57 |
devananda | oooh | 13:57 |
lucasagomes | once it's updated to 3.4.2+ it's going to fail on ubuntu too | 13:58 |
lucasagomes | devananda, http://bugs.python.org/issue23661. | 13:58 |
lucasagomes | that's the version of the mock in python, it's not OS related | 13:58 |
sambetts | lucasagomes: I get the 3.4.2 failures on my local machine :( | 13:58 |
devananda | mordred: ^ | 13:58 |
lucasagomes | sambetts, right | 13:58 |
devananda | lucasagomes: can you start a discussion about that in -infra ? | 13:59 |
NobodyCam | morning sambetts | 13:59 |
lucasagomes | devananda, ack, I will comment on that patch first and jump on -infra | 13:59 |
devananda | ty | 13:59 |
* devananda jumps on a meeting | 13:59 | |
jroll | I wish we could get python cores to land that mock patch, though I doubt they'd backport it all the way back to 3.4.2 | 14:01 |
jroll | g'morning everyone | 14:01 |
dtantsur | jroll, morning | 14:05 |
dtantsur | devananda, ty, yeah let's make people aware of this spec :) | 14:05 |
lazy_prince | jroll: morning | 14:06 |
lazy_prince | jroll: ironic neutron plugin refactored as a ML2 extension driver -> https://github.com/kumarom/ironic-neutron-plugin.git | 14:06 |
*** ijw has joined #openstack-ironic | 14:06 | |
NobodyCam | morning jroll lazy_prince :) | 14:06 |
jroll | lazy_prince: yep, saw that, nice :) | 14:06 |
lazy_prince | NobodyCam: its actually night here.. | 14:07 |
jroll | it's always morning in ironiclandia! | 14:07 |
jroll | lazy_prince: http://www.total-knowledge.com/~ilya/mips/ugt.html :) | 14:08 |
lazy_prince | :) | 14:08 |
*** achanda_ has quit IRC | 14:08 | |
*** rbudden has joined #openstack-ironic | 14:10 | |
NobodyCam | :) jroll ++++ on UGT link!!! hehehehe | 14:10 |
jroll | :P | 14:10 |
*** ijw has quit IRC | 14:12 | |
*** yog_ has quit IRC | 14:13 | |
jroll | any opposition to +A on https://review.openstack.org/#/c/173218/6 | 14:13 |
jroll | 4x +2 on the previous patchset, this one had no changes just a rebase | 14:13 |
* jroll does it | 14:13 | |
jroll | oh I guess I was wrong, I must have been thinking of a different spec... oops. | 14:14 |
jroll | ¯\_(ツ)_/¯ | 14:15 |
dtantsur | lol | 14:16 |
dtantsur | well, this spec does not look too complicated to me | 14:16 |
*** bnemec has joined #openstack-ironic | 14:16 | |
*** dmitchell__ has joined #openstack-ironic | 14:17 | |
*** dmitchell has quit IRC | 14:20 | |
*** [1]cdearborn has quit IRC | 14:20 | |
openstackgerrit | Merged openstack/ironic-specs: Add inband RAID configuration spec for liberty https://review.openstack.org/173218 | 14:21 |
sambetts | dtantsur: If I change NodeNotFoundHook into just a function instead of a class, don't we lose the enforcement that the abc abstract class thing gives us? | 14:22 |
dtantsur | sambetts, it does not gives us much actually | 14:22 |
dtantsur | I doubt it enforces function signatures for example.. | 14:23 |
sambetts | dtantsur: I just realised I forgot the @abc.abstractmethod, <- I'm pretty sure that does | 14:23 |
jroll | it doesn't, nova broke us once by adding an argument to a driver method | 14:23 |
dtantsur | sambetts, that's the only thing it checks IIRC | 14:23 |
jroll | though idk if those are abstract, ignore me | 14:23 |
dtantsur | sambetts, I meant, IIRC it checks for overloading, but not for signatures | 14:24 |
*** cdearborn has joined #openstack-ironic | 14:25 | |
sambetts | dtantsur: Just reading the docs it seems that it just ensures that the function is overloaded by subclasses, nothing more | 14:26 |
dtantsur | sambetts, right. And if we have one function, we're already sure we have a function :) | 14:26 |
mjturek1 | hey all, maurosr and I had noticed that there's a check in devstack that sets the rootfs type to ramfs if the node RAM is set to <1024. We were wondering if this might be a sanity check worth bringing into ironic proper. We're thinking that, during a deploy, if the flavor's RAM is <1024, ramfs is automatically used as the rootfstype rather than what's in ironic.conf. | 14:27 |
openstackgerrit | Julia Kreger proposed openstack/bifrost: Add os_ironic_facts module https://review.openstack.org/191093 | 14:27 |
maurosr | this lines https://github.com/openstack-dev/devstack/blob/master/lib/ironic#L353-L356 specifically | 14:29 |
*** krtaylor has joined #openstack-ironic | 14:32 | |
*** ijw has joined #openstack-ironic | 14:34 | |
TheJulia | mjturek1: I think that would be something that would be more deployment specific detail. Additionally we have many users focusing on use of the Ironic Python Agent which realistically requires a very different footprint to deploy. | 14:35 |
mjturek1 | TheJulia: ahh generally a larger footprint for IPA? | 14:36 |
TheJulia | mjturek1: like 2.5-3GB minimum | 14:37 |
mjturek1 | gotcha | 14:37 |
jroll | o.o | 14:38 |
mjturek1 | TheJulia: maurosr and I are willing to implement the check, think it'd be worth at least showing the community? Of course it's understandable if it's ultimately deemed not worthwhile | 14:39 |
*** zz_jgrimm is now known as jgrimm | 14:39 | |
jroll | TheJulia: that seems... high. | 14:39 |
jroll | is that coreos ramdisk or dib | 14:39 |
TheJulia | jroll: coreos | 14:39 |
jroll | hrm | 14:39 |
dtantsur | I was assuming only dib ramdisk has this problem... | 14:39 |
jroll | I guess maybe with pulling images into the ramdisk | 14:39 |
jroll | it's totally fine with 1gb with cirros... so that plus the size of your image I guess | 14:40 |
*** romainh has left #openstack-ironic | 14:41 | |
*** jgrimm is now known as zz_jgrimm | 14:41 | |
TheJulia | jroll: I seem to remember one of my cohorts ran into issues trying to boot the coreos IPA image in <1.5GB of ram and just didn't get anywhere until we took it to 2GB. | 14:42 |
jroll | TheJulia: odd, the gate runs on 1gb :) | 14:43 |
TheJulia | Hmm.. | 14:43 |
dtantsur | jroll, (maybe stupid question follows) can't we somehow stream an image to the hard driver (aka curl | dd), so that we don't have to keep it in memory? | 14:43 |
jroll | dtantsur: yeah. totally. got some time to work on it? :) | 14:44 |
TheJulia | jroll: very interesting..... | 14:44 |
dtantsur | jroll, likely to have this cycle | 14:44 |
dtantsur | no right now though | 14:44 |
jroll | dtantsur: comstud started working on making image writing better/pluggable before he stopped working on this stuff, never ressurected it | 14:44 |
TheJulia | dtantsur: I think it gets written through qemu-img out to disk | 14:44 |
TheJulia | but if it supports stdin, then that could work | 14:44 |
jroll | it was more about vhd support, but may help here https://review.openstack.org/#/c/128701/ | 14:45 |
jroll | maybe not, dunno | 14:45 |
dtantsur | if nobody takes this task, I'll get to it this cycle. lowering memory requirements is a must.. | 14:45 |
jroll | how many 2gb servers do people deploy these days? :) | 14:45 |
jroll | and/or huge images | 14:45 |
jroll | I guess that might be the bigger issue | 14:46 |
dtantsur | well, people tend to deploy on weird things :) | 14:46 |
jroll | heh, yeah | 14:46 |
jroll | our smallest flavor is 32gb so it hasn't been a priority for us :P | 14:46 |
dtantsur | well, for servers it's valid :) | 14:49 |
dtantsur | jroll, though if people try to deploy Windows image with a lot of stuff installed.... | 14:49 |
*** e0ne is now known as e0ne_ | 14:49 | |
dtantsur | ... and do it on a server with only 16 GiB.... | 14:49 |
jroll | dtantsur: yep. agree | 14:49 |
jlvillal | Good morning Ironic | 14:50 |
TheJulia | dtantsur: I wouldn't mind seeing a smaller footprint for testing bifrost since we currently only provision 1 VM :( | 14:50 |
TheJulia | good morning | 14:50 |
dtantsur | jlvillal, morning, happy Friday | 14:51 |
jlvillal | dtantsur, Thanks :) I hope you enjoy your weekend too! | 14:51 |
dtantsur | it's gonna be pretty hot here :) maybe some swimming, will see | 14:52 |
jlvillal | dtantsur, Warm weather sounds good to me. It will take a lot more warm weather here before I start complaining. | 14:52 |
jlvillal | dtantsur, I will be in Moscow in July. My wife is warning me that it will probably be hot. | 14:53 |
*** yog_ has joined #openstack-ironic | 14:53 | |
*** erelwell is now known as bethelwell | 14:53 | |
jlvillal | TheJulia, Good morning to you too :) | 14:53 |
dtantsur | jlvillal, it can be either deadly hot, or deadly cold with rain. no one can guess :) | 14:53 |
jlvillal | dtantsur, You are not making me excited to go :P | 14:54 |
dtantsur | lol sorry :) | 14:54 |
*** cdearborn has quit IRC | 14:54 | |
dtantsur | jlvillal, May is the best time to go nearly anywhere, other seasons might suck | 14:54 |
*** e0ne_ is now known as e0ne | 14:55 | |
NobodyCam | morning jlvillal | 14:56 |
jlvillal | NobodyCam, Thanks and you too. | 14:57 |
NobodyCam | :) | 14:57 |
*** jcoufal_ has quit IRC | 15:07 | |
sambetts | dtantsur: If I move the Error raise into the helper function in process and don't pass in the failures list, then we won't be able to include the failures that occur in the preprocessing hook loop in the Error thats raised | 15:09 |
openstackgerrit | Christopher Dearborn proposed openstack/ironic: Fix failing unit tests under py34 https://review.openstack.org/190835 | 15:10 |
*** alex_xu is now known as alexus | 15:11 | |
dtantsur | sambetts, right. nevermind then, we can always refactor later | 15:11 |
*** achanda has joined #openstack-ironic | 15:11 | |
sambetts | dtantsur: Yeah, cool, I'll push a new patch set with your comments addressed :) | 15:13 |
openstackgerrit | Julia Kreger proposed openstack/bifrost: Changes to ironic-install module for NUC testing in Vagrant https://review.openstack.org/191109 | 15:15 |
openstackgerrit | Julia Kreger proposed openstack/bifrost: Vagrantfile and vagrant.yml for testing NUC https://review.openstack.org/191110 | 15:15 |
*** yog_ has quit IRC | 15:16 | |
*** e0ne is now known as e0ne_ | 15:18 | |
openstackgerrit | Dmitry Tantsur proposed stackforge/ironic-discoverd: Update devstack plugin for the recent authentication change https://review.openstack.org/191111 | 15:19 |
*** cdearborn has joined #openstack-ironic | 15:19 | |
TheJulia | devananda: ^^^ two reviews posted from your AMT/vagrant stuffs | 15:20 |
*** zhenguo has joined #openstack-ironic | 15:20 | |
*** dprince has quit IRC | 15:22 | |
lazy_prince | night all.. | 15:24 |
* lazy_prince goes home.. | 15:24 | |
*** lazy_prince has quit IRC | 15:25 | |
*** achanda has quit IRC | 15:25 | |
*** e0ne_ is now known as e0ne | 15:27 | |
openstackgerrit | Jim Rollenhagen proposed openstack/ironic-specs: Change release model to independent releases https://review.openstack.org/185171 | 15:32 |
*** hemna has joined #openstack-ironic | 15:32 | |
jroll | ^^ folks I've finally updated the release model spec with feedback from the review and the ML, pls to review :) | 15:33 |
jroll | devananda: ^ especially | 15:33 |
*** yog_ has joined #openstack-ironic | 15:33 | |
*** Marga_ has quit IRC | 15:37 | |
*** rameshg87 has joined #openstack-ironic | 15:39 | |
openstackgerrit | Sam Betts proposed stackforge/ironic-discoverd: Provide hook into process when a node isn't found https://review.openstack.org/190124 | 15:40 |
*** absubram has joined #openstack-ironic | 15:41 | |
rameshg87 | jroll: hi | 15:41 |
rameshg87 | jroll: can you please +A the change https://review.openstack.org/#/c/189241/ if it seems okay | 15:41 |
jroll | hey rameshg87, what's up :) | 15:41 |
jroll | dtantsur: any reason you didn't +A that? | 15:42 |
jroll | rameshg87: will do unless dmitry has a reason not to | 15:42 |
*** absubram has quit IRC | 15:42 | |
rameshg87 | jroll: yeah sure .. | 15:42 |
jroll | bbiaf | 15:43 |
*** absubram has joined #openstack-ironic | 15:43 | |
*** tiagogomes_ has left #openstack-ironic | 15:46 | |
*** hemna has quit IRC | 15:48 | |
dtantsur | see y'all next week, g'night! | 15:49 |
*** dtantsur is now known as dtantsur|afk | 15:49 | |
NobodyCam | have a good weekend dtantsur|afk | 15:49 |
*** hemna has joined #openstack-ironic | 15:49 | |
*** achanda has joined #openstack-ironic | 15:50 | |
lucasagomes | rameshg87, jroll it lgtm, should I wait to +a that? | 15:50 |
dtantsur|afk | jroll, not +A as I don't know IPA well enough :) | 15:51 |
rameshg87 | lucasagomes: looks like I can't test wol driver for a while. I took my old cpu to nearby hardware shop, I expected a SMPS failure, but it seems something wrong with the board, it's not powering on. :( | 15:58 |
lucasagomes | rameshg87, hmm | 15:59 |
lucasagomes | rameshg87, powering on via WOL? or powering on in general? | 15:59 |
rameshg87 | lucasagomes: powering on in general. | 15:59 |
rameshg87 | lucasagomes: not even with power button :( | 15:59 |
lucasagomes | oh damn | 16:01 |
*** subscope has quit IRC | 16:04 | |
*** wuhg has joined #openstack-ironic | 16:05 | |
jroll | dtantsur|afk: I see | 16:10 |
jroll | lucasagomes: yeah go ahead | 16:10 |
lucasagomes | jroll, morning ugt :-) | 16:10 |
lucasagomes | done | 16:10 |
jroll | heh morning :) | 16:11 |
jroll | thanks! | 16:11 |
rameshg87 | \o/ | 16:11 |
*** bnemec has quit IRC | 16:16 | |
openstackgerrit | Laura Moore proposed openstack/ironic-specs: Update of the Ironic ML2 Integration spec https://review.openstack.org/188528 | 16:18 |
rameshg87 | clif_h: hi | 16:20 |
jroll | rameshg87: he's on vacation, what's up | 16:20 |
rameshg87 | jroll: oh okay | 16:21 |
rameshg87 | jroll: wanted to discuss something regarding https://review.openstack.org/#/c/161832/ | 16:21 |
jroll | rameshg87: I can try :) | 16:22 |
rameshg87 | jroll: had some comments on it and thought it was better to go through | 16:22 |
rameshg87 | jroll: yeah sure | 16:22 |
rameshg87 | jroll: first - https://review.openstack.org/#/c/161832/15/ironic/dhcp/neutron.py | 16:22 |
*** dguerri is now known as dguerri` | 16:22 | |
rameshg87 | jroll: is there any real-use case of separate caching vs cleaning network ? | 16:22 |
jroll | rameshg87: so I don't think there's a real use case for different caching vs *provisioning* network | 16:23 |
jroll | rameshg87: but I think there's a use case for different cleaning network | 16:23 |
jroll | as you don't know the state of the machine when you bring it up there | 16:23 |
jroll | and we don't have a concept of a provisioning network yet | 16:23 |
*** e0ne has quit IRC | 16:24 | |
rameshg87 | jroll: okay, so later on it can be merged into provisioning network ? | 16:24 |
jroll | rameshg87: I think so yeah | 16:24 |
jroll | I don't have a strong opinion on that | 16:25 |
rameshg87 | jroll: my question because it adds more methods to our dhcp api and once it is added, it's hard to recall back | 16:25 |
*** viktors is now known as viktors|afk | 16:26 | |
zhenguo | hi all, I create a new project ironic-dashboard as the UI component for Ironic. | 16:26 |
jroll | rameshg87: indeed. not sure if there's a better way to do that | 16:26 |
*** Marga_ has joined #openstack-ironic | 16:27 | |
rameshg87 | okay. may be will keep that aside and see if provisioning network concept takes shape by that time | 16:27 |
zhenguo | please help to review https://review.openstack.org/#/c/191142/ and https://review.openstack.org/#/c/191131/ | 16:27 |
zhenguo | the init repo is here: https://github.com/niuzhenguo/ironic-dashboard | 16:28 |
rameshg87 | jroll: and https://review.openstack.org/#/c/161832/15/ironic/drivers/modules/agent.py L644 | 16:28 |
*** dprince has joined #openstack-ironic | 16:28 | |
rameshg87 | jroll: I meant if we do something in validate(), the API waits till we complete it | 16:29 |
jroll | rameshg87: I don't get your point | 16:30 |
jroll | zhenguo: have you seen krotscheck 's work on this? https://github.com/krotscheck/ironic-webclient | 16:30 |
rameshg87 | jroll: and hence we could fail the API call itself, instead of doing the error checks in vendorpassthru | 16:30 |
jroll | zhenguo: or is this for horizon? | 16:30 |
zhenguo | jroll, yes it's a plugin for horizon | 16:30 |
rameshg87 | jroll: for example we just return in vendor passthru if node.provision_state != states.AVAILABLE | 16:31 |
rameshg87 | jroll: the vendor passthru just silently exists | 16:31 |
jroll | zhenguo: ok | 16:31 |
zhenguo | jroll, like other project's UI project like manila-ui, tuscar-ui, | 16:31 |
rameshg87 | jroll: we could have done it in validate() and thrown an error back to the caller of the API | 16:31 |
jroll | rameshg87: what line are you looking at?' | 16:31 |
jroll | oh sorry 644 | 16:31 |
rameshg87 | jroll: https://review.openstack.org/#/c/161832/15/ironic/drivers/modules/agent.py L644 | 16:31 |
jroll | rameshg87: yeah, looking | 16:32 |
openstackgerrit | Mario Villaplana proposed openstack/ironic-specs: Adds spec for creating a key manager with Barbican to Ironic https://review.openstack.org/188609 | 16:33 |
rameshg87 | jroll: a lot of such checks down moved to validate() will help in better experience with the API | 16:33 |
jroll | rameshg87: I think I agree, it should at least return an error | 16:33 |
rameshg87 | jroll: okay will comment on the review | 16:34 |
jroll | rameshg87: I will too | 16:34 |
rameshg87 | jroll: https://review.openstack.org/#/c/161832/15/ironic/drivers/modules/agent.py L707 | 16:34 |
rameshg87 | jroll: wondering if creation of neutron ports by ironic in available state will lead to race between nova and ironic if node gets selected at the same time | 16:35 |
rameshg87 | jroll: we just lock the ironic node in available state and do stuff, but nova doesn't know about it | 16:35 |
*** zhenguo has quit IRC | 16:35 | |
rameshg87 | do stuff = caching | 16:36 |
jroll | rameshg87: urgh. yeah. | 16:36 |
jroll | I guess we didn't think about that | 16:36 |
jroll | rameshg87: I think the confusion there is because we don't use neutron for dhcp | 16:37 |
rameshg87 | yeah, that solves most of the problems :) | 16:37 |
jroll | heh | 16:37 |
rameshg87 | but yeah | 16:37 |
rameshg87 | if neutron port creation failure in nova just leads to nova picking up another node, then it's fine | 16:37 |
rameshg87 | I mean at least okay in my opinion, because (node in caching) = (not available) | 16:38 |
jroll | rameshg87: yeah, but let's avoid reschedules if possible :) | 16:39 |
*** zz_natorious is now known as natorious | 16:39 | |
jroll | if it does it 3 times in a row the boot will fail | 16:39 |
rameshg87 | yeah | 16:39 |
jroll | we'll have to think about this one... I'll leave a note on the review | 16:39 |
jroll | thanks for catching that :) | 16:39 |
rameshg87 | okay | 16:39 |
rameshg87 | :) | 16:39 |
rameshg87 | jroll: and it sort of seems like we are misusing vendor interface | 16:40 |
jroll | rameshg87: well, the bash ramdisk can't support this :) | 16:41 |
rameshg87 | jroll: agent is not a vendor and our own stuff :) | 16:41 |
jroll | rameshg87: I tend to think vendor passthru name is a bad name | 16:42 |
jroll | rameshg87: it's really more for things specific to a single driver (rather, a single implementation of an interface) | 16:42 |
rameshg87 | yeah it is | 16:43 |
jroll | though I guess I'd like this to be a real endpoint one day | 16:43 |
rameshg87 | jroll: +1 | 16:44 |
rameshg87 | jroll: yeah I will think about nova thing further if I can find anything about it | 16:45 |
rameshg87 | just saw clif_h change on nova side and getting full picture of it | 16:45 |
jroll | rameshg87: I'll think about it as well | 16:46 |
jroll | rameshg87: have you looked at arsenal also? | 16:46 |
*** Sukhdev has joined #openstack-ironic | 16:46 | |
rameshg87 | jroll: what's that ? | 16:46 |
rameshg87 | I just looked at the review link that he posted in comment | 16:46 |
jroll | rameshg87: https://github.com/rackerlabs/arsenal | 16:47 |
jroll | basically a service that discovers images and ironic nodes, and decides how to cache what and on where | 16:47 |
jroll | also has docs https://arsenal.readthedocs.org/en/latest/ | 16:47 |
rameshg87 | oh great .. | 16:47 |
*** trown is now known as trown|lunch | 16:48 | |
rameshg87 | jroll: you guys are miles ahead of ironic everytime :) | 16:48 |
jroll | rameshg87: ha :) | 16:48 |
rameshg87 | jroll: will go through it | 16:48 |
openstackgerrit | Christopher Dearborn proposed openstack/ironic: Fix failing unit tests under py34 https://review.openstack.org/190835 | 16:48 |
rameshg87 | and nice name too | 16:49 |
jroll | rameshg87: turns out you move quickly when profit is a motivator :P | 16:49 |
* rameshg87 is a fan of arsenal club in epl | 16:49 | |
rameshg87 | jroll: :) | 16:49 |
jroll | heh | 16:49 |
*** cdearborn has quit IRC | 16:50 | |
lucasagomes | rameshg87, w00t! WOL upstream | 16:50 |
rameshg87 | \o/ | 16:50 |
lucasagomes | I will drink a pint for that tonight | 16:50 |
rameshg87 | lucasagomes: cheers | 16:50 |
jroll | \o/ | 16:50 |
lucasagomes | no more keep merging it downstream here | 16:50 |
lucasagomes | (it wasn't that painful, it's pretty much self contained. But anyway, glad I don't have to do it anymore) | 16:51 |
rameshg87 | lucasagomes: and you can let those people know who were waiting for this to get upstream :) | 16:51 |
lucasagomes | yup | 16:51 |
* lucasagomes prepares FYI emaisl | 16:52 | |
lucasagomes | emails* | 16:52 |
*** athomas has quit IRC | 16:52 | |
* rameshg87 is sad couldn't still try wol out | 16:52 | |
* rameshg87 will get hardware for it from somewhere | 16:52 | |
* rameshg87 thinks it's time to go to bed | 16:53 | |
*** bethelwell has quit IRC | 16:53 | |
rameshg87 | goodnight ugt jroll lucasagomes ironic | 16:53 |
jroll | see ya rameshg87 :) | 16:53 |
rameshg87 | see you next weekend | 16:53 |
jroll | have a good weekend | 16:53 |
lucasagomes | rameshg87, good ugt night! Enjoy the weekend | 16:53 |
* lucasagomes will go very soon too | 16:53 | |
rameshg87 | happy weekend to both of you too .. | 16:53 |
lucasagomes | thanks | 16:53 |
*** rameshg87 has quit IRC | 16:53 | |
jroll | have a good one lucasagomes \o | 16:53 |
openstackgerrit | Merged openstack/ironic-python-agent: Add power_off command in standby extension https://review.openstack.org/189241 | 16:54 |
lucasagomes | jroll, cheers! | 16:54 |
* jroll dives into work | 16:54 | |
lucasagomes | you too enjoy it | 16:54 |
openstackgerrit | Merged openstack/ironic: Wake-On-Lan Power interface https://review.openstack.org/179078 | 16:54 |
lucasagomes | I mean the weekend, but work as well :-) | 16:54 |
jroll | heh yeah :P | 16:54 |
jlvillal | lucasagomes: Ciao. Enjoy your weekend. I hope you have sun in Ireland :) | 16:55 |
lucasagomes | jlvillal, thanks you too! Have a good night enjoy the weekend | 16:56 |
jlvillal | :) | 16:56 |
lucasagomes | jlvillal, yeah, it's cloudy at the moment. But on the weekend I think the sun will show up! | 16:56 |
* lucasagomes *fingers crossed* | 16:56 | |
jlvillal | lucasagomes: I hope so :) it is supposed to be hot here. 31-32 I think. | 16:56 |
lucasagomes | 31C? | 16:56 |
lucasagomes | It's 17C :-( | 16:57 |
jlvillal | 17C here at the moment. But weekend warmer, I hope. | 16:57 |
lucasagomes | actually anything near 20C is already pretty "hot" here. The thermometers maxes out at 25C here, at least it saves mercury | 16:57 |
*** Marga_ has quit IRC | 16:57 | |
jlvillal | LOL! | 16:57 |
*** Marga_ has joined #openstack-ironic | 16:57 | |
jlvillal | lucasagomes: I lived in Huntingdon, England for three years. So I think somewhat similar. | 16:58 |
lucasagomes | jlvillal, I see, well in 4 years here. The max I've seem was 27C | 16:58 |
jlvillal | :( | 16:58 |
lucasagomes | it's ok... coming from brazil I'm glad it's cold | 16:58 |
jlvillal | lucasagomes: You adapt. Now when it hits 20C you think it is so hot ;) | 16:59 |
lucasagomes | I had enough sun from 3 lives already | 16:59 |
lucasagomes | for* | 16:59 |
lucasagomes | jlvillal, yeah | 16:59 |
lucasagomes | 20C is like wow | 16:59 |
lucasagomes | anyhoo, I see you all later | 17:00 |
lucasagomes | have a good night | 17:00 |
*** lucasagomes is now known as lucas-beer | 17:00 | |
jlvillal | lucasagomes: Thanks. You too. | 17:00 |
sambetts | night all :) | 17:00 |
jlvillal | sambetts: Ciao. | 17:00 |
lucas-beer | sambetts, night | 17:00 |
openstackgerrit | Christopher Dearborn proposed openstack/ironic: Fix failing unit tests under py34 https://review.openstack.org/190835 | 17:15 |
NobodyCam | night sambetts | 17:15 |
NobodyCam | have a good weekend | 17:15 |
*** Sukhdev has quit IRC | 17:27 | |
*** Sukhdev has joined #openstack-ironic | 17:33 | |
*** trown|lunch is now known as trown | 17:38 | |
*** amotoki has quit IRC | 17:41 | |
*** spandhe has joined #openstack-ironic | 17:42 | |
*** achanda has quit IRC | 17:46 | |
*** jistr has quit IRC | 17:48 | |
*** pelix has quit IRC | 17:52 | |
*** Sukhdev has quit IRC | 17:56 | |
*** achanda has joined #openstack-ironic | 17:59 | |
*** e0ne has joined #openstack-ironic | 18:00 | |
*** mitchjameson has joined #openstack-ironic | 18:06 | |
*** e0ne has quit IRC | 18:13 | |
*** achanda has quit IRC | 18:17 | |
*** cdearborn has joined #openstack-ironic | 18:18 | |
*** Sukhdev has joined #openstack-ironic | 18:19 | |
*** e0ne has joined #openstack-ironic | 18:19 | |
*** dprince has quit IRC | 18:20 | |
*** yog_ has quit IRC | 18:22 | |
*** e0ne is now known as e0ne_ | 18:25 | |
openstackgerrit | Imre Farkas proposed openstack/ironic: DRAC: fix set/get boot device for 11g https://review.openstack.org/191177 | 18:27 |
*** valkyrja has quit IRC | 18:27 | |
*** e0ne_ has quit IRC | 18:30 | |
*** dmitchell has joined #openstack-ironic | 18:32 | |
*** Marga_ has quit IRC | 18:32 | |
*** bethelwell has joined #openstack-ironic | 18:33 | |
*** yog_ has joined #openstack-ironic | 18:33 | |
*** dmitchell__ has quit IRC | 18:33 | |
jlvillal | Does anyone know much about ironic/nova/compute/manager.py? Is it owned by Ironic team? Is it a copy of some Nova code? | 18:35 |
jlvillal | lifeless: ^^^ | 18:35 |
NobodyCam | jlvillal: in our repo... | 18:36 |
jlvillal | NobodyCam: Yes | 18:36 |
NobodyCam | that the clustered compute manager | 18:36 |
jlvillal | NobodyCam: Do you know if it is code unique to Ironic? Or something that gets copied in occasionally from Nova. | 18:37 |
jlvillal | I'm thinking unique to Ironic, but not sure. | 18:37 |
*** Marga_ has joined #openstack-ironic | 18:38 | |
*** hemna_ has joined #openstack-ironic | 18:39 | |
NobodyCam | as last update was 12/9/2014 I expect its not copied.. | 18:39 |
jlvillal | NobodyCam: Thank you! | 18:39 |
*** hemna has quit IRC | 18:40 | |
*** bethelwell has quit IRC | 18:41 | |
NobodyCam | jlvillal: I was just looking to see if we had any doc I could point you to | 18:42 |
jlvillal | NobodyCam: Thanks. I'm not quite sure how it is used. I got detoured because I was going to enable flake8 checking for that directory. | 18:43 |
jlvillal | NobodyCam: I only had to fix a few things in that file. | 18:43 |
NobodyCam | the etherpad is still around: https://etherpad.openstack.org/p/ironic-nova-friction | 18:44 |
openstackgerrit | John L. Villalovos proposed openstack/ironic: Enable flake8 checking of ironic/nova/* https://review.openstack.org/191183 | 18:45 |
jlvillal | NobodyCam: ^^^ | 18:46 |
jlvillal | NobodyCam: I will look at the etherpad. Thanks. | 18:47 |
openstackgerrit | John L. Villalovos proposed openstack/ironic: Enable flake8 checking of ironic/nova/* https://review.openstack.org/191183 | 18:48 |
*** mgoddard1 has joined #openstack-ironic | 18:48 | |
*** valkyrja has joined #openstack-ironic | 18:49 | |
*** mgoddard has quit IRC | 18:49 | |
*** mgoddard1 has quit IRC | 18:53 | |
openstackgerrit | John L. Villalovos proposed openstack/ironic: WIP TEST: See what happens if remove ironic/nova https://review.openstack.org/191191 | 19:03 |
jlvillal | NobodyCam: ^^^ Trying something crazy ;) | 19:03 |
* NobodyCam clicks | 19:03 | |
*** hemna__ has joined #openstack-ironic | 19:03 | |
NobodyCam | jlvillal: why? | 19:04 |
*** wuhg has quit IRC | 19:04 | |
jlvillal | NobodyCam: Just curious if it is actually being used. I'm not sure. So I thought I would do a test. | 19:04 |
jlvillal | NobodyCam: py27 tests worked without it... | 19:05 |
*** hemna_ has quit IRC | 19:05 | |
NobodyCam | http://docs.openstack.org/developer/ironic/deploy/install-guide.html#configure-compute-service-to-use-the-bare-metal-service | 19:05 |
jlvillal | NobodyCam: Oh, thanks for that. | 19:06 |
NobodyCam | yes it used | 19:06 |
NobodyCam | mainly by folks with > 1 conductor in the same compute regon | 19:06 |
jlvillal | NobodyCam: Okay, I think I see. | 19:07 |
jlvillal | NobodyCam: Nova calls into it. But Ironic doesn't call it, so that is why the unit tests passed. Thanks. | 19:09 |
jlvillal | NobodyCam: So much to learn! | 19:09 |
NobodyCam | yes that is a out of tree nova driver | 19:09 |
NobodyCam | :-p | 19:09 |
*** dontalton has joined #openstack-ironic | 19:09 | |
*** e0ne has joined #openstack-ironic | 19:09 | |
*** e0ne is now known as e0ne_ | 19:18 | |
jroll | jlvillal: pls don't remove that :( | 19:21 |
jroll | jlvillal: it's an out of tree thing... it's meant so you can run more than one nova-compute host | 19:21 |
jroll | however it gets racy as all hell | 19:21 |
jroll | but it mostly works (we run it) | 19:21 |
jroll | that should be nuked from orbit like this cycle or else, once we figure out the better compute story | 19:22 |
NobodyCam | jroll: +++ | 19:22 |
*** e0ne_ is now known as e0ne | 19:22 | |
*** cdearborn has quit IRC | 19:22 | |
NobodyCam | but I do tend to look at it as an out of tree compute driver :-p ... that may just be me | 19:23 |
*** cdearborn has joined #openstack-ironic | 19:24 | |
*** yog_ has quit IRC | 19:28 | |
*** absubram has quit IRC | 19:37 | |
*** mgoddard has joined #openstack-ironic | 19:52 | |
*** dontalton has quit IRC | 19:52 | |
jlvillal | jroll, I won't remove it! :) I was just doing a test because I didn't know what it was for... | 19:52 |
jroll | gotcha | 20:00 |
*** v92 has quit IRC | 20:07 | |
*** dmellado has quit IRC | 20:07 | |
*** aignatov has quit IRC | 20:07 | |
*** cascone has quit IRC | 20:07 | |
*** Guest19563 is now known as dan | 20:07 | |
*** v92 has joined #openstack-ironic | 20:07 | |
*** dmellado has joined #openstack-ironic | 20:07 | |
*** hemna__ is now known as hemna | 20:09 | |
*** cascone has joined #openstack-ironic | 20:10 | |
*** EmilienM is now known as EmilienM|afk | 20:10 | |
*** aignatov has joined #openstack-ironic | 20:11 | |
*** ifarkas has quit IRC | 20:12 | |
lifeless | jlvillal: its unique | 20:14 |
jlvillal | lifeless: Thank you. | 20:14 |
lifeless | jlvillal: one could argue that it should be in Nova | 20:15 |
lifeless | jlvillal: but that discussion was more cost than benefit back in the split-out-day, so... | 20:15 |
jlvillal | lifeless: My feeling is that probably the Ironic team likes to have it in Ironic. As they can more easily make changes if needed. Though it doesn't seem to change much. | 20:16 |
jlvillal | But I've been known to be wrong :) | 20:16 |
lifeless | :) | 20:17 |
jroll | well, the thing is | 20:17 |
jroll | nova won't accept it :P | 20:17 |
jroll | it's completely broken by their standards, even though it mostly works | 20:17 |
*** Sukhdev has quit IRC | 20:19 | |
*** krtaylor has quit IRC | 20:19 | |
*** e0ne has quit IRC | 20:19 | |
*** natorious is now known as zz_natorious | 20:19 | |
jroll | is anyone here going to the nova midcycle? devananda maybe? | 20:20 |
* jlvillal started to say yes and then realized it was not the Ironic midcycle | 20:25 | |
lifeless | I am, if I get my a into gear and book stuff | 20:27 |
*** greghaynes has quit IRC | 20:27 | |
jroll | heh | 20:27 |
*** alexpilotti has quit IRC | 20:35 | |
*** alexpilotti has joined #openstack-ironic | 20:37 | |
jlvillal | jroll: NobodyCam: Slightly worrying that the Jenkins job passed for the patch removing the ironic/nova/ directory. https://review.openstack.org/191191 | 20:39 |
jroll | jlvillal: right, it's untested | 20:40 |
jroll | or not used in dsvm stuff, I should say | 20:40 |
openstackgerrit | Julia Kreger proposed openstack/bifrost: Support for AMT https://review.openstack.org/191109 | 20:40 |
jlvillal | jroll: Okay. I was thinking it would be tested in one of the tempest jobs. Thanks. | 20:40 |
jroll | jlvillal: nope, our testing story is kind of horrible | 20:40 |
jroll | jlvillal: this is another fun example https://review.openstack.org/#/c/165430/ | 20:41 |
jlvillal | jroll: I'll make a bug about that. Just to trake it. | 20:41 |
jlvillal | s/trake/track/ | 20:41 |
jlvillal | Maybe can figure out how to test it in the future... | 20:41 |
jlvillal | jroll: Uh. That's not good either. | 20:42 |
jroll | yah. | 20:42 |
*** zz_natorious is now known as natorious | 20:47 | |
*** natorious is now known as zz_natorious | 20:48 | |
*** zz_natorious is now known as natorious | 20:49 | |
TheJulia | Anyone fired up the ucs driver recently? | 20:54 |
BadCub | nope | 20:55 |
BadCub | lol | 20:55 |
TheJulia | Looks like the driver is borked | 20:56 |
TheJulia | filing a bug | 20:56 |
*** trown is now known as trown|outttypeww | 21:03 | |
*** davideagnello has quit IRC | 21:08 | |
*** derekh_afk has quit IRC | 21:08 | |
*** davideagnello has joined #openstack-ironic | 21:08 | |
*** krtaylor has joined #openstack-ironic | 21:08 | |
TheJulia | hmm... doesn't actually appear to be the driver... | 21:09 |
*** vishwanathj has quit IRC | 21:13 | |
*** vishwanathj has joined #openstack-ironic | 21:13 | |
openstackgerrit | Julia Kreger proposed openstack/bifrost: Enable agent_ilo driver https://review.openstack.org/191218 | 21:14 |
openstackgerrit | Julia Kreger proposed openstack/bifrost: Adding agent_ucs driver to the driver list https://review.openstack.org/191219 | 21:14 |
*** dguerri` is now known as dguerri | 21:17 | |
spandhe | lucas-beer: hey! yt? | 21:20 |
*** cdearborn has quit IRC | 21:21 | |
*** Sukhdev has joined #openstack-ironic | 21:25 | |
*** zz_jgrimm is now known as jgrimm | 21:33 | |
*** jjohnson2 has quit IRC | 21:34 | |
-openstackstatus- NOTICE: Gerrit will be offline for project renames between 22:00 and 22:30 UTC | 21:43 | |
*** ChanServ changes topic to "Gerrit will be offline for project renames between 22:00 and 22:30 UTC" | 21:43 | |
*** ndipanov has quit IRC | 21:52 | |
*** radek_ has quit IRC | 21:52 | |
-openstackstatus- NOTICE: Gerrit is offline for project renames. ETA 20:30 | 22:04 | |
*** ChanServ changes topic to "Gerrit is offline for project renames. ETA 20:30" | 22:04 | |
*** Marga_ has quit IRC | 22:07 | |
*** Marga_ has joined #openstack-ironic | 22:07 | |
-openstackstatus- NOTICE: Gerrit is offline for project renames. ETA 22:40 | 22:08 | |
jlvillal | Hmmm. Gerrit is down. Sunshine outside. 3pm.... :) | 22:08 |
*** ChanServ changes topic to "Gerrit is offline for project renames. ETA 22:40" | 22:08 | |
TheJulia | lol | 22:11 |
*** Marga_ has quit IRC | 22:12 | |
*** Marga_ has joined #openstack-ironic | 22:12 | |
*** gdmallar has joined #openstack-ironic | 22:30 | |
TheJulia | On a plus side, it is back up and the inspector repo has been renamed | 22:38 |
cinerama | oh yay | 22:41 |
*** ChanServ changes topic to "Gerrit is offline for project renames. ETA 20:30" | 22:42 | |
-openstackstatus- NOTICE: Gerrit is back online. Zuul reconfiguration for renamed projects is still in progress, ETA 23:30. | 22:42 | |
*** ijw has quit IRC | 22:45 | |
*** ChanServ changes topic to "Bare Metal Provisioning | Status: http://bit.ly/ironic-whiteboard | Docs: http://docs.openstack.org/developer/ironic/ | Bugs: https://bugs.launchpad.net/ironic" | 22:50 | |
*** dguerri is now known as dguerri` | 22:54 | |
*** natorious is now known as zz_natorious | 23:05 | |
*** lucas-beer has quit IRC | 23:06 | |
*** rbudden has quit IRC | 23:15 | |
*** hemna has quit IRC | 23:17 | |
*** lucas-beer has joined #openstack-ironic | 23:17 | |
*** rbudden has joined #openstack-ironic | 23:18 | |
*** openstackgerrit has quit IRC | 23:22 | |
*** lucas-beer has quit IRC | 23:22 | |
*** openstackgerrit has joined #openstack-ironic | 23:22 | |
*** davideag_ has joined #openstack-ironic | 23:23 | |
*** Sukhdev has quit IRC | 23:25 | |
*** davideag_ has quit IRC | 23:25 | |
*** davideagnello has quit IRC | 23:27 | |
jlvillal | NobodyCam: I added you as a reviewer to: https://review.openstack.org/#/c/191248/ Updating the text a little for the Ironic meeting. | 23:31 |
*** Marga_ has quit IRC | 23:37 | |
*** Marga_ has joined #openstack-ironic | 23:37 | |
*** Sukhdev has joined #openstack-ironic | 23:38 | |
NobodyCam | hehehe | 23:38 |
NobodyCam | jlvillal: +1'd | 23:39 |
NobodyCam | good catch | 23:39 |
jlvillal | NobodyCam: Thanks :) | 23:39 |
NobodyCam | :) | 23:41 |
*** ijw has joined #openstack-ironic | 23:41 | |
*** Marga_ has quit IRC | 23:47 | |
*** zhenguo has joined #openstack-ironic | 23:48 | |
*** mgoddard has quit IRC | 23:49 | |
*** dan is now known as Guest74409 | 23:49 | |
openstackgerrit | Jeremy Stanley proposed openstack/ironic-inspector: Update .gitreview file for project rename https://review.openstack.org/191271 | 23:50 |
*** dan_ has joined #openstack-ironic | 23:50 | |
*** dan_ is now known as Guest68549 | 23:50 | |
*** bkero has quit IRC | 23:57 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!