| *** Madasi has joined #openstack-ironic | 00:01 | |
| * NobodyCam open frezzer to find several http://www.totalwine.com/_static/webupload/730/1_119555865_3.jpg ... NOTE: answers after this point are not guaranteed correct | 00:04 | |
| jroll | NobodyCam: cool, ty | 00:05 |
|---|---|---|
| jroll | heh | 00:05 |
| NobodyCam | it is after 5 here :) | 00:06 |
| devananda | jroll: so, that 'pxe_*' is in the nova virt driver is bad. but E_HISTORY and E_NOTIME. All of that should move to node.instance_info in Juno | 00:08 |
| devananda | it'll be a fun game of backwards compatibility for a little while, though | 00:08 |
| Shrews | So, after rebuilding my python venv environment, running the py27 tests causes subunit to take 100% cpu and slowly eat all of the memory | 00:08 |
| Shrews | so this is fun | 00:09 |
| Shrews | lifeless: ever seen that? ^^^ i don't think it's specifically testr, but who knows | 00:09 |
| jroll | devananda: yeah, heard - as I said, I'm going to be working on this stuff next | 00:10 |
| jroll | JoshNang: see what Shrews said :o | 00:10 |
| jroll | Shrews: josh saw this yesterday | 00:10 |
| jroll | slash today? | 00:10 |
| Shrews | orly? | 00:10 |
| lifeless | Shrews: subunit takes that? | 00:11 |
| lifeless | Shrews: whats the exact command line in question ? | 00:11 |
| Shrews | lifeless: ps output was: ubuntu 6722 97.6 34.1 1542308 1384244 pts/1 R+ 20:03 4:29 python -m subunit.run discover -t ./ ./ --load-list /tmp/tmpQhUWEa | 00:11 |
| Shrews | 34% of 8G of memory before i killed it | 00:12 |
| lifeless | Shrews: ok so thats the test runner | 00:12 |
| lifeless | Shrews: no reason to believe subunit is the problem :) | 00:12 |
| JoshNang | Shrews: ah! | 00:12 |
| JoshNang | i had this same issue | 00:12 |
| JoshNang | no idea why | 00:12 |
| Shrews | well that's concerning | 00:13 |
| JoshNang | i was working on making a merge for our agent driver and kept running into it randomly. I gave up, figuring it was my code, and I'd come back. | 00:13 |
| Shrews | i removed the .testrepository directory and ran it again. this time it worked as expected | 00:17 |
| JoshNang | hmmm | 00:17 |
| Shrews | not sure if that was dumb luck or related | 00:17 |
| Shrews | lifeless: ^^^^ | 00:17 |
| JoshNang | i had it come back a few times after i thought i "fixed" it. it'd run fine for a couple runs then go crazy | 00:17 |
| NobodyCam | Shrews: are there any circular directory link in play? | 00:20 |
| NobodyCam | s/link/links/ | 00:20 |
| Shrews | NobodyCam: within the .testrepository directory? don't think so. certainly not now that i've removed it :) | 00:21 |
| NobodyCam | :) | 00:21 |
| openstackgerrit | David Shrewsbury proposed a change to openstack/ironic: Encapsulate Ironic client retry logic https://review.openstack.org/83105 | 00:22 |
| lifeless | Shrews: that directory isn't accessed from that process | 00:24 |
| lifeless | Shrews: so its totally unrelated :) | 00:24 |
| lifeless | Shrews: in theory anyhow | 00:24 |
| Shrews | hrm, certainly odd | 00:25 |
| NobodyCam | Shrews: line 88 of client_wrapper could we do a Dynamic Looping Call thing there vs time.sleep | 00:26 |
| openstackgerrit | A change was merged to openstack/ironic: Replace fixtures with mock in test_keystone.py https://review.openstack.org/79702 | 00:26 |
| Shrews | NobodyCam: there's no way (that i saw, at least) to be able to return a return value from the method being called doing it that way. I initially implemented it that way. | 00:27 |
| Shrews | calls like "cli.node.list" need to return a value | 00:27 |
| Shrews | NobodyCam: assuming you're refering to the way the current _retry_if_service_is_unavailable() does it with FixedIntervalLoopingCall | 00:29 |
| *** matsuhashi has joined #openstack-ironic | 00:31 | |
| NobodyCam | I was thinking of a dynamic type but fixed would be better | 00:32 |
| Shrews | NobodyCam: i just left a comment on the review about it | 00:39 |
| Shrews | ok, done doing battle with assorted python issues that had nothing to do with the work i *actually* wanted to accomplish. see you all tomorrow (yay Friday's!) | 00:47 |
| mordred | Shrews: you LOVE python | 00:48 |
| Shrews | mordred: you love java | 00:48 |
| mordred | Shrews: wanna start rewriting opensatck in C++ yet? | 00:48 |
| Shrews | i'll start on that tomorrow | 00:49 |
| openstackgerrit | A change was merged to openstack/ironic: Process public API list as regular expressions https://review.openstack.org/66621 | 00:49 |
| *** derekh has joined #openstack-ironic | 00:57 | |
| *** dguerri has quit IRC | 01:02 | |
| *** dguerri has joined #openstack-ironic | 01:05 | |
| JayF | I just filed https://review.openstack.org/#/c/83638/ to get ironic-python-agent added to projects.txt in the Openstack requirements repo. Please drop a note in there if it shouldn't be there. Thanks! | 01:10 |
| *** eghobo has joined #openstack-ironic | 01:27 | |
| *** derekh has quit IRC | 01:32 | |
| openstackgerrit | Josh Gachnang proposed a change to openstack/ironic-python-agent: Adding DynamicLoopingCall around lookup https://review.openstack.org/83604 | 01:34 |
| *** nosnos has joined #openstack-ironic | 01:37 | |
| *** hemna_ has quit IRC | 02:09 | |
| *** eghobo has quit IRC | 03:01 | |
| *** matsuhashi has quit IRC | 03:04 | |
| *** packet has joined #openstack-ironic | 03:07 | |
| *** matsuhashi has joined #openstack-ironic | 03:10 | |
| *** nosnos has quit IRC | 03:17 | |
| *** harlowja is now known as harlowja_away | 03:19 | |
| *** rloo has quit IRC | 03:20 | |
| *** matsuhashi has quit IRC | 03:31 | |
| *** eghobo has joined #openstack-ironic | 03:35 | |
| *** matsuhashi has joined #openstack-ironic | 03:53 | |
| *** nosnos has joined #openstack-ironic | 04:06 | |
| *** killer_prince has joined #openstack-ironic | 04:09 | |
| *** vkozhukalov has joined #openstack-ironic | 04:15 | |
| *** rameshg87 has joined #openstack-ironic | 04:34 | |
| *** pradipta_away is now known as pradipta | 04:38 | |
| *** packet has quit IRC | 04:43 | |
| *** Shrews has quit IRC | 04:51 | |
| *** Shrews has joined #openstack-ironic | 04:52 | |
| *** loki_ has joined #openstack-ironic | 05:00 | |
| *** killer_prince has quit IRC | 05:43 | |
| *** mrda is now known as mrda_weekend | 06:06 | |
| openstackgerrit | Jenkins proposed a change to openstack/ironic: Imported Translations from Transifex https://review.openstack.org/82993 | 06:07 |
| *** vkozhukalov has quit IRC | 06:15 | |
| *** vkozhukalov1 has joined #openstack-ironic | 06:15 | |
| *** eghobo has quit IRC | 06:27 | |
| *** epim has joined #openstack-ironic | 06:34 | |
| *** dividehex has quit IRC | 06:37 | |
| *** epim has quit IRC | 06:37 | |
| *** saju_m has joined #openstack-ironic | 06:38 | |
| *** dividehex has joined #openstack-ironic | 06:42 | |
| *** vkozhukalov1 has quit IRC | 06:49 | |
| *** killer_prince has joined #openstack-ironic | 06:56 | |
| *** Manishanker has joined #openstack-ironic | 07:06 | |
| *** mdurnosvistov_ has joined #openstack-ironic | 07:15 | |
| *** matsuhashi has quit IRC | 07:20 | |
| *** matsuhas_ has joined #openstack-ironic | 07:24 | |
| GheRivero | morning all | 07:41 |
| *** athomas has joined #openstack-ironic | 08:05 | |
| *** mdurnosvistov_ has quit IRC | 08:13 | |
| *** vkozhukalov has joined #openstack-ironic | 08:16 | |
| *** ndipanov has joined #openstack-ironic | 08:22 | |
| *** lazy_prince has quit IRC | 08:29 | |
| *** jistr has joined #openstack-ironic | 08:33 | |
| *** lazy_prince has joined #openstack-ironic | 08:35 | |
| agordeev | GheRivero: morning :) | 08:37 |
| *** Haomeng has quit IRC | 08:46 | |
| *** 77CAADEUV has joined #openstack-ironic | 08:46 | |
| *** 77CAADEUV has quit IRC | 08:47 | |
| *** ifarkas has joined #openstack-ironic | 08:48 | |
| *** jistr has quit IRC | 08:49 | |
| *** matsuhas_ has quit IRC | 08:53 | |
| *** matsuhas_ has joined #openstack-ironic | 08:57 | |
| *** jistr has joined #openstack-ironic | 08:59 | |
| *** saju_m has quit IRC | 09:05 | |
| *** derekh has joined #openstack-ironic | 09:09 | |
| *** pbrooko has joined #openstack-ironic | 09:19 | |
| *** saju_m has joined #openstack-ironic | 09:25 | |
| mdurnosvistov | Good morning Ironic! :) | 09:25 |
| dtantsur | Morning! | 09:27 |
| *** lucasagomes has joined #openstack-ironic | 09:28 | |
| agordeev | mdurnosvistov dtantsur morning! | 09:40 |
| *** mdurnosvistov has quit IRC | 09:40 | |
| *** max_lobur has joined #openstack-ironic | 09:43 | |
| *** max_lobur1 has joined #openstack-ironic | 09:43 | |
| *** max_lobur has quit IRC | 09:47 | |
| *** sabah has joined #openstack-ironic | 09:56 | |
| *** matsuhas_ has quit IRC | 10:00 | |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Add DiskPartitioner https://review.openstack.org/83396 | 10:04 |
| killer_prince | Can anyone point me to the work being done related to UEFI in ironic and where can I find more details related to it. | 10:07 |
| lucasagomes | killer_prince, I don't think there's much working right now on the UEFI part | 10:10 |
| *** rameshg87 has quit IRC | 10:10 | |
| lucasagomes | killer_prince, I'm doing this plumbing for the GPT paritioning and will implement GPT soon | 10:10 |
| *** romcheg has joined #openstack-ironic | 10:10 | |
| *** rameshg87 has joined #openstack-ironic | 10:11 | |
| lucasagomes | but that's it for me now, but I don't think there's anyone _right now_ working on things like security boot | 10:11 |
| killer_prince | aha... so when GPT disks are created, are you also providing support for creating efi boot partition when disk is GPT formatted..? | 10:12 |
| killer_prince | in any case i would like to get involved in the GPT and EFI related work. if there is any thing with which I can help you, please let me know.. | 10:14 |
| *** nosnos has quit IRC | 10:16 | |
| lucasagomes | killer_prince, right, this would be nice to create an efi boot partition, which would be quite simply with the plumbing work. But as a inital work I would simply support creating GPT partitions | 10:24 |
| lucasagomes | mostly because we want support for partitions bigger than 2TB | 10:24 |
| openstackgerrit | A change was merged to openstack/ironic: Fix seamicro validate() method definition https://review.openstack.org/83598 | 10:24 |
| lucasagomes | killer_prince, sure I let u know about that work | 10:24 |
| lucasagomes | killer_prince, basically creating an efi boot partition is to create a FAT partition no? | 10:26 |
| lucasagomes | and set it as bootable, idk much about the details | 10:26 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Use DiskPartitioner https://review.openstack.org/83399 | 10:29 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Use DiskPartitioner https://review.openstack.org/83399 | 10:30 |
| killer_prince | lucasagomes: yes.. but I looked at https://review.openstack.org/83396 and it looks like there is no provision for anyone to add a partition to a disk with re-creating partition table.. | 10:37 |
| killer_prince | *without re-creating.. | 10:37 |
| lucasagomes | killer_prince, yeah, it's just plumbing working for now to replace the existing way to create partitions | 10:38 |
| lucasagomes | that class is totally generic, and we can set the disklabel to GPT and add_partition(fs_type=fat) | 10:38 |
| lucasagomes | but we need some work on other parts of ironic to tell it that u want to create a efi boot partition | 10:39 |
| lucasagomes | (same for gpt, which I will try to take a look soon) | 10:39 |
| *** pradipta is now known as pradipta_away | 10:41 | |
| killer_prince | cool.. if there is a design document for UEFI support in ironic, can you send me its link.. i would be happy to contribute for GPT and UEFI support. | 10:41 |
| lucasagomes | killer_prince, great! there's none yet unfortunately :( perhaps we should add a blueprint about it at least | 10:42 |
| killer_prince | i guess, theer will be changes needed from the Image side as well and then some change will be needed in the pxe too for UEFI based PXE.. | 10:42 |
| lucasagomes | u wanna start that off? | 10:42 |
| killer_prince | i would be happy to do that... | 10:42 |
| lucasagomes | killer_prince, +1 yeah! also when we add support for local boot | 10:43 |
| lucasagomes | we need to support EFI boot loaders | 10:43 |
| lucasagomes | elilo or something | 10:43 |
| killer_prince | I thought GRUB2 itself has support for EFI. But i could be wrong.. | 10:43 |
| lucasagomes | yeah it might, idk as well | 10:44 |
| killer_prince | I will start on the blueprint for now and put all thats needed for UEFI. Once done you can review and then we can start... | 10:44 |
| * lucasagomes needs to read more about the EFI world heh | 10:45 | |
| lucasagomes | killer_prince, great! thanks for that | 10:45 |
| romcheg | Morning folks | 10:55 |
| agordeev | romcheg: morning! | 10:56 |
| romcheg | lucasagomes: I've seen you removed sleep(3) in 83399. Are you going to add the check we discussed in a separate patch? | 10:57 |
| lucasagomes | romcheg, I put it into the diskpartitioner | 10:57 |
| lucasagomes | romcheg, so I moved it to inside the wrapper as u suggested | 10:58 |
| romcheg | I couldn't find it in the patch set 2 in 83396 | 10:58 |
| lucasagomes | romcheg, but the improvement to use lsof or fuser to detect the disk is busy problem needs to go in another patch | 10:58 |
| lucasagomes | because it's not really related to the work | 10:58 |
| lucasagomes | romcheg, oh | 10:58 |
| lucasagomes | it's patch set 5 now :) | 10:58 |
| lucasagomes | https://review.openstack.org/#/c/83399/5/ironic/common/disk_partitioner.py | 10:58 |
| lucasagomes | romcheg, ohhh | 10:59 |
| romcheg | I can only see 2 of them http://cl.ly/image/2l1o243M0Q2F | 10:59 |
| lucasagomes | ohhh ur talking about 83396 | 10:59 |
| lucasagomes | right I moved it in the 83399 | 10:59 |
| lucasagomes | romcheg, you prefer me to move it it in the 83396? | 11:00 |
| romcheg | Ah, now I see | 11:00 |
| lucasagomes | the reason I did on the other one was because I was touching the work_on_disk() method in the 83399 | 11:00 |
| lucasagomes | and the first one was only to add a stand-alone class | 11:00 |
| lucasagomes | but not touching the existing code | 11:00 |
| romcheg | I think leaving it as is should be fine | 11:01 |
| lucasagomes | ack | 11:01 |
| romcheg | I only have 1 concern | 11:01 |
| lucasagomes | sure | 11:01 |
| *** mdurnosvistov has joined #openstack-ironic | 11:01 | |
| romcheg | So you exec parted N times (where N is the number of partitions). After that you invoke sleep(3) | 11:02 |
| lucasagomes | right | 11:02 |
| romcheg | I think that "device is busy" might appear somewhere between two launches of parted | 11:02 |
| lucasagomes | hmm | 11:03 |
| romcheg | Previously we invoked just one command and waited for some time. | 11:03 |
| lucasagomes | indeed yeah | 11:03 |
| lucasagomes | I haven't seem it yet, but that might be the reason of that sleep | 11:04 |
| romcheg | To be honest I haven't seen this problem myself | 11:04 |
| lucasagomes | I will take a look at it | 11:04 |
| lucasagomes | yeah | 11:04 |
| romcheg | But someone eventually put that sleep for this reason :) | 11:04 |
| lucasagomes | heh yeah | 11:04 |
| lucasagomes | it comes from nova bm | 11:05 |
| lucasagomes | so it will be hard to track it | 11:05 |
| lucasagomes | but yeah, I will put some thought on it | 11:05 |
| lucasagomes | cause sleeping 3 sec for every partition sounds a bit hmm | 11:06 |
| lucasagomes | not cool | 11:06 |
| romcheg | I will post this concern inline. Maybe someone will come and help us | 11:06 |
| romcheg | indeed, not cool | 11:06 |
| romcheg | and there is an invocation that creates a partition table which myght need some time as wel | 11:06 |
| killer_prince | wish it was possible to create multiple partition in a single invocation.. | 11:07 |
| romcheg | killer_prince: sfdisk used to deal with that | 11:08 |
| lucasagomes | killer_prince, I think it's I will take a quick look | 11:08 |
| romcheg | But we need gpt support | 11:08 |
| lucasagomes | I think u can specify multiples mkpart for parted | 11:09 |
| lucasagomes | not sure | 11:09 |
| killer_prince | actually, you can.. | 11:09 |
| lucasagomes | yeah, I will try to consolidate all the commands into one | 11:09 |
| romcheg | lucasagomes: that would be cool | 11:12 |
| *** rameshg87 has left #openstack-ironic | 11:30 | |
| *** Shrews has quit IRC | 11:33 | |
| *** Shrews has joined #openstack-ironic | 11:33 | |
| *** linggao has joined #openstack-ironic | 11:38 | |
| *** loki_ has quit IRC | 11:40 | |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Add DiskPartitioner https://review.openstack.org/83396 | 11:40 |
| lucasagomes | romcheg, ^ | 11:40 |
| romcheg | *is looking* | 11:41 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Use DiskPartitioner https://review.openstack.org/83399 | 11:43 |
| * lucasagomes forgot the unittests urghh | 11:48 | |
| *** saju_m has quit IRC | 11:50 | |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Add DiskPartitioner https://review.openstack.org/83396 | 11:52 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Use DiskPartitioner https://review.openstack.org/83399 | 11:53 |
| romcheg | I was just about to ask :) | 11:53 |
| lucasagomes | heh | 11:53 |
| lucasagomes | and it's not even that early here | 11:54 |
| lucasagomes | :P | 11:54 |
| lucasagomes | maybe it's because it's friday | 11:54 |
| *** Alexei_987 has joined #openstack-ironic | 11:55 | |
| *** rameshg87 has joined #openstack-ironic | 11:56 | |
| *** rameshg87 has left #openstack-ironic | 11:57 | |
| romcheg | yup, it's friday | 11:57 |
| romcheg | A 5L barrel of Paulaner Hefe-Weissbier it waiting for the evening in the fridge :-P | 11:58 |
| lucasagomes | hah nice! | 11:59 |
| *** dguerri is now known as _dguerri | 12:01 | |
| vkozhukalov | hey, guys | 12:08 |
| vkozhukalov | lucasagomes: I saw your review request https://review.openstack.org/#/c/83396/ | 12:08 |
| lucasagomes | vkozhukalov, sure | 12:08 |
| vkozhukalov | right now im working on the same task but for ironic-python-agent | 12:09 |
| lucasagomes | vkozhukalov, ohh | 12:09 |
| lucasagomes | vkozhukalov, do you think we can share some code? | 12:09 |
| lucasagomes | vkozhukalov, that class ^ is pretty much stand-alone | 12:09 |
| vkozhukalov | it would be great to have common code | 12:09 |
| lucasagomes | yeah | 12:09 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Get rid of the swap partition https://review.openstack.org/83726 | 12:09 |
| vkozhukalov | what do you think about how to do that, i mean to share some code | 12:11 |
| lucasagomes | vkozhukalov, what tool ur using for partitioning the disk in ur implementation? | 12:11 |
| vkozhukalov | how to organize it& | 12:11 |
| vkozhukalov | ? | 12:11 |
| lucasagomes | vkozhukalov, I think that's great | 12:11 |
| vkozhukalov | parted | 12:11 |
| vkozhukalov | i'd like to implement a bit more comprehensive tool not only for creating partitions | 12:12 |
| lucasagomes | we actually need some mechanism for that, cause we already some files copied from ironic to the client (common/states.py) | 12:12 |
| vkozhukalov | but also for lvm, software raids etc | 12:12 |
| lucasagomes | vkozhukalov, right, like a DiskManager thing? | 12:12 |
| lucasagomes | oh that's great | 12:12 |
| lucasagomes | btw | 12:12 |
| lucasagomes | I think that would be pretty cool to have | 12:13 |
| * lucasagomes just don't know a good way to share code right now | 12:13 | |
| lucasagomes | unless we make that an standalone tool, and just call it from the client and/or Ironic | 12:14 |
| vkozhukalov | lucasagomes: the same for me (don't know) | 12:14 |
| lucasagomes | maybe we should discuss it in the next meeting? | 12:14 |
| lucasagomes | or devananda ^ ideas? | 12:14 |
| vkozhukalov | ok, will raise this question for meeting | 12:15 |
| lucasagomes | also lifeless input on it would be great ^ | 12:15 |
| vkozhukalov | ok, will ask him as well | 12:16 |
| lucasagomes | vkozhukalov, no worries he will read the log here and comment :) | 12:16 |
| *** killer_prince has quit IRC | 12:18 | |
| *** lazy_prince is now known as killer_prince | 12:18 | |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Get rid of the swap partition https://review.openstack.org/83726 | 12:23 |
| *** lucasagomes is now known as lucas-hungry | 12:26 | |
| *** romcheg1 has joined #openstack-ironic | 12:28 | |
| *** romcheg has quit IRC | 12:30 | |
| *** sabah has quit IRC | 12:34 | |
| mtrisoline | NobodyCam: Sorry, I got pulled away yesterday and I did get an error. But my proof of concept in a dev environment is 100% functional and working. I have one physical server acting as a controller node running qpid and mysql, everything else is running inside of VMWARE. One VM for swift, cinder, and nova respectively. We decided on the nova-network implementation because Neutron is just way more than we need and | 12:35 |
| mtrisoline | NobodyCam: Sorry, I got pulled away yesterday and I did get an error. But my proof of concept in a dev environment is 100% functional and working. I have one physical server acting as a controller node running qpid and mysql, everything else is running inside of VMWARE. One VM for swift, cinder, and nova respectively. We decided on the nova-network implementation because Neutron is just way more than we need and | 12:36 |
| mtrisoline | Sorry for that IRC threw a bunch of errors didn't mean to send that twice. | 12:37 |
| Shrews | Ok, I'm definitely concerned about our tests now. Two of us have encountered something eating all memory during unit tests, and now Jenkins has seen it: http://logs.openstack.org/05/83105/3/check/gate-ironic-python26/dab25e8/console.html | 12:40 |
| Shrews | NobodyCam, devananda, lucas-hungry: ^^^^ | 12:41 |
| *** jdob has joined #openstack-ironic | 12:56 | |
| openstackgerrit | Rohan Kanade proposed a change to openstack/ironic: Adds exact match filters to nova scheduler https://review.openstack.org/83734 | 12:57 |
| openstackgerrit | Jim Rollenhagen proposed a change to openstack/ironic-python-agent: Remove advertise-host from service file https://review.openstack.org/83608 | 13:00 |
| *** zigo has quit IRC | 13:13 | |
| *** zigo has joined #openstack-ironic | 13:14 | |
| *** saju_m has joined #openstack-ironic | 13:15 | |
| *** pbrooko has quit IRC | 13:25 | |
| *** matty_dubs|gone is now known as matty_dubs | 13:27 | |
| *** jgrimm has joined #openstack-ironic | 13:42 | |
| romcheg1 | lucas-hungry: I noticed my patch that fixes https://bugs.launchpad.net/ironic/+bug/1251880 was merged | 13:43 |
| romcheg1 | However the bug is untargeted | 13:43 |
| *** romcheg1 is now known as romcheg | 13:43 | |
| romcheg | lucas-hungry: should I target it to j1? | 13:43 |
| *** lucas-hungry is now known as lucasagomes | 13:43 | |
| lucasagomes | romcheg, hey | 13:43 |
| lucasagomes | romcheg, hmmmm, I think it will go to rc1 no? | 13:44 |
| *** toure has joined #openstack-ironic | 13:44 | |
| lucasagomes | was rc1 released already? | 13:44 |
| romcheg | lucasagomes: I think so, let's check that | 13:44 |
| lucasagomes | ack | 13:44 |
| *** zigo has quit IRC | 13:47 | |
| *** zigo has joined #openstack-ironic | 13:47 | |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Get rid of the swap partition https://review.openstack.org/83726 | 13:50 |
| *** zigo has quit IRC | 13:51 | |
| jroll | morning ironic | 13:55 |
| *** zigo has joined #openstack-ironic | 13:55 | |
| *** jbjohnso_ has joined #openstack-ironic | 13:56 | |
| jbjohnso_ | morning | 13:56 |
| jbjohnso_ | pyghmi 0.9.5.2 should be on pypi... but I might have sinned against process because I forgot how I was supposed to do it | 13:57 |
| *** jistr has quit IRC | 13:57 | |
| lucasagomes | heh | 13:57 |
| lucasagomes | jbjohnso_, does *.5.2 includes the unicode fix? | 13:57 |
| jbjohnso_ | lucasagomes, indeed | 13:58 |
| lucasagomes | for the hmac function? | 13:58 |
| lucasagomes | nice | 13:58 |
| lucasagomes | yeah we might want to pin our requirements to >=9.5.2 then | 13:58 |
| jbjohnso_ | err 5.9.2 | 13:58 |
| lucasagomes | unless we use that fix that yuriyz propoused | 13:58 |
| agordeev | jroll jbjohnso_ morning! | 13:58 |
| jbjohnso_ | 0.5.9.2 | 13:58 |
| lucasagomes | heh 0.5.9.2 :) | 13:59 |
| *** zigo has quit IRC | 13:59 | |
| jbjohnso_ | I'm saving 0.6 for at least full inventory and SEL retrieval | 13:59 |
| lucasagomes | right | 13:59 |
| *** zigo has joined #openstack-ironic | 13:59 | |
| *** jistr has joined #openstack-ironic | 13:59 | |
| jroll | hey agordeev | 13:59 |
| lucasagomes | jrist, morning | 14:00 |
| lucasagomes | jroll, * | 14:00 |
| jrist | haha | 14:00 |
| jroll | vkozhukalov, lucasagomes, I've also been thinking about how to share code between ironic and the agent | 14:00 |
| jrist | my nick strikes again | 14:00 |
| jroll | heh | 14:00 |
| lucasagomes | jrist, hah yeah, sorry buddy! | 14:00 |
| jrist | what does vkozhukalov mean? | 14:00 |
| jrist | something awesome, I hope | 14:00 |
| jroll | it's a name :P | 14:00 |
| lucasagomes | jroll, yeah... ironic, agent, and client would be nice to have common code | 14:01 |
| jroll | anyhow, I don't want to pull ironic into the agent for obvious reasons | 14:01 |
| jroll | for the partition tool, I think that could be a separate library | 14:01 |
| vkozhukalov | jrist: vkozhukalov means Vladimir Kozhukalov :-) | 14:01 |
| jroll | but as far as other code, not sure what to do | 14:01 |
| lucasagomes | jroll, sure, yeah | 14:01 |
| lucasagomes | maybe having an ironic-common repo | 14:01 |
| lucasagomes | idk | 14:01 |
| jrist | vkozhukalov: ha :) | 14:02 |
| jroll | I've already copied some of the common rest stuff :( | 14:02 |
| jrist | vkozhukalov: awesome! | 14:02 |
| jrist | vkozhukalov: that is a great name | 14:02 |
| * jrist waves | 14:02 | |
| lucasagomes | jroll, :( in the client we copied it as well | 14:02 |
| lucasagomes | jroll, like states | 14:02 |
| jroll | ironic-common isn't a horrible idea | 14:02 |
| jroll | yeah | 14:02 |
| lucasagomes | jroll, yeah... idk if it's ideal, but would solve our problem | 14:02 |
| * jroll wonders if there's any other openstack projects that have encountered this | 14:03 | |
| * lucasagomes same | 14:03 | |
| jroll | I can send a mailing list post later after I dig myself out of existing email | 14:03 |
| lucasagomes | jroll, :D sure that would be useful | 14:04 |
| jroll | :) | 14:04 |
| vkozhukalov | jroll: +1 | 14:04 |
| jbjohnso_ | lucasagomes, fyi, I could see one reason to be wary of non-ascii stuff, I'm not sure ipmitool can set such a password. However the underlying BMC generally have such a limitation | 14:06 |
| jbjohnso_ | shouldn't have I should say | 14:06 |
| NobodyCam | good morning Ironic | 14:08 |
| *** russellb is now known as rustlebee | 14:08 | |
| agordeev | NobodyCam: morning! | 14:09 |
| jroll | morning NobodyCam | 14:10 |
| NobodyCam | morning agordeev and jroll :) | 14:11 |
| *** ndipanov is now known as engeepanoff | 14:18 | |
| NobodyCam | ahh just reading some scroll back in OoO .. looks like lifeless found this: https://bugs.launchpad.net/openstack-ci/+bug/1298731 | 14:18 |
| romcheg | lucasagomes: forgot to write that: rc-1 was released yesterday | 14:18 |
| romcheg | so j-1 is the next milestone for that bug | 14:19 |
| lucasagomes | romcheg, ack! thanks for the ingo | 14:20 |
| lucasagomes | info* | 14:20 |
| NobodyCam | morning romcheg :) | 14:20 |
| romcheg | Morning NobodyCam | 14:20 |
| *** openstack has joined #openstack-ironic | 14:21 | |
| *** jistr has quit IRC | 14:21 | |
| *** jistr has joined #openstack-ironic | 14:21 | |
| *** Manishanker has quit IRC | 14:33 | |
| *** dshulyak has quit IRC | 14:37 | |
| *** rwsu has quit IRC | 14:38 | |
| *** rwsu has joined #openstack-ironic | 14:41 | |
| lucasagomes | morning NobodyCam | 14:55 |
| NobodyCam | morning lucasagomes :) | 14:57 |
| NobodyCam | I was just looking into the bug lifeless filed about ssh commands | 14:57 |
| lucasagomes | NobodyCam, in ironic? | 15:00 |
| lucasagomes | NobodyCam, what's the #> | 15:00 |
| lucasagomes | ?* | 15:00 |
| NobodyCam | https://bugs.launchpad.net/openstack-ci/+bug/1298731 | 15:01 |
| Shrews | agordeev: What you propose on adam_g's review is fine. As a "quick start" guide, I think there's a fine line between showing just enough to get the reader doing something, and confusing the reader with too much info. Your suggested change is fine by me, though, if you want to -1 it. | 15:01 |
| agordeev | Shrews: I was thinking that the entire patch is not about explosing significant default values. So it'll better to add what i propose in another patch later, if it's still worth adding | 15:10 |
| *** sphoorti has joined #openstack-ironic | 15:17 | |
| *** packet has joined #openstack-ironic | 15:18 | |
| openstackgerrit | Rohan Kanade proposed a change to openstack/ironic: Adds exact match filters to nova scheduler https://review.openstack.org/83734 | 15:25 |
| openstackgerrit | Jim Rollenhagen proposed a change to openstack/ironic-python-agent: Added PyPy env to tox.ini https://review.openstack.org/83605 | 15:26 |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Use GB instead of MB for swap https://review.openstack.org/83788 | 15:33 |
| NobodyCam | br | 15:35 |
| NobodyCam | brb | 15:35 |
| *** saju_m has quit IRC | 15:35 | |
| *** eghobo has joined #openstack-ironic | 15:41 | |
| *** packet has quit IRC | 15:41 | |
| *** toure has left #openstack-ironic | 15:43 | |
| NobodyCam | gota reboot | 15:55 |
| *** athomas has quit IRC | 16:01 | |
| *** sphoorti has quit IRC | 16:04 | |
| *** sphoorti has joined #openstack-ironic | 16:04 | |
| *** matty_dubs is now known as matty_dubs|lunch | 16:09 | |
| NobodyCam | oh thats much better ... guess I need to reboot every month or so :-p | 16:13 |
| JayF | jroll: for https://review.openstack.org/#/c/83605/ why did jenkins think tests passed and you didn't? Are we missing something when running tests? | 16:14 |
| JayF | (in jenkins) | 16:14 |
| NobodyCam | post bbt walkies..bbiafm | 16:15 |
| devananda | morning, all | 16:17 |
| *** epim has joined #openstack-ironic | 16:18 | |
| JayF | jroll: or does that mean that tests all pass on pypy now for the agent? if so that'd be awesome | 16:18 |
| * devananda catches up on reviews and emails from the night | 16:21 | |
| lucasagomes | devananda, morning | 16:22 |
| devananda | lucasagomes: the DiskPartitioner series of patches -- Juno material, I presume? | 16:23 |
| lucasagomes | devananda, yeah | 16:23 |
| lucasagomes | devananda, should I set the target to j1 in the bugs? | 16:23 |
| devananda | lucasagomes: cool. do you think the noswap fix needs to be in icehouse? | 16:23 |
| devananda | lucasagomes: yep. that'd be helpful for tracking | 16:24 |
| lucasagomes | devananda, hmm... I'm more concerned about the fix of the partitining in general | 16:24 |
| lucasagomes | devananda, right now it's broken in trunk (if you create a swap partition with >= 1024 it breaks) | 16:24 |
| devananda | oh | 16:24 |
| devananda | that's bad | 16:25 |
| lucasagomes | yeah | 16:25 |
| lucasagomes | nova bm is broken too | 16:25 |
| devananda | i would call that RC critical for us, even though it's an "enhancement" over nova bm | 16:25 |
| devananda | like | 16:25 |
| devananda | anyone with real hardware is going to want swap > 1GB | 16:25 |
| lucasagomes | devananda, yeah I set it as critical | 16:25 |
| devananda | ok | 16:26 |
| lucasagomes | devananda, also, our nova ironic driver is not looking at the swap option in the flavor | 16:26 |
| lucasagomes | and sending it to ironic | 16:26 |
| devananda | hah | 16:26 |
| lucasagomes | so that would need to be included as well | 16:26 |
| devananda | right | 16:26 |
| devananda | and how ever we fix this, we will want to support that disk layout | 16:26 |
| devananda | eg during rebuild... if at all possible | 16:26 |
| lucasagomes | yeah | 16:27 |
| lucasagomes | https://bugs.launchpad.net/ironic/+bug/1297925 | 16:27 |
| devananda | so changing the order of partitions on disk has some ramifications | 16:27 |
| NobodyCam | good morning devananda | 16:27 |
| lucasagomes | devananda, it does yeah, all this disk operations right now is messy, I'm trying to add some consistency in the series of patches | 16:28 |
| lucasagomes | for e.g having root as the last partition always | 16:28 |
| lucasagomes | this also allows cloud-init to resize it | 16:28 |
| lucasagomes | if needed | 16:28 |
| devananda | right | 16:28 |
| devananda | good stuff | 16:28 |
| lucasagomes | and parameters unit size needs to be more consistent as well, all GB | 16:29 |
| devananda | so, i'd like your thouts on fixing the "sfdisk breaks if swap >= 1GB" by itself | 16:30 |
| devananda | in icehouse | 16:30 |
| lucasagomes | right | 16:30 |
| lucasagomes | it sounds important, the problem with my fix is that it was also a bit of plumbing to be able to add other stuff on top of the change (gpt, get rid of swap etc) | 16:31 |
| lucasagomes | maybe for icehouse we can just include a simple fix | 16:31 |
| lucasagomes | like replacing sfdisk with parted but doing something similar with what we have in trunk right now | 16:32 |
| JayF | Parted is what the Ironic Python Agent uses now to inject partitions for configdrive -> http://git.openstack.org/cgit/openstack/ironic-python-agent/tree/ironic_python_agent/shell/copy_configdrive_to_disk.sh#n24 | 16:33 |
| JayF | parted script mode is pretty great in my experience | 16:33 |
| lucasagomes | JayF, yeah | 16:33 |
| lucasagomes | JayF, https://review.openstack.org/83396 | 16:33 |
| JayF | That was an open tab for me to review yesterday but I never quite got there | 16:34 |
| JayF | I'll leave it up today and try to get there :D | 16:34 |
| lucasagomes | :) ack thanks | 16:35 |
| *** comstud is now known as bearhands | 16:35 | |
| NobodyCam | devananda: is the swap size > 1 gb a issue on vm only? Ie. when the nodes hdd is created it uses disk_size and not disk_size+swap? | 16:35 |
| *** vkozhukalov has quit IRC | 16:35 | |
| devananda | NobodyCam: AIUI, it's a problem with how we're using sfdisk in general | 16:36 |
| devananda | lucasagomes: ^ ? | 16:36 |
| lucasagomes | right is not a disk space problem | 16:37 |
| lucasagomes | that 1MB lead partition is making things to fail | 16:37 |
| lucasagomes | and seems to be a bug in sfdisk | 16:37 |
| NobodyCam | devananda: side question: have you worked with the infra repos much.. with reguard to https://bugs.launchpad.net/openstack-ci/+bug/1298731 | 16:37 |
| NobodyCam | lucasagomes: what is the actual error? | 16:38 |
| devananda | NobodyCam: https://bugs.launchpad.net/ironic/+bug/1297925 | 16:38 |
| devananda | NobodyCam: I am not familiar with the issue in testenv/tripleoci | 16:39 |
| NobodyCam | devananda: do you know where testenv gets its keys from? | 16:39 |
| *** lucasagomes has quit IRC | 16:41 | |
| devananda | NobodyCam: nope | 16:42 |
| *** lucasagomes has joined #openstack-ironic | 16:43 | |
| lucasagomes | devananda, NobodyCam my connection dropped :( | 16:43 |
| lucasagomes | what was my last message? | 16:43 |
| NobodyCam | devananda, NobodyCam my connection dropped :( | 16:45 |
| NobodyCam | :-p | 16:45 |
| lucasagomes | lol | 16:46 |
| lucasagomes | before that hah | 16:46 |
| jroll | JayF: the latter :D | 16:46 |
| NobodyCam | oh | 16:46 |
| NobodyCam | and seems to be a bug in sfdisk | 16:46 |
| lucasagomes | yeah | 16:46 |
| NobodyCam | hehe | 16:46 |
| JayF | jroll: yeah your comments when you came into the office were highly relevant :) I've already +2'd that pypy patch | 16:46 |
| *** mdurnosvistov has quit IRC | 16:47 | |
| JayF | Ironic-python-agent already works with pypy with a small patch (that jroll has in) so we're going to make pypy a voting job for it after that gets merged. Woo. | 16:47 |
| *** engeepanoff has quit IRC | 16:47 | |
| *** ndipanov has joined #openstack-ironic | 16:48 | |
| *** ndipanov has quit IRC | 16:48 | |
| openstackgerrit | Jim Rollenhagen proposed a change to openstack/ironic-python-agent: Remove advertise-host from service file https://review.openstack.org/83608 | 16:53 |
| *** matty_dubs|lunch is now known as matty_dubs | 16:58 | |
| *** epim has quit IRC | 17:01 | |
| NobodyCam | lucasagomes: you still have your sfdisk test env? | 17:06 |
| lucasagomes | NobodyCam, yup | 17:06 |
| NobodyCam | if you have a free minute can you run the current sfdisk command but add -i to the command? | 17:08 |
| NobodyCam | ie sfdisk -uMi /dev/vdb | 17:09 |
| lucasagomes | NobodyCam, sure | 17:09 |
| NobodyCam | oh I forgot TGIF! | 17:10 |
| lucasagomes | NobodyCam, http://paste.openstack.org/show/74579/ | 17:11 |
| NobodyCam | lucasagomes: Thank you :) | 17:11 |
| lucasagomes | yay friday :) | 17:11 |
| NobodyCam | lucasagomes: thank you :) | 17:11 |
| lucasagomes | np | 17:12 |
| linggao | Hi matty_dubs, | 17:12 |
| NobodyCam | morning linggao :) | 17:12 |
| linggao | morning NobodyCam. Hey. | 17:13 |
| NobodyCam | :) | 17:14 |
| *** harlowja_away is now known as harlowja | 17:15 | |
| linggao | matty_dubs, this is reagarding the console support sjing and you were working on, are you planing to pick it up from where sjing left? If not, I'll pick it up. | 17:15 |
| *** ifarkas has quit IRC | 17:16 | |
| matty_dubs | linggao: I had intended to, but hadn't been focused on it right now because of feature freeze. | 17:17 |
| matty_dubs | If you'd prefer to work on it, feel free to do so -- I can share some notes. | 17:17 |
| *** saju_m has joined #openstack-ironic | 17:19 | |
| linggao | I can be either way as I told sjing. Since I am trying to get my feet wet again and you guys are busy with the release, maybe this is a good start point for me if it is okay with you. | 17:20 |
| linggao | matty_dubs, do you have a note on you how you tested it? | 17:23 |
| *** epim has joined #openstack-ironic | 17:23 | |
| matty_dubs | linggao: Sure. https://review.openstack.org/#/c/64100/ is much of the way there; I think the last big hurdle is around spawning and controlling the process. I was starting to build a Conductor task to manage the process when the -2 hit and I moved my focus elsewhere. | 17:24 |
| matty_dubs | linggao: I can write something up re: testing. Do you have access to a box with an IPMI serial-over-LAN console? | 17:24 |
| linggao | Yes, I have. | 17:24 |
| matty_dubs | linggao: Perfect! Let me take a moment to remember exactly what I did. O:-) Then I'll write it up and send it over. | 17:25 |
| linggao | thanks a lot matty_dubs! | 17:25 |
| matty_dubs | linggao: Thanks for picking this up! And let me know if I can help any once you get going. | 17:26 |
| *** epim_ has joined #openstack-ironic | 17:27 | |
| linggao | matty_dubs, sure. thanks a lot for all the help you have done. | 17:28 |
| *** _dguerri has quit IRC | 17:28 | |
| *** epim has quit IRC | 17:28 | |
| *** epim_ is now known as epim | 17:28 | |
| *** Shrews has quit IRC | 17:29 | |
| *** mdurnosvistov has joined #openstack-ironic | 17:32 | |
| *** stevehuang has joined #openstack-ironic | 17:34 | |
| *** jistr has quit IRC | 17:34 | |
| * devananda does a quick refactoring of scheduler related thigns | 17:34 | |
| *** saju_m has quit IRC | 17:35 | |
| *** saju_m has joined #openstack-ironic | 17:36 | |
| *** k4n0 has joined #openstack-ironic | 17:43 | |
| matty_dubs | linggao: I think https://gist.github.com/n1zyy/9838658 covers everything I did for setting up console with that patch. | 17:48 |
| *** krtaylor has quit IRC | 17:55 | |
| *** derekh has quit IRC | 17:56 | |
| *** mkerrin has quit IRC | 18:02 | |
| openstackgerrit | Josh Gachnang proposed a change to openstack/ironic-python-agent: Adding DynamicLoopingCall around lookup https://review.openstack.org/83604 | 18:03 |
| linggao | matty_dubs, I'll try with the instructions. thanks. | 18:05 |
| lucasagomes | ok! I'm done for today | 18:08 |
| lucasagomes | have a good night and good weekend everybody! | 18:09 |
| lucasagomes | NobodyCam, devananda matty_dubs linggao etc... ! night, enjoy the weekend! | 18:09 |
| linggao | lucasagomes, have a great weekend! | 18:09 |
| *** lucasagomes is now known as lucas-afk | 18:10 | |
| matty_dubs | You too, lucas-afk ! | 18:11 |
| linggao | matty_dubs, the setting is within the venv with Ironic only or is it in a real world with keystone and nova and other stuff? | 18:11 |
| matty_dubs | linggao: I only had to add it to Ironic. | 18:12 |
| matty_dubs | Assuming you already have Ironic set up. | 18:12 |
| matty_dubs | No changes needed outside Ironic. | 18:12 |
| linggao | ok | 18:14 |
| NobodyCam | Night lucas-afk | 18:16 |
| NobodyCam | have a good weekend] | 18:16 |
| devananda | k4n0: reviews posted on the exact scheduler patch. I know it's late for you -- do you mind if we continue the work on this patch today? | 18:24 |
| k4n0 | devananda: I can stick around, got lots of coffee :), I will fix the review comments now | 18:25 |
| devananda | k4n0: lol :) thanks! | 18:26 |
| openstackgerrit | Devananda van der Veen proposed a change to openstack/ironic: ironic_host_manager was missing two stats https://review.openstack.org/83851 | 18:26 |
| openstackgerrit | Devananda van der Veen proposed a change to openstack/ironic: Clean up IronicNodeStates.update_from_compute_node https://review.openstack.org/83852 | 18:26 |
| openstackgerrit | Devananda van der Veen proposed a change to openstack/ironic: Refactor nova.virt.ironic.driver get_host_stats https://review.openstack.org/83853 | 18:26 |
| devananda | k4n0: these might help you a bit ^ | 18:26 |
| k4n0 | ok, rebasing on top then | 18:26 |
| devananda | k4n0: i haven't tested them yet, fwiw | 18:27 |
| k4n0 | devananda: ok, ill be careful | 18:27 |
| devananda | but in theory, the only one with a functional change (and it's one that your patches require) is the first one | 18:27 |
| devananda | so if you base on just https://review.openstack.org/83851, shoudl be fine | 18:27 |
| *** Alexei_987 has quit IRC | 18:27 | |
| k4n0 | devananda: ok | 18:28 |
| devananda | linggao: it looks like you are testing teh scheduler fix that k4n0 posted. I just tossed up some related patches, if you have time to review them, that'd be great | 18:28 |
| NobodyCam | brb | 18:30 |
| *** vkozhukalov has joined #openstack-ironic | 18:30 | |
| linggao | devananda, yes. I will. | 18:32 |
| linggao | brb | 18:37 |
| *** openstackgerrit has quit IRC | 18:48 | |
| *** openstackgerrit has joined #openstack-ironic | 18:48 | |
| *** max_lobur1 has quit IRC | 18:50 | |
| *** dguerri has joined #openstack-ironic | 18:50 | |
| NobodyCam | devananda: good catch with 83851 :) | 18:52 |
| NobodyCam | brb | 19:00 |
| *** harlowja is now known as harlowja_away | 19:00 | |
| *** harlowja_away is now known as harlowja | 19:17 | |
| openstackgerrit | Rohan Kanade proposed a change to openstack/ironic: Adds exact match filters to nova scheduler https://review.openstack.org/83734 | 19:20 |
| NobodyCam | uuummmm bagel :-p | 19:21 |
| *** packet has joined #openstack-ironic | 19:33 | |
| *** romcheg has quit IRC | 19:33 | |
| openstackgerrit | A change was merged to openstack/ironic-python-agent: Added PyPy env to tox.ini https://review.openstack.org/83605 | 19:34 |
| k4n0 | devananda: can you please review the latest patchset | 19:55 |
| *** epim has quit IRC | 20:07 | |
| *** krtaylor has joined #openstack-ironic | 20:07 | |
| *** romcheg has joined #openstack-ironic | 20:15 | |
| *** vkozhukalov has quit IRC | 20:21 | |
| *** max_lobur has joined #openstack-ironic | 20:31 | |
| *** max_lobur has quit IRC | 20:32 | |
| *** max_lobur has joined #openstack-ironic | 20:32 | |
| *** max_lobur has quit IRC | 20:32 | |
| *** max_lobur has joined #openstack-ironic | 20:32 | |
| linggao | k4n0, are the 2 new CONF settings under default section for your patch? | 20:38 |
| k4n0 | linggao: yes, its from devananda's review comments | 20:38 |
| linggao | k4n0, then where do they get used? | 20:39 |
| linggao | The base class does not check these 2 settings. | 20:39 |
| k4n0 | linggao: they will be used in nova.conf right? | 20:40 |
| linggao | The base class has: if filter_cls_names is None: | 20:40 |
| linggao | filter_cls_names = CONF.scheduler_default_filters | 20:40 |
| linggao | user can set it up in nova.conf, but no place in code checks these settings. | 20:41 |
| k4n0 | linggao: i will have to push new nova patchset to fix this | 20:42 |
| linggao | :) | 20:42 |
| linggao | k4n0, in the init function of the ironic_host_manager, read the settins and then assign CONF.scheduler_default_filters = CONF.baremetal_scheduler_default_filters | 20:45 |
| linggao | if scheduler_use_baremetal_filters == TRUE | 20:45 |
| k4n0 | linggao: got it, so we dont need to change nova patch now | 20:46 |
| linggao | k4n0, right. | 20:46 |
| *** eguz has joined #openstack-ironic | 20:52 | |
| *** eghobo has quit IRC | 20:56 | |
| *** lucas-afk has quit IRC | 21:01 | |
| *** matty_dubs is now known as matty_dubs|gone | 21:01 | |
| *** sphoorti has quit IRC | 21:03 | |
| *** Shrews has joined #openstack-ironic | 21:05 | |
| Shrews | If anyone has tried to ping me today, I have been (and continue to be) a victim of the hp cloud outage. | 21:07 |
| Shrews | Fun | 21:07 |
| * jroll inserts rackspace cloud advert | 21:08 | |
| jroll | :P | 21:08 |
| devananda | bearhands: hi! spare a few minutes for scheduler questions? | 21:09 |
| bearhands | sure | 21:09 |
| *** jbjohnso_ has quit IRC | 21:09 | |
| openstackgerrit | A change was merged to openstack/python-ironicclient: Avoid traceback with insufficient auth credentials https://review.openstack.org/79494 | 21:09 |
| devananda | bearhands: i'm looking at k4n0's work to add exact match scheduler filters | 21:09 |
| devananda | bearhands: the filters look good. but i dont see how the current ones are being loaded, so i'm not exactly sure how we should be loading the new ones | 21:10 |
| openstackgerrit | A change was merged to openstack/ironic-python-agent: Remove advertise-host from service file https://review.openstack.org/83608 | 21:10 |
| devananda | bearhands: eg, https://github.com/openstack/nova/blob/master/nova/scheduler/host_manager.py#L45 | 21:10 |
| devananda | doesn't list any of the weight stuff | 21:10 |
| devananda | er | 21:10 |
| devananda | not weight | 21:10 |
| bearhands | weight classes are line 56 | 21:10 |
| devananda | the cpu/ram/disk filters | 21:10 |
| bearhands | oh | 21:10 |
| devananda | yea. wrong thing, sorry | 21:10 |
| bearhands | well, so those are just the defaults | 21:10 |
| *** jdob has quit IRC | 21:11 | |
| devananda | is matching by available cpu/ram/disk not the default? | 21:11 |
| bearhands | ComputeFilter | 21:11 |
| bearhands | sorry | 21:11 |
| bearhands | or RamFilter also,I guess | 21:12 |
| devananda | not ComputeCapabilitiesFilter either -- that just checks extra specs | 21:12 |
| bearhands | I'm not sure disk is by default | 21:12 |
| devananda | so there's RamFilter, DiskFilter, COreFilter | 21:12 |
| devananda | but it doesn' tlook like they're loaded by default?? | 21:12 |
| bearhands | RamFilter is | 21:12 |
| bearhands | line 49 | 21:12 |
| bearhands | but it looks like cores and disk are not | 21:12 |
| devananda | ahh | 21:12 |
| devananda | missed that | 21:12 |
| devananda | ok | 21:12 |
| bearhands | cores doesn't tend to matter, since they are virtual | 21:12 |
| *** romcheg has quit IRC | 21:12 | |
| bearhands | (until ironic maybe) | 21:13 |
| bearhands | i'm not sure why Disk is not in the default list though | 21:13 |
| devananda | thanks | 21:15 |
| devananda | bearhands: so the list i'm proposing here - -https://review.openstack.org/#/c/83734/3/ironic/nova/scheduler/ironic_host_manager.py | 21:16 |
| devananda | does that seem reasonable to you for ironic's needs | 21:16 |
| devananda | lifeless: same question for you ^ | 21:17 |
| bearhands | yes, your comment there seems to make sense to me | 21:18 |
| bearhands | definitely should have those other filters, IMO | 21:18 |
| devananda | bearhands: ack. thanks much | 21:19 |
| devananda | k4n0: comments left. are you around to fix it up? | 21:20 |
| k4n0 | devananda: yup, fixing them | 21:20 |
| devananda | awesome | 21:21 |
| *** romcheg has joined #openstack-ironic | 21:22 | |
| openstackgerrit | Rohan Kanade proposed a change to openstack/ironic: Adds exact match filters to nova scheduler https://review.openstack.org/83734 | 21:24 |
| openstackgerrit | Josh Gachnang proposed a change to openstack/ironic: Adding swift temp url support https://review.openstack.org/81391 | 21:26 |
| openstackgerrit | A change was merged to openstack/ironic: Imported Translations from Transifex https://review.openstack.org/82993 | 21:27 |
| *** stevehuang has quit IRC | 21:28 | |
| devananda | i need to run out for food | 21:31 |
| devananda | but if anyone's looking for something to do | 21:31 |
| devananda | deleting obsolete translation lines would be helpful | 21:31 |
| devananda | basically duplicate https://review.openstack.org/#/c/81925/1 but for Ironic | 21:31 |
| devananda | as a potential fix to https://bugs.launchpad.net/ironic/+bug/1298645 | 21:31 |
| devananda | I'm still waiting on clarkb to confirm whether he thinkgs it's a viable solution.... | 21:32 |
| devananda | bbiab | 21:32 |
| lifeless | devananda: replied | 21:34 |
| devananda | lifeless: keep it in sync with ...? | 21:35 |
| lifeless | devananda: the list in the default scheduler | 21:37 |
| lifeless | like retryfilter etc | 21:37 |
| devananda | ah | 21:37 |
| devananda | so i dont know that that's a long-term solution | 21:38 |
| devananda | obviously this needs to get into nova | 21:38 |
| k4n0 | lifeless: the default filters in nova havent changed since 2012 | 21:38 |
| *** eguz has quit IRC | 21:39 | |
| * devananda really goes afk now :) | 21:41 | |
| NobodyCam | have a goodweekend devananda | 21:44 |
| NobodyCam | oh just food.. then I should see you b4 eod | 21:45 |
| *** mdurnosvistov has quit IRC | 21:48 | |
| linggao | devananda, I have tested your recent 3 changes I1b49accf, I7d66ea5e, I0d055dba with k4n0's exact filters with my kluge env. Got an error. I left an comment in your patch I1b49accf, | 22:09 |
| linggao | If add those 4 missing items back to the dict, everything works fine. | 22:09 |
| linggao | I have to go. | 22:10 |
| linggao | Have a good weekend. | 22:10 |
| k4n0 | linggao: thanks for the review | 22:11 |
| *** eghobo has joined #openstack-ironic | 22:13 | |
| linggao | k4n0, :). Thank you for your hard work. | 22:13 |
| *** eghobo has quit IRC | 22:13 | |
| *** eghobo has joined #openstack-ironic | 22:13 | |
| *** linggao has quit IRC | 22:14 | |
| NobodyCam | just incase anyone wanted to know I have fully deployed seed / undercloud with ironic : a6d0f7ee-1a24-47b8-abf2-198657449190 | overcloud | CREATE_IN_PROGRESS | 2014-03-28T22:27:14Z | 22:28 |
| jroll | \o/ | 22:28 |
| NobodyCam | :) | 22:29 |
| JoshNang | woo! | 22:30 |
| NobodyCam | I have to dbl check but I think All *critical* patches have landed and you can use ironic with devtest simply by setting `export USE_IRONIC=1` (with out quots ofc) | 22:30 |
| mordred | NobodyCam: wow. exciting! | 22:31 |
| NobodyCam | :) | 22:34 |
| openstackgerrit | A change was merged to openstack/ironic: Remove some obsolete settings from DevStack doc https://review.openstack.org/83540 | 22:38 |
| adam_g | devananda, was thinking about nova driver unit test coverage. remembered this crazy thing keystone is/was doing where the test suite would go out to the internets and checkout copies of keystoneclient into its tree during test runs | 22:40 |
| NobodyCam | just worked with lifeless and https://review.openstack.org/#/c/83906 should fix up the ssh issue with ironic in ci | 22:47 |
| *** eguz has joined #openstack-ironic | 22:48 | |
| *** max_lobur has quit IRC | 22:48 | |
| NobodyCam | and now adding a note to our ssh driver about that | 22:49 |
| *** packet has quit IRC | 22:51 | |
| *** eghobo has quit IRC | 22:52 | |
| mordred | adam_g: I keep meaning to delete taht code | 22:52 |
| adam_g | mordred, yeah--it drove me absolutely nuts trying to package for buildds | 22:52 |
| adam_g | but something similar may be useful as a short term thing till ironic driver lives in nova | 22:52 |
| mordred | adam_g: do you want to check out versions of nova? | 22:53 |
| devananda | adam_g: that's what i've been saying | 22:54 |
| adam_g | mordred, yeah. the driver and its tests now live in the ironic tree @ ironic/nova | 22:55 |
| devananda | adam_g: temporary fix. check out nova into ironic's venv during a test run | 22:55 |
| devananda | then run nova's unit tests | 22:55 |
| adam_g | devananda, right | 22:55 |
| adam_g | devananda, that could just be done with tox + a script in tools/ no? | 22:55 |
| devananda | right | 22:55 |
| devananda | exactly | 22:55 |
| devananda | we should do that | 22:55 |
| adam_g | ill play with that this afternoon | 22:55 |
| devananda | thanks | 22:55 |
| devananda | it's better than the nothing we have now | 22:55 |
| adam_g | yeah | 22:55 |
| adam_g | even if it takes a good 7 minutes to checkout nova :) | 22:55 |
| devananda | heh | 22:56 |
| openstackgerrit | Chris Krelle proposed a change to openstack/ironic: Add note to ssh about impact on ci testing https://review.openstack.org/83926 | 22:56 |
| adam_g | or maybe thats changed | 22:56 |
| NobodyCam | devananda: ^^^ is just a note | 22:56 |
| NobodyCam | adam_g: 7 minutes seems long to me? | 23:02 |
| adam_g | NobodyCam, oh ya, you're right. didn't notice how fast it is lately 0m57.922s | 23:09 |
| NobodyCam | happy dance: http://paste.openstack.org/show/t38Hj7wkxmfB4aT8WQJQ/ | 23:10 |
| NobodyCam | :) adam_g if your good with seven I sure I dig up a old dialup account for ya :) | 23:10 |
| *** dhellmann is now known as dhellmann_ | 23:13 | |
| openstackgerrit | Devananda van der Veen proposed a change to openstack/ironic: Refactor nova.virt.ironic.driver get_host_stats https://review.openstack.org/83853 | 23:16 |
| NobodyCam | go get food... | 23:35 |
| devananda | ? | 23:36 |
| NobodyCam | darn mouse scroll the irssi screen and I wasn't watching | 23:36 |
| NobodyCam | last line was you saying you needed to go get food | 23:36 |
| NobodyCam | doh | 23:36 |
| devananda | :p | 23:36 |
| openstackgerrit | Devananda van der Veen proposed a change to openstack/ironic: Remove obsolete po entries - they break translation jobs https://review.openstack.org/83935 | 23:46 |
| NobodyCam | lol I was just going to look at that one :-p | 23:46 |
| devananda | :) | 23:47 |
| NobodyCam | was just thinging of the grep | sort | line to use | 23:47 |
| NobodyCam | thinking even | 23:47 |
| NobodyCam | wow https://bugs.launchpad.net/ironic/+bug/1298645 | 23:49 |
| NobodyCam | no no | 23:49 |
| NobodyCam | wow -15234 | 23:49 |
| NobodyCam | devananda: has clark confirmed that will fix? | 23:50 |
| devananda | NobodyCam: http://paste.openstack.org/show/74602/ | 23:51 |
| devananda | NobodyCam: and yes, he said that it probably will fix it | 23:51 |
| NobodyCam | :) | 23:53 |
| NobodyCam | you used Releated not Closes ? :-p | 23:57 |
| NobodyCam | on the bug # | 23:57 |
| devananda | yes | 23:59 |
| devananda | because I'm not sure it actually closes it | 23:59 |
| devananda | and I am going to push this through to find out... afaik, there's no way to test it | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
