| *** ellenh has quit IRC | 00:01 | |
| *** hemna is now known as hemna_ | 00:12 | |
| *** ellenh has joined #openstack-ironic | 00:13 | |
| *** romcheg has joined #openstack-ironic | 00:14 | |
| *** rloo has quit IRC | 00:16 | |
| *** rloo has joined #openstack-ironic | 00:16 | |
| *** romcheg has quit IRC | 00:18 | |
| *** ellenh has quit IRC | 00:19 | |
| *** matsuhashi has joined #openstack-ironic | 00:27 | |
| *** nosnos has joined #openstack-ironic | 00:32 | |
| openstackgerrit | Adam Gandelman proposed a change to openstack/ironic: Cleanup virt driver tests and verify final spawn https://review.openstack.org/99518 | 00:33 |
|---|---|---|
| openstackgerrit | Adam Gandelman proposed a change to openstack/ironic: Destroy instance to clear node state on failure https://review.openstack.org/99519 | 00:33 |
| *** romcheg has joined #openstack-ironic | 00:57 | |
| *** romcheg has quit IRC | 01:02 | |
| *** igordcard has quit IRC | 01:28 | |
| *** rwsu has quit IRC | 01:42 | |
| russell_h | agordeev: hey ping me at some point in the next 4 hours or so if you want me to re-review https://review.openstack.org/#/c/92847/ | 01:44 |
| russell_h | cycle times on this are high because our timezones are totally opposite I think | 01:45 |
| openstackgerrit | lvdongbing proposed a change to openstack/ironic: remove default=None for config options https://review.openstack.org/99532 | 01:51 |
| *** romcheg has joined #openstack-ironic | 01:52 | |
| *** romcheg has quit IRC | 01:56 | |
| *** hemna has joined #openstack-ironic | 02:05 | |
| *** eghobo has quit IRC | 02:36 | |
| *** romcheg has joined #openstack-ironic | 02:45 | |
| *** rloo has quit IRC | 02:48 | |
| *** romcheg has quit IRC | 02:49 | |
| *** coolsvap|afk is now known as coolsvap | 03:04 | |
| *** romcheg has joined #openstack-ironic | 03:13 | |
| *** romcheg has quit IRC | 03:18 | |
| *** hemna has quit IRC | 03:24 | |
| openstackgerrit | Josh Gachnang proposed a change to openstack/ironic: Adding swift temp url support https://review.openstack.org/81391 | 03:32 |
| lifeless | devananda: http://www.blackhat.com/presentations/bh-dc-07/Heasman/Paper/bh-dc-07-Heasman-WP.pdf <- homework :) | 03:37 |
| *** romcheg has joined #openstack-ironic | 03:40 | |
| *** pcrews has quit IRC | 03:40 | |
| openstackgerrit | Josh Gachnang proposed a change to openstack/ironic: Adding swift temp url support https://review.openstack.org/81391 | 03:44 |
| *** romcheg has quit IRC | 03:44 | |
| *** nosnos has quit IRC | 03:44 | |
| *** eghobo has joined #openstack-ironic | 03:47 | |
| *** vinbs has joined #openstack-ironic | 03:49 | |
| *** coolsvap is now known as coolsvap|afk | 04:01 | |
| *** romcheg has joined #openstack-ironic | 04:07 | |
| openstackgerrit | A change was merged to openstack/ironic: Add node id to DEBUG messages in impitool https://review.openstack.org/98928 | 04:08 |
| *** Poornima has joined #openstack-ironic | 04:09 | |
| vinbs | Hi Haomeng :) | 04:10 |
| vinbs | Hello Ironic | 04:11 |
| *** matsuhashi has quit IRC | 04:11 | |
| *** romcheg has quit IRC | 04:11 | |
| *** lazy_prince has joined #openstack-ironic | 04:13 | |
| *** harlowja is now known as harlowja_away | 04:16 | |
| Haomeng | vinbs: hi | 04:18 |
| Haomeng | vinbs: :) | 04:18 |
| vinbs | Haomeng: I'm a bit confused about the pxe boot process that the baremetal node goes through | 04:35 |
| Haomeng | vinbs: welcome :) | 04:35 |
| Haomeng | vinbs: any concern | 04:35 |
| vinbs | Haomeng, I'm using ubuntu server image as the pxe image source but I see the server booting up with cirros! | 04:36 |
| vinbs | Haomeng, is that how it's supposed to happen? | 04:36 |
| Haomeng | vinbs: we will use two image, one is for deployment, another one is user image that will install into the baremetal node | 04:36 |
| Haomeng | Haomeng: so for boot option, we will input a user image which has properties to find the deploy image | 04:37 |
| Haomeng | vinbs: so what your image id is used for nova boot? | 04:38 |
| Haomeng | Haomeng: we can doublecheck by glance command 'glance image-list' or ' nova image-list' to see if the correct image we are using | 04:38 |
| vinbs | so the one would be 'pxe_image_name' and the other would be 'pxe_image_source' that I give in Ironic node-create command? | 04:38 |
| Haomeng | vinbs: from ironic side, these are deploy images I think | 04:40 |
| Haomeng | how dou you launch the ironic booting? | 04:40 |
| Haomeng | with nova boot command? | 04:40 |
| vinbs | Haomeng, I do it through horizon and I use ubuntu server image | 04:41 |
| vinbs | http://paste.openstack.org/show/83744/ | 04:41 |
| Haomeng | looks fine | 04:42 |
| Haomeng | strange | 04:42 |
| vinbs | Haomeng, here's the image list and the command i used for enrolling the node | 04:42 |
| Haomeng | where the cirros from? | 04:42 |
| Haomeng | did you login correct ip? | 04:42 |
| vinbs | Haomeng, yes I have logged in correct IP. I did have cirros image before, but I deleted the cirros image from glance | 04:43 |
| vinbs | Haomeng, and added the ubuntu server image | 04:43 |
| Haomeng | vinbs: maybe that is a bug, ironic cache the image | 04:43 |
| Haomeng | vinbs: can you check it from tftp folder? | 04:43 |
| *** dwalleck has joined #openstack-ironic | 04:44 | |
| Haomeng | vinbs: you can check the ironic log to trace the imag id | 04:45 |
| Haomeng | such as "8614c1a9-7983-4d2a-adae-df04af070460" | 04:45 |
| *** dwalleck has joined #openstack-ironic | 04:45 | |
| vinbs | Haomeng, yes.. let me do it | 04:45 |
| Haomeng | vinbs: :) | 04:45 |
| Haomeng | vinbs: let me check ironic code to see where is the cache folder | 04:46 |
| *** matsuhashi has joined #openstack-ironic | 04:48 | |
| vinbs | Haomeng, and one more thing, when my baremetal boots up with cirros, I get this error "wget: can't connect to remote host (169.254.169.254): Network is unreachable " | 04:48 |
| *** nosnos has joined #openstack-ironic | 04:48 | |
| vinbs | Haomeng, followed by this: cloud-userdata: failed to read instance id | 04:49 |
| vinbs | Haomeng, and then the cirros login screen comes up | 04:49 |
| Haomeng | vinbs: that is not ironic issue, that is cirros image issue i think, you can try to deploy it with kvm directly, we will get same issue I think | 04:49 |
| vinbs | Haomeng, that is what I thought. So booting with a ubuntu or fedora image might resolve this issue right? | 04:50 |
| Haomeng | vinbs: I think we can ignore the 'wget 169.254...' issue | 04:51 |
| vinbs | Haomeng, okay | 04:52 |
| Haomeng | vinbs: :) | 04:52 |
| Haomeng | vinbs: let me help you find the image cache logic | 04:52 |
| Haomeng | vinbs: maybe it is a cache bug | 04:53 |
| vinbs | Haomeng, cool :) | 04:53 |
| vinbs | Haomeng, I searched the ironic logs for the ubuntu image id and I see that there are requests sent with this image id | 04:54 |
| Haomeng | vinbs: that is strage, is it the id which is ubuntu image loaded by glance? | 04:55 |
| openstackgerrit | OpenStack Proposal Bot proposed a change to openstack/ironic: Updated from global requirements https://review.openstack.org/96228 | 04:55 |
| openstackgerrit | OpenStack Proposal Bot proposed a change to openstack/ironic-python-agent: Updated from global requirements https://review.openstack.org/88722 | 04:55 |
| Haomeng | can you try to upload another ubuntu image into glance again? | 04:55 |
| Haomeng | vinbs: or just run nova boot to creaet a vm which is from your ubuntu image? | 04:55 |
| Haomeng | vinbs: to see if it works with nova vm? | 04:56 |
| vinbs | Haomeng, okay.. let me try that :) | 04:56 |
| Haomeng | vinbs: if you make sure the image you are using is correct on, no need to verify with nova vms | 04:56 |
| vinbs | Haomeng, first I'll clear up the existing image from glance and try again by uploading ubuntu image again | 04:57 |
| Haomeng | ok | 04:58 |
| Haomeng | this is our ironic image cache path - CONF.pxe.tftp_master_path | 04:58 |
| Haomeng | vinbs: you can check it, and clean them | 04:59 |
| vinbs | you mean master_images folder in the /tftpboot folder? | 05:01 |
| Haomeng | vinbs: yes, we can find some sub-folders with id as names | 05:02 |
| *** coolsvap|afk is now known as coolsvap | 05:02 | |
| *** k4n0 has joined #openstack-ironic | 05:04 | |
| vinbs | I do see two files with some ids which are different from that of the ubuntu image | 05:05 |
| *** pradipta_away is now known as pradipta | 05:06 | |
| *** eghobo has quit IRC | 05:08 | |
| vinbs | Haomeng, at one point I had deleted the ironic database and set it up again.. that might have been the reason for this mess up with the images? | 05:08 |
| *** eghobo has joined #openstack-ironic | 05:09 | |
| *** rakesh_hs has joined #openstack-ironic | 05:22 | |
| *** pradipta is now known as pradipta_away | 05:24 | |
| *** sysexit has joined #openstack-ironic | 05:32 | |
| Haomeng | vinbs: vinbs: yes, delete them and run again to see if that is using ubuntu image this time | 05:34 |
| *** romcheg has joined #openstack-ironic | 05:37 | |
| *** dwalleck has quit IRC | 05:41 | |
| *** romcheg has quit IRC | 05:42 | |
| vinbs | Haomeng, I deleted the images from cache | 05:56 |
| vinbs | Haomeng, I restarted the services too, now when I launch an instance it's still looking for that image which was present in the cache | 05:57 |
| vinbs | Haomeng, I see an error in nova-compute saying <image-id> image was not found | 05:57 |
| Haomeng | vinbs: strange | 05:59 |
| Haomeng | vinbs: what imageid you are passing to launch nova boot? | 05:59 |
| Haomeng | vinbs: and did you pass the image which used to boot ? | 06:00 |
| Haomeng | vinbs: and check if this image-id is defined in our ironic conf file | 06:00 |
| *** romcheg has joined #openstack-ironic | 06:01 | |
| vinbs | Haomeng, I passed the ubuntu server image-id | 06:01 |
| openstackgerrit | OpenStack Proposal Bot proposed a change to openstack/ironic: Imported Translations from Transifex https://review.openstack.org/96063 | 06:02 |
| vinbs | Haomeng but i haven't tested it with vm nova boot yet | 06:02 |
| Haomeng | vinbs: I under that is new one which we upload by glance this time | 06:02 |
| Haomeng | ok, how do you launch the deployment? | 06:02 |
| vinbs | Haomeng, through Horizon itself | 06:03 |
| Haomeng | vinbs: yes, horizon will call nova api I think | 06:03 |
| Haomeng | vinbs: what case you are testing? for booting vm or ironic baremetel deployment? | 06:04 |
| vinbs | Haomeng, yes, I see the folder with ironic node-id created in the tftpboot directory | 06:04 |
| vinbs | Haomeng, ironic baremetal deployment | 06:05 |
| Haomeng | vinbs: and for this time, what is image-id? | 06:05 |
| Haomeng | vinbs: is it new one you have uploaded image? | 06:05 |
| Haomeng | vinbs: you can run glance image-list to double check | 06:05 |
| *** romcheg has quit IRC | 06:05 | |
| Haomeng | vinbs: and check what image id you defined in your ironic.conf | 06:06 |
| vinbs | Haomeng, ok.. let me check that | 06:06 |
| Haomeng | vinbs: strange | 06:07 |
| vinbs | Haomeng, I haven't defined any image id in ironic.conf | 06:08 |
| Haomeng | vinbs: ok, ignore this | 06:08 |
| *** radsy has quit IRC | 06:09 | |
| vinbs | Haomeng, this is the command i use to add the ubuntu image to glance: glance image-create --name "ubuntu-14.04-server-amd64" --disk-format qcow2 --container-format bare --is-public True --progress < ubuntu-14.04-server-amd64.iso | 06:14 |
| vinbs | Haomeng, do you see anything wrong with that? | 06:14 |
| Haomeng | vinbs: that is fine | 06:14 |
| Haomeng | vinbs: should we update image properties to set the deploy image id used by ironic? | 06:15 |
| vinbs | Haomeng, after this i use disk image builder to create the ramdisk and kernel images | 06:15 |
| Haomeng | vinbs: ok | 06:16 |
| vinbs | Haomeng, you mean like we update deploy kernel id in the image properties of the actual image we are going to boot? | 06:16 |
| Haomeng | vinbs: let me check the documents, I remember we have to associate the deploy ramdisk/kernel id with the user real image by glance image-update command | 06:17 |
| vinbs | Haomeng, yes I do associate the kernel id and ramdisk id with my ubuntu image after adding them to glance | 06:18 |
| Haomeng | vinbs: what documents are you following now? can you try with devstack+ironic - http://docs.openstack.org/developer/ironic/dev/dev-quickstart.html#deploying-ironic-with-devstack | 06:21 |
| Haomeng | vinbs: devstack scripts suppport our ironic, we can use it directy, that is easy to deploy ironic node | 06:21 |
| Haomeng | vinbs: however devstack is using a vm to simulate the baremetal I think | 06:23 |
| vinbs | Haomeng, but devstack is mostly used with virtual baremetal nodes right? | 06:23 |
| vinbs | Haomeng, yes | 06:23 |
| vinbs | Haomeng, but I need to test with real hardware | 06:23 |
| Haomeng | vinbs: yes, but we can take reference for the scripts what actions should be done | 06:23 |
| Haomeng | vinbs: yes | 06:23 |
| vinbs | Haomeng, yes.. let me take a look at those scripts then | 06:24 |
| Haomeng | vinbs: I think only the flavor is different between the virtual or physical baremetal deployments | 06:24 |
| Haomeng | vinbs: and the way to get mac address for baremetal | 06:25 |
| Haomeng | vinbs: I think your env has something issue, but not clear idea how to fix | 06:26 |
| vinbs | Haomeng, I'll try to clear up and reinstall services. Maybe that will resolve the issue | 06:27 |
| Haomeng | vinbs: If I can login your system, I'd like to debug with you together | 06:27 |
| *** romcheg has joined #openstack-ironic | 06:28 | |
| vinbs | Haomeng: it's in a lab setup which doesn't have access to outside internet | 06:28 |
| vinbs | Haomeng, let me check if I can get that access | 06:28 |
| Haomeng | vinbs: ok, so it is diffcult to make progress with irc commnuitation:) | 06:29 |
| Haomeng | vinbs: dont worry | 06:29 |
| vinbs | Haomeng, thanks a lot :) | 06:30 |
| Haomeng | vinbs: :) | 06:30 |
| vinbs | Haomeng, I'll see if I can get the access to the machine for you.. that will be very helpful for me | 06:30 |
| Haomeng | vinbs: this is devstack ironic scripts, that is good reference for us for ironic operations | 06:30 |
| Haomeng | vinbs: ok | 06:31 |
| Haomeng | vinbs: this line - https://github.com/openstack-dev/devstack/blob/master/lib/ironic#L380 | 06:32 |
| Haomeng | vinbs: will set the flavor properties to associate the deploy ramdisk/kernel id | 06:32 |
| *** romcheg has quit IRC | 06:32 | |
| vinbs | Haomeng, yes.. but even after doing this I used to get an error saying kernel_id is not defined | 06:33 |
| vinbs | Haomeng, then I came to know that I need to associate kernel_id and ramdisk_id with the real image | 06:33 |
| Haomeng | vinbs: what log file indicate 'kernel_id' not defined? | 06:33 |
| Haomeng | vinbs: can you share the log with paste.openstack.org | 06:34 |
| vinbs | Haomeng, https://ask.openstack.org/en/question/30707/failed-to-deploy-error-kernel_id/ | 06:34 |
| Haomeng | vinbs: ok, let me check | 06:34 |
| vinbs | Haomeng, so after doing the association that issue resolved | 06:34 |
| Haomeng | vinbs: ? | 06:35 |
| Haomeng | vinbs: you mean your issue is resolved? | 06:35 |
| vinbs | Haomeng, I mean I was getting a kernel_id issue before, I resolved that by associating the kernel_id and ramdisk_id to the boot image | 06:36 |
| vinbs | Haomeng, but the issue I have now is different | 06:36 |
| Haomeng | vinbs: yes your are correct | 06:36 |
| vinbs | Haomeng, i still have this issue where it is looking for a different image | 06:37 |
| Haomeng | vinbs: that is very strange:) | 06:38 |
| Haomeng | vinbs: confusing where the different image from | 06:38 |
| Haomeng | vinbs: that is magic:) | 06:38 |
| vinbs | Haomeng, lol yeah :D | 06:39 |
| vinbs | Haomeng, I grep-ed for that image id in the entire etc folder just to make sure i haven't defined it in any of the config files | 06:40 |
| vinbs | Haomeng, didn't find it anywhere | 06:40 |
| *** dwalleck has joined #openstack-ironic | 06:42 | |
| *** sysexit has quit IRC | 06:44 | |
| *** vinbs is now known as vinbs_|_lunch | 06:45 | |
| *** vinbs_|_lunch is now known as vinbs_lunch | 06:45 | |
| *** dwalleck has quit IRC | 06:46 | |
| *** sysexit has joined #openstack-ironic | 06:47 | |
| openstackgerrit | lvdongbing proposed a change to openstack/ironic: remove default=None for config options https://review.openstack.org/99532 | 07:07 |
| *** eguz has joined #openstack-ironic | 07:11 | |
| *** eguz has quit IRC | 07:11 | |
| *** jcoufal has joined #openstack-ironic | 07:13 | |
| *** eghobo has quit IRC | 07:15 | |
| mrda | devananda: I've been trying but I got at least one reviewer to look at https://review.openstack.org/#/c/95024 :) | 07:19 |
| *** Mikhail_D_ltp has joined #openstack-ironic | 07:23 | |
| *** jcoufal has quit IRC | 07:33 | |
| Mikhail_D_ltp | Good morning folks! :) | 07:34 |
| mrda | Hey Mikhail_D_ltp | 07:40 |
| mrda | and with that... goodnight ironic! | 07:40 |
| *** mrda is now known as mrda-away | 07:40 | |
| *** jcoufal has joined #openstack-ironic | 07:43 | |
| *** dwalleck has joined #openstack-ironic | 07:44 | |
| *** dwalleck has quit IRC | 07:49 | |
| Mikhail_D_ltp | mrda-away: goodnight! :) | 07:51 |
| *** jcoufal has quit IRC | 07:51 | |
| *** ndipanov has joined #openstack-ironic | 07:54 | |
| *** jcoufal has joined #openstack-ironic | 08:00 | |
| *** jistr has joined #openstack-ironic | 08:02 | |
| *** derekh_ has joined #openstack-ironic | 08:11 | |
| *** lucasagomes has joined #openstack-ironic | 08:12 | |
| *** jcoufal has quit IRC | 08:14 | |
| *** romcheg has joined #openstack-ironic | 08:15 | |
| *** Poornima has quit IRC | 08:17 | |
| *** agordeev has quit IRC | 08:17 | |
| *** pelix has joined #openstack-ironic | 08:26 | |
| *** foexle has joined #openstack-ironic | 08:35 | |
| *** datajerk has quit IRC | 08:36 | |
| *** igordcard has joined #openstack-ironic | 08:43 | |
| *** jcoufal has joined #openstack-ironic | 08:43 | |
| *** martyntaylor has joined #openstack-ironic | 08:45 | |
| *** dwalleck has joined #openstack-ironic | 08:46 | |
| *** dwalleck has quit IRC | 08:50 | |
| *** datajerk has joined #openstack-ironic | 08:51 | |
| *** max_lobur has joined #openstack-ironic | 08:55 | |
| *** coolsvap is now known as coolsvap|afk | 08:55 | |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Use GB instead of MB for swap https://review.openstack.org/83788 | 09:01 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Fractional sizes for root_gb and ephemeral_gb https://review.openstack.org/99426 | 09:01 |
| *** coolsvap|afk is now known as coolsvap | 09:06 | |
| openstackgerrit | Mikhail Durnosvistov proposed a change to openstack/ironic: Check whether specified FS is supported https://review.openstack.org/98102 | 09:09 |
| openstackgerrit | Imre Farkas proposed a change to openstack/ironic-specs: More robust device status checking with fuser https://review.openstack.org/99605 | 09:19 |
| *** max_lobur1 has joined #openstack-ironic | 09:32 | |
| *** max_lobur has quit IRC | 09:35 | |
| *** dwalleck has joined #openstack-ironic | 09:46 | |
| *** dwalleck has quit IRC | 09:51 | |
| *** Alexei_987 has joined #openstack-ironic | 09:52 | |
| *** agordeev has joined #openstack-ironic | 10:14 | |
| *** matsuhashi has quit IRC | 10:20 | |
| *** matsuhashi has joined #openstack-ironic | 10:20 | |
| *** romcheg has quit IRC | 10:23 | |
| *** matsuhashi has quit IRC | 10:25 | |
| *** shausy has joined #openstack-ironic | 10:27 | |
| *** vinbs_lunch is now known as vinbs | 10:27 | |
| vinbs | Haomeng, I had to change the image_id in the flavor key too after I changed the image in glance | 10:28 |
| vinbs | Haomeng, it worked fine after that | 10:29 |
| vinbs | Haomeng, so there are basically three places where I need to change the image_id whenever I change the real boot image in glance | 10:29 |
| vinbs | Haomeng, once in ironic node-create command, where I associate the ironic node with the image id | 10:30 |
| vinbs | Haomeng, secondly in the baremetal flavor-key where I mention the kernel and ramdisk id | 10:30 |
| Haomeng | vinbs: cool | 10:31 |
| vinbs | Haomeng, thirdly in need to associate the kernel id and ramdisk id with the boot image using glance image-update | 10:31 |
| Haomeng | vinbs: yes | 10:31 |
| Haomeng | vinbs: I guess you take the devstack ironic scripts as reference and get idea, righ? | 10:32 |
| vinbs | Haomeng, yes :) that and a bit of going over the logs | 10:32 |
| *** rakesh_hs has quit IRC | 10:32 | |
| vinbs | Haomeng, it was a relief to finally see the launch instance move from spawning stage to running stage :D | 10:33 |
| Haomeng | vinbs: :) | 10:33 |
| Haomeng | vinbs: contgs:) | 10:33 |
| Haomeng | vinbs: you get good luck:) | 10:34 |
| vinbs | Haomeng, thanks a lot for the help :) | 10:34 |
| Haomeng | vinbs: you are welcome, and it is that you help yourseff | 10:34 |
| Haomeng | vinbs: :) | 10:34 |
| *** sysexit has quit IRC | 10:40 | |
| lucasagomes | vinbs, the flavor is the deploy kernel & ramdisk | 10:42 |
| lucasagomes | vinbs, not really related to the image u want to deploy | 10:42 |
| *** martyntaylor has quit IRC | 10:43 | |
| *** martyntaylor has joined #openstack-ironic | 10:47 | |
| *** Poornima has joined #openstack-ironic | 10:53 | |
| *** vinbs has quit IRC | 10:56 | |
| *** sysexit has joined #openstack-ironic | 10:56 | |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic-specs: iPXE boot https://review.openstack.org/98822 | 10:58 |
| openstackgerrit | jan grant proposed a change to openstack/ironic: WIP: Serialise paramiko requests. https://review.openstack.org/99626 | 11:05 |
| openstackgerrit | Mikhail Durnosvistov proposed a change to openstack/ironic: Old value 'updated_at' field returned after update https://review.openstack.org/75430 | 11:08 |
| openstackgerrit | Mikhail Durnosvistov proposed a change to openstack/ironic: Get rid object model `dict` methods part 1 https://review.openstack.org/60025 | 11:08 |
| openstackgerrit | David Shrewsbury proposed a change to openstack/ironic: Replace mknod() with chmod() https://review.openstack.org/99484 | 11:08 |
| *** nosnos has quit IRC | 11:13 | |
| openstackgerrit | Mikhail Durnosvistov proposed a change to openstack/ironic: Old value 'updated_at' field returned after update https://review.openstack.org/75430 | 11:14 |
| openstackgerrit | Mikhail Durnosvistov proposed a change to openstack/ironic: Get rid object model `dict` methods part 1 https://review.openstack.org/60025 | 11:14 |
| *** romcheg has joined #openstack-ironic | 11:26 | |
| *** k4n0 has quit IRC | 11:31 | |
| *** Haomeng has quit IRC | 11:33 | |
| *** coolsvap is now known as coolsvap|afk | 11:36 | |
| openstackgerrit | Mikhail Durnosvistov proposed a change to openstack/ironic: Check whether specified FS is supported https://review.openstack.org/98102 | 11:37 |
| openstackgerrit | Mikhail Durnosvistov proposed a change to openstack/ironic: Sync processutils from oslo code https://review.openstack.org/98342 | 11:37 |
| openstackgerrit | Yuriy Zveryanskyy proposed a change to openstack/ironic: Fix utils.execute() for consistency with Oslo code https://review.openstack.org/99636 | 11:47 |
| *** max_lobur1 has quit IRC | 11:51 | |
| *** yuriyz has quit IRC | 12:14 | |
| *** max_lobur has joined #openstack-ironic | 12:15 | |
| *** jdob has joined #openstack-ironic | 12:17 | |
| Shrews | Hmm, we're going to have to drastically rework our gate tests, me thinks | 12:20 |
| Shrews | http://bit.ly/1oRZHUI | 12:23 |
| *** linggao has joined #openstack-ironic | 12:29 | |
| lucasagomes | yeah :( | 12:34 |
| lucasagomes | Shrews, check Sean email | 12:34 |
| lucasagomes | Shrews, and morning :0 | 12:34 |
| lucasagomes | :)* | 12:34 |
| Shrews | lucasagomes: yeah, that's where i got that from | 12:34 |
| Shrews | lucasagomes: morning :) | 12:34 |
| Shrews | lucasagomes: Are you familiar with the "oslotest overlap" Sean mentions in his review to disable that job? | 12:37 |
| Shrews | https://review.openstack.org/99630 | 12:38 |
| lucasagomes | Shrews, not really trying to find out | 12:41 |
| *** coolsvap|afk has quit IRC | 12:46 | |
| *** openstackgerrit has quit IRC | 12:46 | |
| *** rakesh_hs has joined #openstack-ironic | 12:46 | |
| *** coolsvap|afk has joined #openstack-ironic | 12:46 | |
| *** openstackgerrit has joined #openstack-ironic | 12:48 | |
| rushiagr | hello folks | 12:48 |
| rushiagr | I am trying out ironic with servers in my lab | 12:48 |
| rushiagr | they are all HP proliant gen8 servers | 12:49 |
| rushiagr | any idea how to access them from outside via IPMI? | 12:49 |
| rushiagr | username/password/hostname combination via lanplus interface is not working | 12:49 |
| *** yuriyz has joined #openstack-ironic | 12:50 | |
| rushiagr | on the other hand, from inside the system, I am able to access ipmi things | 12:50 |
| rushiagr | the above two methods are tried using ipmitool | 12:51 |
| romcheg | rushiagr: Hi! | 12:51 |
| romcheg | What error do you get? | 12:51 |
| lucasagomes | Shrews, dont know really :( | 12:52 |
| Shrews | rushiagr: IIRC, i read something recently where ipmi over lan can be disabled on hp systems. Not sure where you would check that, though. | 12:53 |
| rushiagr | romcheg: hi! | 12:54 |
| *** rloo has joined #openstack-ironic | 12:54 | |
| * Shrews been toying with the idea of buying a proliant microserver for playing | 12:54 | |
| rushiagr | romcheg: I can access the ILO web interface | 12:54 |
| rushiagr | romcheg: the error is | 12:55 |
| romcheg | yuriyz: Did you enable anything in the web interface? | 12:55 |
| rushiagr | t: http://csrc.nist.gov/groups/ | 12:55 |
| rushiagr | STM/cmvp/documents/140-1/140-1val.zip | 12:55 |
| yuriyz | https://srv10-kha-ipmi.kha.mirantis.net/ | 12:55 |
| rushiagr | oops, sory | 12:55 |
| yuriyz | engineer | 12:55 |
| rushiagr | Error: Unable to establish IPMI v2 / RMCP+ session | 12:55 |
| romcheg | yuriyz: Stop right now, this is not a private window! | 12:55 |
| romcheg | :) | 12:56 |
| boris-42 | romcheg yuriyz lol | 12:56 |
| *** Poornima has quit IRC | 12:56 | |
| rushiagr | I was trying "ipmitool -I lanplus -U <username> -P <password> -H xx.xx.xx.xx sdr list | 12:57 |
| rushiagr | I found one document which says that only KCS interface is enabled | 12:57 |
| rushiagr | does that mean only after logging into the system I can access the interface? I need to access it via outside of that machine | 12:58 |
| rushiagr | romcheg: ^ | 12:58 |
| romcheg | rushiagr: http://cl.ly/image/3k1k3b2q2L30 could you check please if that is enabled? | 12:58 |
| romcheg | It's in the iLO web interface Administration->Access Settings | 12:59 |
| *** agordeev has quit IRC | 13:00 | |
| *** agordeev has joined #openstack-ironic | 13:01 | |
| rushiagr | romcheg: yes, this is enabled | 13:01 |
| romcheg | rushiagr: hmm, let me think | 13:03 |
| rushiagr | romcheg: unfortunately, there isnt much info provided in the Web UI with regards to IPMI :( | 13:04 |
| rushiagr | I can log into the machine and do stuff using ipmitool, to extract more information from the machine, if it is needed | 13:05 |
| yuriyz | rushiagr, do you have admin IPMI access? Try add -L OPERATOR otherwise | 13:05 |
| rushiagr | yuriyz: I have admin access | 13:08 |
| NobodyCam | good morning IRonic | 13:11 |
| * rushiagr will be afk for an hour | 13:12 | |
| yuriyz | rushiagr, your command syntax OK, maybe a problem with network access | 13:12 |
| *** matty_dubs|gone is now known as matty_dubs | 13:12 | |
| openstackgerrit | Chris Krelle proposed a change to openstack/ironic: Let ipmitool natively retry commands https://review.openstack.org/96902 | 13:30 |
| openstackgerrit | Imre Farkas proposed a change to openstack/ironic: Fix workaround for the "device is busy" problem https://review.openstack.org/90126 | 13:37 |
| openstackgerrit | Chris Krelle proposed a change to openstack/ironic: Enforce a minimum time between all IPMI commands https://review.openstack.org/96558 | 13:39 |
| openstackgerrit | Imre Farkas proposed a change to openstack/ironic-specs: More robust device status checking with fuser https://review.openstack.org/99605 | 13:42 |
| openstackgerrit | Jim Rollenhagen proposed a change to openstack/ironic-python-agent: Use processutils properly https://review.openstack.org/99665 | 13:50 |
| openstackgerrit | Jim Rollenhagen proposed a change to openstack/ironic-python-agent: Better errors for execute() failures https://review.openstack.org/99666 | 13:50 |
| openstackgerrit | Jim Rollenhagen proposed a change to openstack/ironic-python-agent: Better errors for execute() failures https://review.openstack.org/99666 | 13:52 |
| openstackgerrit | Jim Rollenhagen proposed a change to openstack/ironic-python-agent: Use processutils properly https://review.openstack.org/91202 | 13:52 |
| jroll | morning y'all | 13:52 |
| NobodyCam | good morning jroll | 13:54 |
| *** reaper has joined #openstack-ironic | 13:56 | |
| *** jistr has quit IRC | 14:00 | |
| *** jistr has joined #openstack-ironic | 14:02 | |
| *** pcrews has joined #openstack-ironic | 14:10 | |
| *** romcheg has quit IRC | 14:14 | |
| lucasagomes | morning NobodyCam jroll yuriyz romcheg | 14:16 |
| NobodyCam | morning lucasagomes | 14:16 |
| *** rloo has quit IRC | 14:27 | |
| *** jcoufal has quit IRC | 14:27 | |
| *** rloo has joined #openstack-ironic | 14:28 | |
| *** dwalleck has joined #openstack-ironic | 14:29 | |
| *** dwalleck has joined #openstack-ironic | 14:29 | |
| *** rloo has quit IRC | 14:29 | |
| *** rloo has joined #openstack-ironic | 14:30 | |
| *** Shiv has joined #openstack-ironic | 14:31 | |
| *** Shiv has quit IRC | 14:33 | |
| *** coolsvap|afk is now known as coolsvap | 14:33 | |
| BadCub | morning everyone | 14:34 |
| *** Shiv has joined #openstack-ironic | 14:34 | |
| rushiagr | my servers are within a VPN. Can it be due to some blocked network protocol? I can see that though I can not ping the iLO IP, I can SSH into it | 14:34 |
| rushiagr | what protocol does IPMI uses? UDP? | 14:34 |
| *** rloo has quit IRC | 14:34 | |
| *** romcheg has joined #openstack-ironic | 14:35 | |
| *** rloo has joined #openstack-ironic | 14:35 | |
| *** lazy_prince has quit IRC | 14:37 | |
| NobodyCam | morning BadCub | 14:41 |
| romcheg | Morning NobodyCam! | 14:42 |
| NobodyCam | morning romcheg | 14:43 |
| NobodyCam | rushiagr: yes ..port 623 / udp | 14:43 |
| lucasagomes | jroll, ping, mind if I rebase Josh's PXE patch? | 14:44 |
| *** mkerrin has quit IRC | 14:44 | |
| NobodyCam | romcheg: I see that Russia admitted it is supporting separatists | 14:45 |
| *** martyntaylor has quit IRC | 14:47 | |
| romcheg | NobodyCam: I can see 3 Russian tanks in Ukraine http://www.youtube.com/watch?feature=player_detailpage&v=fsZPyM9nNFw#t=37 :( | 14:47 |
| rushiagr | NobodyCam: okay, thanks | 14:48 |
| *** martyntaylor has joined #openstack-ironic | 14:49 | |
| rushiagr | I am seeing that when I execute 'ipmitool lan print' from inside that machine, the 'Auth Type Support' is blank | 14:49 |
| *** coolsvap is now known as coolsvap|afk | 14:49 | |
| Shrews | So it would seem that my rebuild tempest test has random failures. After the initial boot, it verifies ephemeral partition is mounted, touches a file on that partition, then a REBUILD is issued. After the rebuild, sometimes the touched file is not there. Could this be a filesystem sync issue? | 14:49 |
| NobodyCam | romcheg: oh wow... I glad I'm not the person in the little car :-p | 14:50 |
| * Shrews thought "sync; sync; sync" was no longer necessary | 14:51 | |
| rushiagr | paste.openstack.org/show/83818 | 14:52 |
| rushiagr | here is the output | 14:52 |
| rloo | Shrews: the test has the flag to preserve ephemeral? | 14:54 |
| *** dwalleck has quit IRC | 14:54 | |
| Shrews | rloo: yes. that's what it is testing. and i verify it is mounted before and after rebuild | 14:55 |
| NobodyCam | mornign rloo | 14:55 |
| rloo | hello everyone, NobodyCam! | 14:55 |
| Shrews | rloo: if you're curious: https://review.openstack.org/94439 | 14:55 |
| Shrews | rloo: and morning | 14:56 |
| *** jistr has quit IRC | 14:56 | |
| rloo | Shrews: I'm curious but that would mean I'd need to learn about tempest. Which I will one day, just not ready to do it today ;) | 14:56 |
| *** jistr has joined #openstack-ironic | 14:56 | |
| Shrews | rloo: lol... i don't blame you | 14:57 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Factoring out PXE and TFTP functions https://review.openstack.org/90233 | 14:57 |
| lucasagomes | jroll, rebased ^ | 14:57 |
| *** mdorman has joined #openstack-ironic | 14:58 | |
| rushiagr | any idea if the above is related to closed port? | 15:00 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Add iPXE support for Ironic https://review.openstack.org/99318 | 15:00 |
| *** max_lobur has quit IRC | 15:02 | |
| NobodyCam | question I was just looking at the ipmi specs, any know if there is a standard for key name attributes like boot_device.. and if so where can I find sucha list | 15:03 |
| *** rakesh_hs has quit IRC | 15:03 | |
| NobodyCam | s/any/anyone/ | 15:04 |
| Shrews | must be hard question day | 15:07 |
| *** coolsvap|afk is now known as coolsvap | 15:08 | |
| *** hemna has joined #openstack-ironic | 15:13 | |
| *** shausy has quit IRC | 15:13 | |
| *** foexle has quit IRC | 15:17 | |
| jroll | lucasagomes: morning! I'm never opposed to people rebasing things for me :) | 15:19 |
| jroll | lucasagomes: were there conflicts? | 15:19 |
| lucasagomes | jroll, heh cool, just rebased that minutes ago | 15:20 |
| lucasagomes | pxe.py | 15:20 |
| lucasagomes | and at the neutron tests | 15:20 |
| *** Mikhail_D_ltp has quit IRC | 15:20 | |
| devananda | g'morning, all | 15:20 |
| * devananda caches up on scrollback | 15:20 | |
| lucasagomes | devananda, morning | 15:20 |
| lucasagomes | jroll, got to change something at the test_tftp.py as well due "TypeError: acquire() got an unexpected keyword argument 'node_ids'" | 15:21 |
| openstackgerrit | jan grant proposed a change to openstack/ironic: WIP: Serialise paramiko requests. https://review.openstack.org/99626 | 15:23 |
| jroll | lucasagomes: ok, I'll double check when I get to the office | 15:23 |
| jroll | thanks again :) | 15:23 |
| lucasagomes | np thank u | 15:23 |
| devananda | actually, gotta step away again... bbi20m | 15:23 |
| *** Mikhail_D_ltp has joined #openstack-ironic | 15:24 | |
| NobodyCam | good morning devananda | 15:25 |
| jroll | morning devananda | 15:25 |
| lucasagomes | if someone wants to review something: https://review.openstack.org/90233 :) would be nice to get it merged soon | 15:28 |
| *** Kai14 has joined #openstack-ironic | 15:29 | |
| *** Shiv has quit IRC | 15:30 | |
| *** Shiv has joined #openstack-ironic | 15:31 | |
| *** Mikhail_D_ltp has quit IRC | 15:31 | |
| *** sysexit has quit IRC | 15:32 | |
| *** Shiv has quit IRC | 15:33 | |
| *** Mikhail_D_ltp has joined #openstack-ironic | 15:33 | |
| * devananda adds ^ to the 'pad | 15:37 | |
| *** Mikhail_D_ltp has quit IRC | 15:37 | |
| *** Mikhail_D_ltp has joined #openstack-ironic | 15:37 | |
| JayF | I bet Josh would love to come back from vacation to see #90233 merged :) | 15:38 |
| Shrews | devananda: If you haven't seen it yet, there's an email from Sean where he explains why he made our gate tests non-voting again. | 15:39 |
| devananda | .... | 15:39 |
| devananda | Shrews: thanks | 15:39 |
| *** Mikhail_D_ltp has quit IRC | 15:39 | |
| lucasagomes | JayF, +1 | 15:40 |
| lucasagomes | JayF, I tested it locally here, now it works fine | 15:40 |
| JayF | well it failed tempest after the rebase :x | 15:41 |
| devananda | ah, i see | 15:41 |
| *** dwalleck has joined #openstack-ironic | 15:41 | |
| lucasagomes | lemme recheck that | 15:41 |
| JayF | yeah I just read that email and it seems relevant | 15:42 |
| lucasagomes | devananda, btw, check Sean's email in the openstack-dev | 15:42 |
| lucasagomes | yeah like Shrews said... sorry didn't see that message | 15:42 |
| * lucasagomes is a spammer :P | 15:42 | |
| Shrews | http://img2.wikia.nocookie.net/__cb20110629041027/icarly/images/4/49/Monty_python_spam.jpg | 15:44 |
| matty_dubs | lucasagomes: Hey, did you see the email Sean sent to openstack-dev? | 15:44 |
| matty_dubs | ;) | 15:44 |
| lucasagomes | :P | 15:44 |
| *** eghobo has joined #openstack-ironic | 15:45 | |
| * NobodyCam thinks: lets get some real spam goning.... now where did i put that power MTA linc..... | 15:47 | |
| JayF | Hey, I've worked with email marketing before. I may or may not have benchmarked mail systems that sent mail at a 6k messages/second :) (not that it's useful to send mail that fast... just fun) | 15:48 |
| * NobodyCam may or may not have also worked in email marketing too | 15:49 | |
| NobodyCam | power MTA/pytohn injectors/ 10k m/s | 15:50 |
| NobodyCam | per server | 15:50 |
| JayF | heh | 15:50 |
| JayF | mine was just a chrooted postfix, with the chroot living entirely on a ramdisk :) | 15:50 |
| NobodyCam | lol so long as you have enough domanins to send to | 15:50 |
| * BadCub cringes at the memories of email marketing days | 15:50 | |
| *** shiv has joined #openstack-ironic | 15:51 | |
| NobodyCam | JayF: oh not that was something I never thought of back in the day | 15:51 |
| NobodyCam | s/not/now/ | 15:52 |
| NobodyCam | chrroting a ramdisk | 15:52 |
| BadCub | NobodyCam: yu were too busy keeping the nation's largest spammer operational. No time to think LOL | 15:52 |
| NobodyCam | shhhhh | 15:52 |
| JayF | Heh. I quit my email marketing job because someone paid their way out of being shutdown for abuse :( | 15:53 |
| *** matty_dubs is now known as matty_dubs|gone | 15:53 | |
| * BadCub quit after too many depositions | 15:53 | |
| BadCub | I worked in teh legal department for the nation's largest spammer. er, affiliate network LOL | 15:54 |
| NobodyCam | JayF: ahh you remember http://www.danhatesspam.com | 15:55 |
| BadCub | Gods I hated Dan... | 15:55 |
| JayF | I don't know anything about that guy | 15:55 |
| NobodyCam | lol | 15:55 |
| JayF | I remember the woman who was Postmaster at AOL | 15:55 |
| JayF | and basically she'd ask us to jump and we'd ask how high | 15:56 |
| JayF | because she could just flip off all delivery from your domain to aol on a whim | 15:56 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Use GB instead of MB for swap https://review.openstack.org/83788 | 15:56 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Fractional sizes for root_gb and ephemeral_gb https://review.openstack.org/99426 | 15:56 |
| BadCub | He was the largest spam litigater in the USA for years. Sued everyone for everything | 15:56 |
| JayF | and that was the way she got your attention. Do something wrong? Get tons of bounce messages. | 15:56 |
| lucasagomes | devananda, btw about https://review.openstack.org/#/c/98120/ , are we holding it ? | 15:56 |
| * NobodyCam recalls all the back off and delivery rules he wrote... | 15:57 | |
| lucasagomes | or it just because of the gate problems? | 15:57 |
| * BadCub recalls all the C&Ds he received on a daily basis | 15:57 | |
| *** rakesh_hs has joined #openstack-ironic | 15:58 | |
| NobodyCam | :-p | 15:59 |
| NobodyCam | bbt.... brb | 15:59 |
| devananda | lucasagomes: did you see nova has the same db migration issue we do | 16:00 |
| lucasagomes | devananda, oh thought they were fine since they use the old code | 16:01 |
| *** shiv has quit IRC | 16:01 | |
| lucasagomes | and didn't migrate to alembic yet | 16:01 |
| devananda | lucasagomes: it's not alembic that's the cause, it's the oslo-incubator db migration code | 16:01 |
| devananda | see https://bugs.launchpad.net/nova/+bug/1328997 | 16:02 |
| * lucasagomes clicks | 16:02 | |
| devananda | adam_g: around? | 16:02 |
| lucasagomes | ah good and now that it's marked as critical I suspect it's going to be fixed soon :) | 16:03 |
| devananda | see the lengthy discussion? yea | 16:07 |
| *** sysexit has joined #openstack-ironic | 16:07 | |
| devananda | gate bug in nova gets fixes fast these days | 16:07 |
| devananda | we need to track that and, if they dont, port the fix to ironic | 16:07 |
| devananda | our db migration test code is very similar | 16:07 |
| lucasagomes | +1 will keep an eye on that | 16:07 |
| *** linggao has quit IRC | 16:11 | |
| devananda | J1 milestone tag done! | 16:14 |
| devananda | https://launchpad.net/ironic/juno/juno-1 | 16:14 |
| devananda | 62 bugs fixed -- great work, everyone! | 16:14 |
| *** ndipanov has quit IRC | 16:14 | |
| rloo | Yay! | 16:16 |
| Shrews | this one... i no like. i has a sad :( https://bugs.launchpad.net/ironic/+bug/1199522 | 16:16 |
| romcheg | Morning devananda! | 16:16 |
| *** ellenh has joined #openstack-ironic | 16:16 | |
| romcheg | Thierry got to Launchpad… | 16:17 |
| rloo | Shrews: why sad? | 16:17 |
| Shrews | rloo: breaks unit tests on macs | 16:17 |
| rloo | Shrews: oh, so sorry :-( | 16:18 |
| Shrews | which is not important to anyone other than myself and NobodyCam :) | 16:18 |
| rloo | Shrews: don't you have a bug open to fix unit tests on macs? | 16:18 |
| Shrews | rloo: yes | 16:18 |
| rloo | Shrews: what's important to you and NobodyCam is important to the rest of us. Just not as important ;) | 16:18 |
| NobodyCam | ditto----Great job everyone!!!! | 16:19 |
| lucasagomes | nice | 16:19 |
| NobodyCam | lol | 16:19 |
| *** jistr has quit IRC | 16:19 | |
| * NobodyCam thinks everyone should have a mac | 16:19 | |
| openstackgerrit | Jay Faulkner proposed a change to openstack/ironic-python-agent: Fix incorrect log statement https://review.openstack.org/99706 | 16:21 |
| lucasagomes | NobodyCam, oh please hah no | 16:21 |
| NobodyCam | :-p | 16:22 |
| JayF | ^ that is an easy/cheap review (https://review.openstack.org/#/c/99706/) | 16:22 |
| NobodyCam | JayF: yep easy -1, I want to say switching format to : 5S | 16:25 |
| NobodyCam | lol | 16:25 |
| NobodyCam | j/k | 16:25 |
| JayF | NobodyCam: yep, you're an openstack dev /s | 16:26 |
| NobodyCam | :-p | 16:29 |
| *** sysexit has quit IRC | 16:29 | |
| jroll | devananda: nice :) | 16:31 |
| *** Mikhail_D_ltp has joined #openstack-ironic | 16:34 | |
| *** romcheg has quit IRC | 16:35 | |
| *** rloo has quit IRC | 16:36 | |
| *** rloo has joined #openstack-ironic | 16:37 | |
| *** openstackgerrit has quit IRC | 16:38 | |
| *** sysexit has joined #openstack-ironic | 16:39 | |
| *** Kai14 has quit IRC | 16:44 | |
| *** mkerrin has joined #openstack-ironic | 16:53 | |
| *** harlowja_away is now known as harlowja | 16:54 | |
| devananda | adam_g: do you happen to have the ovs/libvirt bug reference handy? (the reason we're installing UCA in devstack-gate) | 16:55 |
| *** jcoufal has joined #openstack-ironic | 16:58 | |
| *** rwsu has joined #openstack-ironic | 17:00 | |
| NobodyCam | devananda: this does not seem right to me... I seem to recall a bug about it.. do you happen to recall which # it is? | 17:01 |
| NobodyCam | http://logs.openstack.org/02/96902/13/check/check-tempest-dsvm-ironic-postgres/af373f4/logs/screen-ir-api.txt.gz | 17:02 |
| devananda | NobodyCam: what's the issue? | 17:03 |
| devananda | I dont see any errors in there | 17:03 |
| NobodyCam | ahh just warns | 17:03 |
| *** eghobo has quit IRC | 17:04 | |
| devananda | huh | 17:04 |
| devananda | but a whole lot of tempest failures | 17:04 |
| *** eghobo has joined #openstack-ironic | 17:04 | |
| devananda | wtf | 17:04 |
| devananda | so we disabled the fake driver by default | 17:04 |
| devananda | but devstack is supposed to be adding it | 17:05 |
| NobodyCam | :( i'm looking for the root cause of: http://logs.openstack.org/02/96902/13/check/check-tempest-dsvm-ironic-postgres/af373f4/console.html | 17:05 |
| *** Mikhail_D_ltp has quit IRC | 17:05 | |
| *** Nisha has joined #openstack-ironic | 17:06 | |
| *** Mikhail_D_ltp has joined #openstack-ironic | 17:08 | |
| lucasagomes | alright, it's time to me to go home, came to the office today | 17:08 |
| lucasagomes | have a good night everybody! | 17:08 |
| NobodyCam | oh conductor blew up | 17:08 |
| NobodyCam | night lucasagomes | 17:08 |
| NobodyCam | just hadent gotten to that log yet | 17:08 |
| *** lucasagomes has quit IRC | 17:09 | |
| devananda | NobodyCam: yea. that's reasonable. | 17:13 |
| devananda | oh, hah | 17:14 |
| NobodyCam | ya it did what it should have | 17:14 |
| devananda | so, right. conductor is refusing to start because the pxe_ipmitool driver is enabled, but there's no ipmitool on the system | 17:14 |
| devananda | exactly | 17:14 |
| devananda | so we need to disable pxe_ipmitool in devstack-gate runs | 17:14 |
| *** derekh_ has quit IRC | 17:14 | |
| * NobodyCam is thinking default should be disable ALL driver and force operators to enable only what they have | 17:15 | |
| * devananda hacks up a patch | 17:16 | |
| *** Mikhail_D_ltp has quit IRC | 17:21 | |
| devananda | https://review.openstack.org/99718 | 17:22 |
| devananda | so there's two ways | 17:24 |
| *** Mikhail_D_ltp has joined #openstack-ironic | 17:24 | |
| NobodyCam | fyi +1'd that patch | 17:26 |
| devananda | i'm chatting with infra about it actually | 17:27 |
| devananda | may abandon it ... | 17:27 |
| NobodyCam | ya I would lean toward only default being fake and all other driver must be enabled | 17:27 |
| devananda | see conversation in infra | 17:28 |
| * NobodyCam runs out of the room in to -infra | 17:28 | |
| *** rwsu has quit IRC | 17:30 | |
| devananda | any redhat users -- what's the name of the ipmi package on RH/Fedora? "ipmitool" ? | 17:30 |
| *** shausy has joined #openstack-ironic | 17:31 | |
| *** rwsu has joined #openstack-ironic | 17:32 | |
| NobodyCam | looks like ipmitool | 17:32 |
| *** martyntaylor has left #openstack-ironic | 17:32 | |
| devananda | rpm.pbone.net says "yes" | 17:32 |
| jroll | who feels like reviewing some specs? :) https://review.openstack.org/#/c/98506/ https://review.openstack.org/#/c/98930/ https://review.openstack.org/#/c/99235/ | 17:34 |
| jroll | one of those is a nova spec for the ironic virt driver | 17:34 |
| jroll | but would also like ironic eyes on it | 17:34 |
| adam_g | devananda, here now | 17:35 |
| *** Mikhail_D_ltp has quit IRC | 17:35 | |
| devananda | adam_g: hi! have a link to the LP bug for why we're pulling in UCA packages? | 17:36 |
| adam_g | devananda, i do not. it was never a bug, but a feature shortcoming of the older libvirt shipped with precise. | 17:38 |
| devananda | adam_g: ah, i thought there was something on LP tracking that | 17:38 |
| adam_g | devananda, there were/are a bunch of bugs about issues with the precise libvirt package. | 17:39 |
| *** dwalleck_ has joined #openstack-ironic | 17:39 | |
| adam_g | specifically its libvirt support for OVS virtual ports that we need for the devstack stuff | 17:39 |
| *** Nisha has quit IRC | 17:39 | |
| devananda | adam_g: it might be useful to reference one or more of those in the devstack-gate.yaml file | 17:39 |
| *** Mikhail_D_ltp has joined #openstack-ironic | 17:40 | |
| devananda | adam_g: so that folks undersatnd why we're pulling UCA in | 17:40 |
| *** dwalleck_ has quit IRC | 17:40 | |
| devananda | adam_g: until infra has support for 14.04, we're probably going to continue to need to pull in UCA - -but not in the way we are today | 17:40 |
| *** dwalleck_ has joined #openstack-ironic | 17:40 | |
| *** dwalleck has quit IRC | 17:41 | |
| adam_g | devananda, there is a comment there now about the OVS support. the bugs i mention actually dont affect us, we pull in a newer package where they've all been fixed. :) | 17:41 |
| devananda | adam_g: there's been a fairly lengthy discussion in infra, between fungi mordred and myself, about precaching these packages | 17:41 |
| devananda | adam_g: AIUI, we pull i nthe newer package from UCA precisely to get that fix | 17:41 |
| adam_g | devananda, speaking of UCA, the GPG errors still seem to be issues with the key package--it seems the fix that merged the other day hasn't synced out to all slaves yet | 17:41 |
| devananda | adam_g: that's what i mean -- linking to those bugs would inform folks /why/ we're using UCA | 17:41 |
| devananda | adam_g: right. fungi just poked jenkins02, which was the only one that was still reporting that failure | 17:42 |
| devananda | and fwiw, that's essentially why sdague made our tempest job non-voting this morning :( | 17:42 |
| devananda | i'm trying to wrap that up and get it voting again | 17:42 |
| adam_g | devananda, the *only* reason we are using the UCA package is for the libvirt OVS support.. the bugs i speak of all are all things that were uncovered long ago in other areas of testing and aren't really relevant. bug im happy to clarify in the comments | 17:43 |
| adam_g | and file a bug just to track "we need 14.04" | 17:43 |
| devananda | adam_g: thanks | 17:44 |
| devananda | it's really about the tracking / communication of why we're doing this | 17:44 |
| adam_g | sure | 17:44 |
| *** Mikhail_D_ltp has quit IRC | 17:45 | |
| devananda | NobodyCam: https://review.openstack.org/99723 | 17:46 |
| rloo | devananda: are we OK to approve things? (I don't see anything in the topic but thought I'd check) | 17:49 |
| *** Mikhail_D_ltp has joined #openstack-ironic | 17:49 | |
| devananda | rloo: there's still a note on the etherpad :) | 17:50 |
| devananda | rloo: we should be somewhat conservative still | 17:50 |
| devananda | gate queue is > 24 hours long | 17:50 |
| rloo | devananda: the IronicReviewDay etherpad? I'd prefer if you updated the topic. too many places to think to look and not sure which has been updated. | 17:51 |
| rloo | devananda: ok, i'll just review a few things then. | 17:52 |
| devananda | rloo: the whiteboard,a ctually | 17:52 |
| rloo | devananda: ok, I see it there. Is that where we should be looking then, not the topic? | 17:52 |
| devananda | rloo: i did change the topi. it now says "Status: ## link ##" | 17:53 |
| devananda | since several folks have complained that putting a long status message in the room topic goes unnoticed | 17:53 |
| rloo | NobodyCam: your https://review.openstack.org/#/c/93133 hasn't merged yet :-( | 17:53 |
| *** Mikhail_D_ltp has quit IRC | 17:53 | |
| devananda | i'm open to ideas, but it seems that i can't please everyone on this | 17:53 |
| rloo | devananda: ha ha. ok, i was used to seeing the status there. No worries, it doesn't matter to me. Just that it changed. So whiteboard is fine. Also means that others can update it, which we couldn't with the topic. | 17:54 |
| *** Mikhail_D_ltp has joined #openstack-ironic | 17:55 | |
| NobodyCam | devananda: queue > 24 hours not < ... | 17:57 |
| NobodyCam | would that mean we shoud approve everything | 17:57 |
| *** Mikhail_D_ltp has quit IRC | 17:58 | |
| NobodyCam | oh nm I miss read | 17:58 |
| NobodyCam | :-p | 17:58 |
| *** zul has quit IRC | 18:03 | |
| *** pelix has quit IRC | 18:03 | |
| adam_g | devananda, reading -infra backscroll. monty's suggestion should work. will give it a shot now | 18:08 |
| *** Mikhail_D_ltp has joined #openstack-ironic | 18:09 | |
| *** jdob has quit IRC | 18:12 | |
| *** jdob has joined #openstack-ironic | 18:13 | |
| *** openstackgerrit has joined #openstack-ironic | 18:13 | |
| *** jcoufal has quit IRC | 18:17 | |
| *** shausy has quit IRC | 18:19 | |
| openstackgerrit | Ellen Hui proposed a change to openstack/ironic: Add more INFO logging to ironic/common https://review.openstack.org/99495 | 18:22 |
| NobodyCam | brb | 18:24 |
| *** jdob has quit IRC | 18:25 | |
| *** jdob has joined #openstack-ironic | 18:25 | |
| *** bandicot has joined #openstack-ironic | 18:27 | |
| *** notarealperson has joined #openstack-ironic | 18:36 | |
| NobodyCam | oh its kinda chilly out there today | 18:37 |
| *** mdorman has quit IRC | 18:45 | |
| *** mdorman has joined #openstack-ironic | 18:47 | |
| *** Mikhail_D_ltp has quit IRC | 18:47 | |
| * NobodyCam expects flood, starts filling snadbags | 18:47 | |
| NobodyCam | lol | 18:47 |
| *** ellenh has quit IRC | 18:47 | |
| adam_g | devananda, https://bugs.launchpad.net/openstack-ci/+bug/1329430 feel free to tag as an ironic bug if you see fit | 18:52 |
| *** dwalleck_ has quit IRC | 18:55 | |
| Shrews | adam_g: when you were working on the tempest basic ops test, did you ever see this: http://logs.openstack.org/39/94439/6/check/check-tempest-dsvm-virtual-ironic-nv/e999951/console.html#_2014-06-12_16_34_49_003 | 18:55 |
| Shrews | adam_g: not sure why that's failing | 18:55 |
| *** dwalleck has joined #openstack-ironic | 18:55 | |
| NobodyCam | oh thats a new one for me | 18:56 |
| *** rakesh_hs has quit IRC | 18:57 | |
| Shrews | it's weird b/c the advance ops test passes (now that I sync the filesystem) | 18:58 |
| *** max_lobur has joined #openstack-ironic | 18:58 | |
| *** dwalleck_ has joined #openstack-ironic | 18:59 | |
| *** dwalleck has quit IRC | 18:59 | |
| adam_g | Shrews, hmm. that would mean nova scheduling failed to cast to the compute node, or ironic failed to associate the instance with node? | 19:00 |
| *** Alexei_987 has quit IRC | 19:00 | |
| Shrews | i have yet to see it pass | 19:00 |
| adam_g | Shrews, is this running witht he other advanced test? | 19:00 |
| Shrews | adam_g: yep. advance runs before it and passes | 19:00 |
| adam_g | Shrews, nova probably hasns't reclaimed the free resources from ironic after the previous instance terminated | 19:01 |
| adam_g | and can't schedule because the periodic task hasns't run yet | 19:01 |
| Shrews | adam_g: I lost you at "Shrews" | 19:02 |
| *** sysexit has quit IRC | 19:02 | |
| adam_g | Shrews, so you're running with only a single ironic node. the first test spawns the instance, tests, then terminates it. while the instance is running, n-cpu reports ironic has 0 free resources, they're all in use | 19:03 |
| adam_g | the test finishes and deletes the instance | 19:03 |
| Shrews | oh. i assumed there were more ironic nodes | 19:03 |
| adam_g | but until the next n-cpu periodic task runs to update available resources, nova still thinks there are 0 resources availbale | 19:03 |
| Shrews | adam_g: is there something i need to do in the tests to fix that? | 19:04 |
| adam_g | Shrews, IRONIC_VM_COUNT in devstack can be adjusted to spawn more, but i dont know how well the slave nodes would handle multiples | 19:04 |
| Shrews | oh geez | 19:04 |
| adam_g | Shrews, you can adjust the either the boot or terminate instance methods in tempest to wait until resources are available before boot / resources get freed after terminate? | 19:04 |
| *** eguz has joined #openstack-ironic | 19:05 | |
| adam_g | via the nova hypervisor-stats call | 19:05 |
| *** sysexit has joined #openstack-ironic | 19:05 | |
| adam_g | or during the test setup have it poll hypervisor-stats until the resources reported in the flavor ref are available in nova | 19:05 |
| Shrews | adam_g: ok. that sounds like the better path. thanks | 19:06 |
| adam_g | Shrews, tripleo does something similar. may want to add a new configurable timeout for that | 19:06 |
| *** eguz has quit IRC | 19:06 | |
| *** eghobo has quit IRC | 19:08 | |
| *** romcheg has joined #openstack-ironic | 19:14 | |
| *** sseago has quit IRC | 19:14 | |
| *** ellenh has joined #openstack-ironic | 19:20 | |
| *** sysexit has quit IRC | 19:24 | |
| *** dwalleck_ has quit IRC | 19:26 | |
| *** dwalleck has joined #openstack-ironic | 19:26 | |
| *** dwalleck has quit IRC | 19:26 | |
| *** romcheg has quit IRC | 19:28 | |
| *** sseago has joined #openstack-ironic | 19:30 | |
| openstackgerrit | Ellen Hui proposed a change to openstack/ironic: Add more INFO logging to ironic/conductor https://review.openstack.org/99744 | 19:39 |
| *** romcheg has joined #openstack-ironic | 19:53 | |
| *** bandicot has quit IRC | 19:54 | |
| *** romcheg has quit IRC | 20:00 | |
| *** Mikhail_D_ltp has joined #openstack-ironic | 20:04 | |
| NobodyCam | brb | 20:06 |
| *** bandicot has joined #openstack-ironic | 20:14 | |
| *** bandicot has quit IRC | 20:14 | |
| *** boris-42 has quit IRC | 20:18 | |
| *** romcheg has joined #openstack-ironic | 20:18 | |
| *** boris-42 has joined #openstack-ironic | 20:18 | |
| *** matty_dubs|gone is now known as matty_dubs | 20:27 | |
| NobodyCam | oh did we get our own change queue! sweet | 20:32 |
| *** dwalleck has joined #openstack-ironic | 20:32 | |
| openstackgerrit | A change was merged to openstack/ironic: Set instance default_ephemeral_device https://review.openstack.org/96498 | 20:34 |
| *** dwalleck has quit IRC | 20:34 | |
| jroll | orly | 20:34 |
| *** romcheg1 has joined #openstack-ironic | 20:34 | |
| NobodyCam | :) | 20:34 |
| jroll | amazing | 20:34 |
| jroll | we can do work again | 20:34 |
| jroll | land all the things! \o | 20:34 |
| NobodyCam | woo hoo our jobs have a chance of landing now | 20:34 |
| *** openstackgerrit has quit IRC | 20:35 | |
| *** romcheg has quit IRC | 20:36 | |
| NobodyCam | lol /me was starting to think the gate was test until it fails... :p | 20:36 |
| *** openstackgerrit has joined #openstack-ironic | 20:36 | |
| NobodyCam | oh fish feeding time... brb | 20:36 |
| *** dwalleck has joined #openstack-ironic | 20:36 | |
| *** romcheg1 has quit IRC | 20:38 | |
| *** dwalleck has quit IRC | 20:41 | |
| NobodyCam | ok back | 20:42 |
| openstackgerrit | Ruby Loo proposed a change to openstack/ironic: Use .png instead of .gif images https://review.openstack.org/99759 | 20:45 |
| * devananda wanders away for a bit | 20:46 | |
| *** dwalleck has joined #openstack-ironic | 20:53 | |
| *** dwalleck has quit IRC | 20:54 | |
| NobodyCam | anyone have time to review a devstack patch.... https://review.openstack.org/#/c/99723/ | 20:54 |
| *** dwalleck has joined #openstack-ironic | 20:55 | |
| rloo | NobodyCam: I have a bit of time now, but it is devstack :-) Will take a look... | 20:56 |
| NobodyCam | rloo: We'll need it!! | 20:59 |
| *** jdob has quit IRC | 20:59 | |
| rloo | NobodyCam: done! | 21:02 |
| openstackgerrit | OpenStack Proposal Bot proposed a change to openstack/ironic: Updated from global requirements https://review.openstack.org/96228 | 21:02 |
| NobodyCam | woo hoo :) | 21:02 |
| NobodyCam | Ty rloo | 21:02 |
| * NobodyCam thinks in about 5 minutes three patches will land al in a row! | 21:04 | |
| rloo | NobodyCam: we still shouldn't approve anything right? | 21:05 |
| jroll | rloo: it looks like it might be ok now | 21:05 |
| jroll | we have our own queue now | 21:05 |
| rloo | we do? How did we score that? | 21:06 |
| jroll | not a clue | 21:06 |
| rloo | and how did you know that we have our own queue? | 21:06 |
| NobodyCam | looks like we now have a queue with DIB :) | 21:06 |
| jroll | rloo: http://status.openstack.org/zuul/ | 21:07 |
| jroll | bottom center | 21:07 |
| NobodyCam | there are only three job in it | 21:07 |
| jroll | (also NobodyCam said so) | 21:07 |
| NobodyCam | lol | 21:07 |
| rloo | so the queue runs ironic/dib changes, but won't be affected by 'issues' in any of the other changes in the other gate/queue? | 21:08 |
| jroll | I think that's the idea | 21:08 |
| NobodyCam | we should ask in infra.. I am only going off the zuul page | 21:08 |
| rloo | hmm. Well, I think I won't approve until I see how long it takes for the three in the gate, to get through. | 21:09 |
| openstackgerrit | A change was merged to openstack/ironic: Stop ipmitool.validate from touching the BMC https://review.openstack.org/99121 | 21:10 |
| openstackgerrit | A change was merged to openstack/ironic: remove default=None for config options https://review.openstack.org/99532 | 21:10 |
| jroll | welp, that was quick :) | 21:10 |
| openstackgerrit | A change was merged to openstack/ironic: Wipe any metadata from a nodes disk https://review.openstack.org/93133 | 21:10 |
| jroll | rloo: +A away :D | 21:10 |
| rloo | ha ha | 21:11 |
| NobodyCam | try one | 21:11 |
| NobodyCam | first | 21:11 |
| NobodyCam | :-p | 21:11 |
| NobodyCam | :) | 21:11 |
| Shrews | omg. 93133 was merged | 21:11 |
| NobodyCam | yep | 21:12 |
| Shrews | awww, i was hoping to see it reach 100 patch sets | 21:12 |
| NobodyCam | lol ... /me wounder if we should declare a holiday | 21:12 |
| NobodyCam | lol... you just wanted the dbl party | 21:12 |
| rloo | I think we have to wait for 100 for a holiday. | 21:13 |
| *** eghobo has joined #openstack-ironic | 21:14 | |
| NobodyCam | fungi's offical answer: 21:13 | fungi > NobodyCam: is anything here ever permanent? it's a result of taking oslo cross-tests off those projects for the gate pipeline, which caused them to no longer have jobs in common with anything in the main integrated gate queue | 21:14 |
| jroll | oh :( | 21:17 |
| NobodyCam | yea :/ | 21:18 |
| jroll | so let's land all the things that are ready now | 21:19 |
| jroll | while we have it | 21:19 |
| openstackgerrit | Ruby Loo proposed a change to openstack/ironic: Driver interface's validate should return nothing https://review.openstack.org/97855 | 21:29 |
| *** petertoft has joined #openstack-ironic | 21:30 | |
| rloo | I had to rebase that ^^, and now all the +2's are gone :-( | 21:30 |
| * jroll would +2 if he could | 21:31 | |
| rloo | thx jroll. I've seen others sometimes approve but not sure I should do it to my own. Oh well. | 21:32 |
| jroll | /shrug | 21:32 |
| JayF | I personally would not +2 or +A my own code | 21:32 |
| jroll | there's some cases where I would, but I'm evil | 21:33 |
| jroll | (I actually wouldn't in an openstack project, but yeah) | 21:33 |
| petertoft | Just to note that — Ironic is live in HP's Helion OpenStack Distribution: https://ask.openstack.org/en/question/32113/announcing-hp-helion-openstack-community-edition-ga/. Give it a try if you have a few servers, we're looking for experiences and feedback. | 21:34 |
| jroll | like, this is a review where the 2x +2 thing is silly: https://review.openstack.org/#/c/99706/ | 21:34 |
| jroll | petertoft: interesting! | 21:34 |
| openstackgerrit | Adam Gandelman proposed a change to openstack/ironic-specs: External event callback API https://review.openstack.org/99770 | 21:34 |
| jroll | petertoft: is ironic just being used in a triple-o fashion there, I guess? | 21:35 |
| petertoft | Yep. | 21:35 |
| jroll | cool :) | 21:35 |
| petertoft | Cutting edge stuff. Thanks for the efforts of this team. | 21:35 |
| *** Mikhail_D_ltp has quit IRC | 21:36 | |
| NobodyCam | hey hey petertoft welcome to the iRonic channel | 21:37 |
| NobodyCam | thats awesome news | 21:38 |
| Shrews | petertoft: send a few moonshots to my house and i'd be thrilled to test it :) | 21:38 |
| NobodyCam | hehehe | 21:38 |
| matty_dubs | Same here! ;) | 21:39 |
| petertoft | Yeh, just send me your postal address. They go via regular mail :) | 21:39 |
| NobodyCam | thou I could make a good case for the moonshot dev tower thing-y I've tale of | 21:39 |
| *** max_lobur1 has joined #openstack-ironic | 21:40 | |
| rloo | is JoshNang around? | 21:42 |
| *** max_lobur has quit IRC | 21:42 | |
| Shrews | NobodyCam: oh? small chassis? | 21:43 |
| jroll | rloo: he's on PTO | 21:43 |
| jroll | rloo: I can cover for him in the meantime :) | 21:43 |
| rloo | jroll: ok thx. Would you please look at my comment in https://review.openstack.org/#/c/98922/ | 21:43 |
| rloo | jroll: he's got a lot of +1, so not sure... | 21:44 |
| jroll | rloo: hmm, good point | 21:44 |
| * jroll wonders what other projects do | 21:44 | |
| jroll | rloo: my thinking is: if auth_strategy was 'foobar', this patch behaves the same as before this patch, yes? | 21:46 |
| jroll | (since that was never checked before) | 21:47 |
| rloo | jroll: well, there was no setting. it uses keystone. | 21:47 |
| rloo | jroll: am looking through other confs to see if there are similar cases. | 21:47 |
| jroll | ah, right. | 21:47 |
| jroll | same, I checked nova and they just have 'if auth_strategy == keystone' | 21:48 |
| jroll | neutronclient at some point raised an exception if it was anything but keystone | 21:48 |
| jroll | noauth is not well supported in openstack | 21:48 |
| NobodyCam | Shrews: ya.. I hear it tower case size with four slots | 21:49 |
| NobodyCam | s/it/it is/ | 21:50 |
| Shrews | NobodyCam: Must. Have. Now. | 21:50 |
| rloo | jroll: so we have an 'auth_strategy=keystone' in ironic.api.app. It just checks if == keystone. | 21:51 |
| jroll | hmm | 21:51 |
| jroll | rloo: would you prefer to refactor it that way? | 21:52 |
| rloo | I don't know. | 21:52 |
| jroll | idk either. | 21:53 |
| jroll | :P | 21:53 |
| rloo | jroll: so if you specify 'neutron', you want to do that stuff. but if you specified 'foobar', doing that will end up err'ing. but if we treat foobar as noauth, there is a security risk there. | 21:53 |
| jroll | right | 21:53 |
| rloo | jroll: so maybe i open another bug about it, and +2 this? | 21:53 |
| rloo | jroll: since I guess we have a similar issue with keystone. | 21:54 |
| jroll | yeah, that's fine | 21:54 |
| rloo | jroll: or I do nothing about it and go away for the weekend... | 21:54 |
| jroll | also, if we treat foobar as noauth here, there's no risk because this is the client | 21:54 |
| jroll | the security risk is in setting a server as noauth | 21:54 |
| jroll | if we accidentally use noauth here, we just aren't sending credentials | 21:54 |
| jroll | noooooooooo don't go :P | 21:54 |
| rloo | jroll: the help msg says "noauth is insecure and should only be used for testing" | 21:55 |
| jroll | ugh | 21:55 |
| jroll | running neutron in noauth mode is insecure | 21:55 |
| jroll | forgetting to send credentials when talking to neutron is not insecure | 21:55 |
| * jroll wants to change that help message | 21:56 | |
| rloo | jroll: please change it then. | 21:56 |
| jroll | I will | 21:56 |
| jroll | wondering if that should say anything about security | 21:56 |
| rloo | jroll: look at the commit msg too. | 21:56 |
| jroll | like 'running neutron in noauth mode is insecure' | 21:56 |
| jroll | yeah, I saw that. not as bad because it's not presented to a deployer. | 21:56 |
| rloo | jroll: exactly. | 21:57 |
| jroll | grr | 21:57 |
| jroll | rloo: so, anything you'd like touched up in the actual code, since I'm uploading a new one anyway? | 21:57 |
| rloo | jroll: so I was somewhat serious about taking off for the weekend. I have to go in a few minutes (when the dinner bell rings), and I'm off after that. I might look in later this evening though. | 21:57 |
| jroll | no worries, don't worry about checking in on me :) | 21:58 |
| rloo | jroll: if you change, i guess it'll need 2 +2s anyway. | 21:58 |
| jroll | but if you want me to add something like 'elif auth == 'keystone'' do it now | 21:58 |
| jroll | yeah | 21:58 |
| rloo | jroll: I think, to be consistent with the other keystone one, I'd say do something like if CONF.neutron.auth_strategy != 'keystone'. | 21:59 |
| jroll | hmm, I was thinking `elif CONF.neutron.auth_strategy == 'keystone' and context.auth_token is None:` | 22:00 |
| jroll | on line 64 | 22:00 |
| rloo | jroll: you could do that, but then what if it is foobar? | 22:00 |
| jroll | then it uses auth_strategy of None | 22:00 |
| rloo | oh, then the last else kicks in. | 22:01 |
| jroll | (who knows what that does, but hey) | 22:01 |
| jroll | I do know that is not the same as noauth | 22:01 |
| *** mrda-away is now known as mrda | 22:01 | |
| mrda | Morning ironic! | 22:01 |
| jroll | \o mrda | 22:02 |
| NobodyCam | morning mrda | 22:02 |
| mrda | o/ | 22:02 |
| rloo | jroll: to be honest, I don't know what the diff is between 'noauth' and None. I think it might not work if auth_token is None (the last else part). | 22:03 |
| openstackgerrit | Jim Rollenhagen proposed a change to openstack/ironic: Allow noauth for Neutron https://review.openstack.org/98922 | 22:03 |
| jroll | rloo: see how that looks ^ | 22:03 |
| jroll | rloo: it's unclear to me, too, but it seems to me that they are different | 22:04 |
| jroll | bah, forgot to generate sample config | 22:04 |
| rloo | jroll: I suspect/guess that if it is foobar and context.auth_token is None, that won't work. So it'll fail later on. Which may be fine, not sure. | 22:05 |
| rloo | jroll: dinner bell has rung. Gotta go. thx :-) | 22:06 |
| jroll | probably. but like, same behavior as before :) | 22:06 |
| jroll | cheers, have a good weekend! | 22:06 |
| jroll | (I wish mine started right now) | 22:06 |
| NobodyCam | night rloo | 22:13 |
| *** matty_dubs is now known as matty_dubs|gone | 22:21 | |
| NobodyCam | lol watching jenkins is very much like watching grass grow | 22:22 |
| mrda | NobodyCam: https://plus.google.com/111308284400884743702/posts/UBHf9VBfTvS | 22:23 |
| NobodyCam | lol | 22:24 |
| *** petertoft has quit IRC | 22:26 | |
| *** radsy has joined #openstack-ironic | 22:34 | |
| *** radsy has joined #openstack-ironic | 22:34 | |
| *** dwalleck has quit IRC | 22:39 | |
| *** dwalleck has joined #openstack-ironic | 22:42 | |
| *** dwalleck has quit IRC | 22:51 | |
| NobodyCam | devananda: around? | 23:07 |
| devananda | NobodyCam: ya | 23:08 |
| NobodyCam | just a off the wall question (from watching jenkins)... :-p .... why do we have https://github.com/openstack/ironic/blob/master/requirements.txt#L30 when its not enabled by default? | 23:09 |
| NobodyCam | takes forever to build ... lol | 23:10 |
| jroll | pyghmi takes forever to build? | 23:10 |
| jroll | or ironic? | 23:10 |
| NobodyCam | pyghmi ... I'm watching a check-tripleo-ironic-undercloud-precise run atm | 23:11 |
| NobodyCam | and it builds it for seed, undercloud | 23:11 |
| NobodyCam | just seem like a waste of testing cycles | 23:12 |
| jroll | "pure python implementation of IPMI protocol" what is it building | 23:12 |
| jroll | oh, requires pycrypto | 23:13 |
| jroll | but it's not used | 23:13 |
| * jroll sends a patch | 23:13 | |
| devananda | NobodyCam: I have a patch up that removes it ... got -1'd ... :p | 23:14 |
| NobodyCam | from us or the pycrypto | 23:15 |
| devananda | NobodyCam: https://review.openstack.org/#/c/92819/2 | 23:15 |
| devananda | NobodyCam: feel free to rebase (there's a conflict now) and update as needed | 23:15 |
| jroll | devananda: nobody -1'd it except for lucas and yourself? | 23:16 |
| jroll | and you fixed lucas' issue with it | 23:16 |
| NobodyCam | ya but I agree with "I think we should remove pyghmi from requirements..." | 23:17 |
| NobodyCam | if you wish it us it install it | 23:17 |
| jroll | bah, pycrypto is used, I forgot it's imported from Crypto | 23:18 |
| NobodyCam | we've started to fail drivers on load now, like seamicro, ipmitool | 23:18 |
| NobodyCam | until we replace ipmitool with pyghmi I dont see a need to have it in the requirments file | 23:20 |
| jroll | +1 | 23:21 |
| devananda | right | 23:22 |
| devananda | so I think failing drivers on load that are not usable is, well, NOT doing that was terrible | 23:23 |
| devananda | so yes | 23:23 |
| openstackgerrit | Adam Gandelman proposed a change to openstack/ironic: Destroy instance to clear node state on failure https://review.openstack.org/99519 | 23:24 |
| openstackgerrit | Adam Gandelman proposed a change to openstack/ironic: Cleanup virt driver tests and verify final spawn https://review.openstack.org/99518 | 23:24 |
| devananda | adam_g: why "node.provision_state = ironic_states.ACTIVE" in nova/tests/virt/irnoic/test_driver? | 23:30 |
| adam_g | devananda, hmm. copy/paste from an existing spawn test. | 23:31 |
| devananda | adam_g: heh | 23:32 |
| adam_g | devananda, not sure why that is there to begin with. | 23:32 |
| *** mdorman has quit IRC | 23:34 | |
| openstackgerrit | Chris Krelle proposed a change to openstack/ironic: Mock pyghmi lib in unit tests if not present https://review.openstack.org/92819 | 23:37 |
| jroll | is patching sys.modules really the best way to accomplish this? | 23:39 |
| jroll | devananda: ^ | 23:40 |
| NobodyCam | oh wait I think I boo boo'ed | 23:48 |
| lifeless | mock.mock! | 23:49 |
| openstackgerrit | Adam Gandelman proposed a change to openstack/ironic: Destroy instance to clear node state on failure https://review.openstack.org/99519 | 23:49 |
| openstackgerrit | Adam Gandelman proposed a change to openstack/ironic: Cleanup virt driver tests and verify final spawn https://review.openstack.org/99518 | 23:49 |
| openstackgerrit | Chris Krelle proposed a change to openstack/ironic: Mock pyghmi lib in unit tests if not present https://review.openstack.org/92819 | 23:51 |
| devananda | jroll: yes | 23:54 |
| jroll | hrm | 23:54 |
| jroll | I don't have a better way in mind, but seems like there must be something | 23:54 |
| devananda | jroll: this is better than what i originally thought of | 23:54 |
| devananda | jroll: take a look at the old review if you want. i think there was some discussion | 23:55 |
| russell_h | adam_g: this callback API is interesting | 23:55 |
| jroll | devananda: ok | 23:55 |
| russell_h | adam_g: I'm thinking about if we could replace a lot of our use of vendor_passthru with that | 23:55 |
| devananda | lifeless: debug/verbose=True is the default now in tripleo? | 23:55 |
| lifeless | no | 23:55 |
| russell_h | adam_g: (both in the PXE driver where there is basically a "ok, now deploy" call, and in the agent driver where we have even more callbacks) | 23:56 |
| lifeless | devananda: why? | 23:56 |
| devananda | lifeless: noticing your patch. i had though tripleo explicitly didn't want debug=true to be default | 23:56 |
| lifeless | devananda: its not | 23:56 |
| lifeless | devananda: that patch doesn't make it default | 23:57 |
| devananda | oh | 23:57 |
| adam_g | russell_h, yeah. im in favor of adding more potential use cases to the spec. a quick look at the reference agent driver showed a lot more neutron usage that would be prone to similar races we've already seen | 23:57 |
| lifeless | that patch adds debug=True if the heat metadata ironic.debug is truthy | 23:57 |
| lifeless | likewise for verbose | 23:57 |
| lifeless | and otherwise leaves the setting absent from the file entirely | 23:58 |
| russell_h | adam_g: yes | 23:58 |
| *** igordcard has quit IRC | 23:58 | |
| jroll | adam_g: would love for neutron to call back | 23:58 |
| adam_g | russell_h, im also wondering if it should model the nova os-external-events so closely or diverge a bit and also track events in the context of not just instance, but nodes as well | 23:58 |
| jroll | that's a pain point for us today | 23:58 |
| devananda | lifeless: ack. i fail at reading :) | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
