| *** Haomeng has joined #openstack-ironic | 00:15 | |
| *** shakamunyi has joined #openstack-ironic | 00:23 | |
| *** ams0 has joined #openstack-ironic | 00:38 | |
| *** ams0 has quit IRC | 00:43 | |
| *** shakamunyi has quit IRC | 00:48 | |
| *** shakamunyi has joined #openstack-ironic | 01:13 | |
| *** lazy_prince is now known as killer_prince | 01:27 | |
| *** ams0 has joined #openstack-ironic | 01:38 | |
| *** Haomeng has quit IRC | 01:41 | |
| *** ams0 has quit IRC | 01:43 | |
| *** radsy has quit IRC | 01:55 | |
| *** Haomeng has joined #openstack-ironic | 01:55 | |
| *** Haomeng has quit IRC | 02:08 | |
| *** coolsvap|afk is now known as coolsvap | 02:35 | |
| *** shakamunyi has quit IRC | 02:57 | |
| *** Haomeng has joined #openstack-ironic | 03:01 | |
| *** blamar has quit IRC | 03:08 | |
| *** Haomeng has quit IRC | 03:34 | |
| *** ams0 has joined #openstack-ironic | 03:39 | |
| *** ams0 has quit IRC | 03:43 | |
| *** killer_prince is now known as lazy_prince | 04:07 | |
| *** ramineni has joined #openstack-ironic | 04:38 | |
| *** lazy_prince has quit IRC | 04:58 | |
| *** ams0 has joined #openstack-ironic | 05:08 | |
| *** Mikhail_D_ltp has joined #openstack-ironic | 05:08 | |
| *** pradipta_away is now known as pradipta | 05:16 | |
| *** ams0 has quit IRC | 05:26 | |
| *** ams0 has joined #openstack-ironic | 05:26 | |
| *** vinbs_ has joined #openstack-ironic | 05:30 | |
| *** ams0 has quit IRC | 05:30 | |
| *** rameshg87 has joined #openstack-ironic | 05:40 | |
| vinbs_ | hey, anybody has a working localrc/local.conf file for a real baremetal setup using ironic? | 05:47 |
|---|---|---|
| *** Mikhail_D_ltp has quit IRC | 05:54 | |
| *** ams0 has joined #openstack-ironic | 05:56 | |
| openstackgerrit | OpenStack Proposal Bot proposed a change to openstack/ironic: Imported Translations from Transifex https://review.openstack.org/88508 | 06:07 |
| *** Haomeng has joined #openstack-ironic | 06:23 | |
| *** Mikhail_D_wk has joined #openstack-ironic | 06:30 | |
| Mikhail_D_wk | Good morning all! :) | 06:31 |
| *** ams0 has quit IRC | 06:32 | |
| Haomeng | Mikhail_D_wk: morning:) | 06:33 |
| *** Nisha has joined #openstack-ironic | 06:44 | |
| *** lazy_prince has joined #openstack-ironic | 07:02 | |
| *** dshulyak has joined #openstack-ironic | 07:13 | |
| *** Manishanker has joined #openstack-ironic | 07:14 | |
| GheRivero | morning all! | 07:19 |
| *** stack__ has quit IRC | 07:22 | |
| *** stack__ has joined #openstack-ironic | 07:23 | |
| *** romcheg has joined #openstack-ironic | 07:29 | |
| Mikhail_D_wk | GheRivero: morning :) | 07:35 |
| *** ams0 has joined #openstack-ironic | 07:38 | |
| *** ndipanov has joined #openstack-ironic | 07:47 | |
| *** stack__ has quit IRC | 07:53 | |
| *** openstackgerrit has quit IRC | 07:53 | |
| *** stack__ has joined #openstack-ironic | 07:54 | |
| *** jistr has joined #openstack-ironic | 07:57 | |
| *** mrda is now known as mrda-away | 07:58 | |
| *** stack__ has quit IRC | 08:07 | |
| *** stack__ has joined #openstack-ironic | 08:10 | |
| dtantsur | Morning Ironic, morning Haomeng, GheRivero, Mikhail_D_wk! | 08:12 |
| Haomeng | dtantsur: morning:) | 08:13 |
| *** foexle has joined #openstack-ironic | 08:19 | |
| Mikhail_D_wk | dtantsur: morning :) | 08:27 |
| *** ams0 has quit IRC | 08:27 | |
| Nisha | dtantsur: Morning | 08:31 |
| Nisha | please confirm if its you who had given comments in https://review.openstack.org/#/c/89500/4 | 08:31 |
| *** ams0 has joined #openstack-ironic | 08:34 | |
| *** ams0 has quit IRC | 08:39 | |
| *** ams0 has joined #openstack-ironic | 08:40 | |
| dtantsur | Nisha, hi! That was me, right :) | 08:43 |
| *** ams0 has quit IRC | 08:44 | |
| *** ams0_ has joined #openstack-ironic | 08:47 | |
| Nisha | Hi Dtantsur, | 08:51 |
| Nisha | regarding comments on ribcl.py | 08:51 |
| Nisha | dtantsur: one of the comment was "If I understand correctly, _validate_message either returns `message` which always has `tag` attribute or None. Maybe just check for None here?" | 08:52 |
| dtantsur | aha, sure. Should I clarify? | 08:52 |
| Nisha | No...Actually the function will never return None. it will be either the message or it will response from ILO under a tag | 08:53 |
| Nisha | For example: if the message is "Host power is already ON.", it returns this message | 08:54 |
| dtantsur | Nisha, line 225: return None | 08:54 |
| Nisha | this is abt function _parse_output() and not _validate_message() | 08:56 |
| dtantsur | ok, I see, I didn't notice reassigning `message` variable (that's why I would prefer calling new variable `text` or anything). You can ignore this. Still this function does return None | 08:56 |
| Nisha | No | 08:56 |
| dtantsur | I understand, I was talking about line #170, where you check result of call to _validate_message | 08:57 |
| Nisha | that is explicitly checked against None, correct | 08:57 |
| dtantsur | Now I understand and you can disregard my comment on line #170, except for you'd better leave a comment there, explaining what _validate_message returns and in what case | 08:58 |
| Nisha | So i can ignore this comment? | 08:58 |
| dtantsur | yes | 08:58 |
| Nisha | In the same function, parse_output(), there is a comment on line 171 | 08:58 |
| Nisha | "This code looks like you only return the _last_ value of 'd', i.e. the last XML, right? If so, please put the explanation in the docstring." | 08:59 |
| *** ams0_ has quit IRC | 08:59 | |
| dtantsur | Nisha, yes, this is probably correct code, but it needs some clarification | 08:59 |
| dtantsur | as you assign to variable every loop iteration, only the last value will be used, right? | 08:59 |
| Nisha | Yes, this is correct code. The response xml from ILO is a group of xmls\ | 09:00 |
| *** ams0 has joined #openstack-ironic | 09:00 | |
| Nisha | No, it either returns actual response for the command executed or any message received | 09:01 |
| Nisha | the actual response is never the last XML | 09:02 |
| Nisha | Let me know if that clarifies | 09:03 |
| dtantsur | Nisha, ok, you have multiple XMLs, on every loop you assign variables d and resp_message, if there are suitable values, right? So the result will be one of the last assignment | 09:03 |
| dtantsur | you don't break loop after finding appropriate d or resp_message, so they will retain their last values | 09:04 |
| dtantsur | while that may be the correct behavior, I'd like to see clarification in docstring on how exactly this works | 09:04 |
| *** ams0 has quit IRC | 09:05 | |
| Nisha | wait, could you go to /opt/stack/ironic/ironic/tests/drivers/ilo/utils.py | 09:05 |
| *** ams0 has joined #openstack-ironic | 09:05 | |
| Nisha | and see example response xmls from ILO | 09:05 |
| Nisha | d will never have two values, it can be only one value even if i loop without breaking | 09:06 |
| *** ams0_ has joined #openstack-ironic | 09:07 | |
| Nisha | in an non-error case, the actual response is not the last one and is fine. | 09:09 |
| Nisha | in error case, the error message string is the last one | 09:09 |
| dtantsur | Nisha, than why not just break on success? This will be more obvious to a reader and save you some CPU time :) | 09:09 |
| Nisha | No, we need to loop till end to find out if there is any error returned by ILO | 09:09 |
| *** ams0 has quit IRC | 09:10 | |
| Nisha | In some cases like when u just set the value, there is no actual response from the ILO its only first xml repeated till last | 09:11 |
| Nisha | there is no point where you can determine whether its successful or not unless u loop till end | 09:11 |
| dtantsur | Nisha, ok, all this make sense. Could you please put into docstring /comments whatever you told me now? | 09:12 |
| Nisha | ok | 09:12 |
| Nisha | thanks for the review | 09:12 |
| agordeev | morning Ironic :) | 09:13 |
| dtantsur | Nisha, u'r welcome | 09:13 |
| dtantsur | agordeev, morning | 09:13 |
| Mikhail_D_wk | agordeev: morning :) | 09:14 |
| romcheg | Morning folks | 09:15 |
| dtantsur | romcheg, morning | 09:16 |
| agordeev | morning dtantsur Mikhail_D_wk romcheg Nisha | 09:22 |
| *** ifarkas has joined #openstack-ironic | 09:40 | |
| dtantsur | adam_g, around? I'm stuck with http://logs.openstack.org/87/85387/20/check/check-tempest-dsvm-virtual-ironic/679c5e9/, this seems to be a bunch of failures, not related too much to my change | 09:53 |
| dtantsur | Btw, folks, are we having a meeting today? I don't understand the outcome of ML discussion :) | 09:57 |
| *** Nisha has quit IRC | 10:07 | |
| *** coolsvap is now known as coolsvap|afk | 10:40 | |
| agordeev | dtantsur: hah, i'm interested in getting the answer for your question too :) | 10:52 |
| *** ams0_ has quit IRC | 11:02 | |
| *** ams0 has joined #openstack-ironic | 11:02 | |
| *** ramineni has quit IRC | 11:05 | |
| *** ams0 has quit IRC | 11:06 | |
| NobodyCam | good morning Ironic | 11:31 |
| agordeev | NobodyCam: good morning! | 11:31 |
| NobodyCam | morning agordeev | 11:32 |
| NobodyCam | :) | 11:32 |
| rameshg87 | good morning NobodyCam: | 11:36 |
| dtantsur | morning NobodyCam! | 11:40 |
| NobodyCam | morning rameshg87 and dtantsur :) | 11:43 |
| rameshg87 | NobodyCam, do you have a few minutes. i just wanted to talk regarding the ilo deploy that we are planning. wanted to check with you since you were involved in the review last time | 11:44 |
| rameshg87 | others are also welcome to tell their thoughts :-) | 11:44 |
| NobodyCam | rameshg87: give me a couple of minutes to get setup and get some coffee if thats ok | 11:45 |
| rameshg87 | NobodyCam: sure :-) | 11:46 |
| NobodyCam | grrr vpn wont connect over my version wifi :( | 11:57 |
| NobodyCam | well, I'll have to think of how to get around that | 12:00 |
| *** ams0 has joined #openstack-ironic | 12:00 | |
| NobodyCam | rameshg87: what about the ilo deploy did you want to chat about;;:) | 12:01 |
| *** ams0_ has joined #openstack-ironic | 12:02 | |
| Shrews | morning all | 12:03 |
| NobodyCam | good morning Shrews :) | 12:03 |
| Shrews | NobodyCam: morning. where abouts are you today? | 12:05 |
| rameshg87 | yes NobodyCam: :-) | 12:06 |
| *** ams0 has quit IRC | 12:06 | |
| NobodyCam | Shrews: south atlanata rv park :-p | 12:06 |
| rameshg87 | earlier, we had posted a patch and tried to submit ilo deploy for icehouse release | 12:06 |
| NobodyCam | rameshg87: ack I recall that | 12:07 |
| rameshg87 | the solution was to embed the disk image within the virtual media and then attach it to the BMC (iLO) of the proliant server | 12:07 |
| Shrews | NobodyCam: If you get the chance, would love to see 90429 pushed through today. Already has 3 +2's. | 12:08 |
| *** Nisha has joined #openstack-ironic | 12:09 | |
| NobodyCam | Shrews: 3??? /me looks | 12:09 |
| rameshg87 | but then there were concerns that retrieving the image through management network would be much slower | 12:09 |
| Nisha | dtantsur: Could you please look at the the review comments resolution comments? | 12:10 |
| rameshg87 | so we were just thinking that we could move the downloading of the image to the data network | 12:10 |
| rameshg87 | the baremetal node can directly download the image from glance using curl and then write it on the disk using the data network | 12:10 |
| rameshg87 | NobodyCam: any thoughts on that ? | 12:10 |
| NobodyCam | rameshg87: yes, if I recall correctly the bmc was on like a 10/100/1000 connection where the other interfaces were 10,000 | 12:11 |
| rameshg87 | we would just use the virtual media to boot up the kernel and ramdisk | 12:11 |
| rameshg87 | NobodyCam: yes | 12:11 |
| dtantsur | Nisha, re returning None: if you think you'll never receive malformed response (like no xml's) - it's ok for me. I leave it up to you. | 12:11 |
| NobodyCam | rameshg87: I like that. I believe the ipa folks are working towards that too | 12:12 |
| Nisha | Ok Thanks. Shall we go ahead for posting new patch? | 12:12 |
| dtantsur | yep | 12:12 |
| rameshg87 | NobodyCam: yeah, i know, but IPA I think is not ready yet .. am i correct ? | 12:12 |
| Nisha | Thank you. If any comments let me know, so that i can resolve comments if any. | 12:13 |
| NobodyCam | rameshg87: correct. but I beleieve that is the correct solution | 12:13 |
| NobodyCam | (pulling the image from the node) | 12:13 |
| rameshg87 | NobodyCam: we could add a new element in diskimage builder for that and use it there. | 12:14 |
| rameshg87 | NobodyCam: we had a new element ready which does so, that would require minimal changes to pull the image from glance over the data network | 12:14 |
| rameshg87 | NobodyCam: we were thinking for now if we can add a new element in diskimage-builder for this until IPA gets ready | 12:15 |
| NobodyCam | rameshg87: could the current deolpy-ironic element be modified to support both methods | 12:15 |
| *** ams0_ has quit IRC | 12:15 | |
| *** ams0 has joined #openstack-ironic | 12:16 | |
| *** rloo has joined #openstack-ironic | 12:16 | |
| rameshg87 | NobodyCam: do you mean at a solution level, a single kernel/ramdisk can support both the deploy method based on arguments or something ? | 12:16 |
| rameshg87 | like a single kernel/ramdisk can be used for both pxe deploy method (which uses iscsi from conductor node) and ilo deploy (which will use direct download of image from bmc) | 12:17 |
| NobodyCam | no that I read it back.. .that may bloat the deploy ramdisk | 12:18 |
| *** jdob has joined #openstack-ironic | 12:18 | |
| NobodyCam | rameshg87: so you are thinking a new deploy-ironic-ilo type element | 12:19 |
| rameshg87 | NobodyCam: yes, that was my thought | 12:19 |
| *** Nisha has quit IRC | 12:19 | |
| rameshg87 | NobodyCam: that would have an init script which downloads the image using curl and then writes it to the disk | 12:20 |
| NobodyCam | rameshg87: it would also partition the disk too? | 12:20 |
| NobodyCam | or whole disk images only | 12:20 |
| rameshg87 | NobodyCam: currently we have whole disk image support only. but it would modify the partition table to create swap and other required partitions on the machine | 12:21 |
| *** Tushar_ has joined #openstack-ironic | 12:24 | |
| *** Tushar_ has left #openstack-ironic | 12:24 | |
| NobodyCam | rameshg87: I think you'll get better acceptance if it supported more then just whole disk images | 12:26 |
| rameshg87 | NobodyCam: usage-wise, is fs-images preferred more than whole disk images ? | 12:27 |
| rameshg87 | NobodyCam: is booting from the local disk not preferred or something ? | 12:27 |
| NobodyCam | we need to support both. | 12:27 |
| NobodyCam | rameshg87: booting from local disk can hae issues, say in a DC power failure, we don't want the customer node comming up before the mangmt nodes | 12:29 |
| *** ams0_ has joined #openstack-ironic | 12:30 | |
| NobodyCam | I thought we were looking at pxe booting and returning localboot 0 from pxe? | 12:30 |
| NobodyCam | (node zero may not be correct, just an example) | 12:31 |
| rameshg87 | NobodyCam: ah, what does "localboot 0" mean ? boot from first partition ? | 12:31 |
| rameshg87 | NobodyCam: okay, got it .. | 12:33 |
| *** ams0 has quit IRC | 12:33 | |
| NobodyCam | it tells the node to boot from local disk not network | 12:33 |
| NobodyCam | :-p | 12:33 |
| rameshg87 | NobodyCam: we would like to create a pxe-less environment. the problem statement was to avoid pxe because some customers want to avoid it. | 12:34 |
| NobodyCam | is ok to local boot the node ... we just need to tell it to do so | 12:34 |
| NobodyCam | rameshg87: can ilo control the boot device | 12:34 |
| rameshg87 | NobodyCam: yes it can .. | 12:34 |
| NobodyCam | ie have ironic tell it to boot from local disk (like pxe) | 12:35 |
| rameshg87 | NobodyCam: yes, ilo can control the boot order | 12:36 |
| *** vinbs has joined #openstack-ironic | 12:36 | |
| rameshg87 | NobodyCam: so let me just check if i got it right | 12:36 |
| * Shrews summons openstackgerrit bot | 12:37 | |
| *** vinbs_ has quit IRC | 12:37 | |
| rameshg87 | NobodyCam: for pxe, the tftp server on the ironic controls the boot of the bm node. so when ironic conductor node is down, the bm node won't boot. when ironic conductor comes up, the bm node can boot. am i correct ? | 12:37 |
| NobodyCam | yes | 12:38 |
| rameshg87 | NobodyCam: so, ironic doesn't directly control the boot of the bm node. i mean ironic doesn't directly ask the bm node to boot. | 12:38 |
| NobodyCam | that way customer node can not come up before the mngt nodes | 12:38 |
| *** vinbs_ has joined #openstack-ironic | 12:39 | |
| NobodyCam | ? not sure I get the last statment. | 12:40 |
| rameshg87 | NobodyCam: okay, let me put it this way | 12:40 |
| lazy_prince | in my opinion, if a customer wants to avoid pxe, then he is also taking a conscious risk of booting bm before mgmt can come up... | 12:40 |
| *** linggao has joined #openstack-ironic | 12:40 | |
| *** vinbs has quit IRC | 12:40 | |
| *** vinbs_ is now known as vinbs | 12:41 | |
| rameshg87 | tftp boot happens automatically without involvement of ironic - when ironic node comes up, bm node can also boot | 12:41 |
| NobodyCam | lazy_prince: that is a hugh risk | 12:42 |
| rameshg87 | similarly for ilo (one alternative) - ilo could host a virtual media containing "localboot 0" on ISOLINUX and then make the bm node boot from virtual media everytime. | 12:42 |
| rameshg87 | in that way, the node won't boot because the virtual media won't be accessible until the ironic node is up | 12:42 |
| NobodyCam | rameshg87: that sounds good to me :) | 12:42 |
| rameshg87 | NobodyCam: and the single virtual media image can be used for all bm nodes because it will just have "localboot 0" in it | 12:43 |
| NobodyCam | yep | 12:44 |
| lazy_prince | the only catch is that if bm boot can fallback to localdisk on failure. | 12:44 |
| *** romcheg1 has joined #openstack-ironic | 12:44 | |
| rameshg87 | lazy_prince: how does pxe handle it now ? | 12:45 |
| NobodyCam | lazy_prince: yes we will have to have instructions telling admins to ensure correct fallback config | 12:45 |
| NobodyCam | rameshg87: now everythong pxe boots | 12:45 |
| NobodyCam | we set netboot only .. no fall back | 12:46 |
| rameshg87 | NobodyCam: do we remove local hdd from the boot order ? | 12:46 |
| NobodyCam | currently yes | 12:46 |
| rameshg87 | NobodyCam: if pxe boot fails (assume ironic node is just coming up after a power failure), will the bm node try booting from pxe repeatedly again and again ? | 12:46 |
| NobodyCam | yes | 12:46 |
| NobodyCam | (most I have testeded do) | 12:47 |
| lazy_prince | so right now, only PXE supplies the boot image and since there is no boot image on disk, the only option to boot is net boot.. | 12:47 |
| NobodyCam | maybe some older one only try once | 12:47 |
| NobodyCam | lazy_prince: :) yep | 12:47 |
| rameshg87 | NobodyCam: okay, we can have a similar logic for ilo also. we could just have virtual media boot enabled :-) | 12:47 |
| lazy_prince | but with whole disk images with boot loaders, the entire game changes.. | 12:48 |
| NobodyCam | rameshg87: sweet :) | 12:48 |
| lazy_prince | if pxe fails, it can fall back to disk and continue boot process from disk.. | 12:48 |
| NobodyCam | lazy_prince: if the admin can disable that fallbace | 12:48 |
| NobodyCam | fallback | 12:48 |
| NobodyCam | gah .. need more coffee.. | 12:49 |
| NobodyCam | lazy_prince: the admin should disable that fallback | 12:49 |
| rameshg87 | NobodyCam: :-) | 12:49 |
| rameshg87 | NobodyCam: is there an option right now ? | 12:49 |
| NobodyCam | option for? | 12:50 |
| rameshg87 | option for configuring fallback for nodes | 12:50 |
| NobodyCam | rameshg87: that would be in the nodes bios | 12:50 |
| rameshg87 | NobodyCam: just wondering if ironic does that (in maintenence mode or something :) ) | 12:51 |
| lazy_prince | well.. it depends on the system BIOS and the fact that whether the BISO allows BOOT order configuration changes/reordering. if changes are allowed then admin can control it but if only reordering is allowed, then it will fallback to local disk.. as it can not be removed from the boot order... | 12:51 |
| NobodyCam | rameshg87: not yet | 12:51 |
| *** pradipta is now known as pradipta_away | 12:51 | |
| NobodyCam | lazy_prince: :( that sounds like the old AMI bios.. is that still the case . I have not seen that in sometime now | 12:52 |
| NobodyCam | lazy_prince: yes That is a issue | 12:53 |
| rameshg87 | lazy_prince: just checking if we can remove from the boot order. i hope it should | 12:53 |
| lazy_prince | Well.. i can still find some HP servers where you can only reorder boot entries in BIOS and you cannot disable them... | 12:54 |
| lazy_prince | one such example is HP Moonshot servers.. | 12:54 |
| NobodyCam | we should chat about that at the summit.. | 12:54 |
| NobodyCam | oh really | 12:54 |
| NobodyCam | I have not been able to access the MS servers | 12:54 |
| NobodyCam | how do MoonShot server handle power outages? | 12:55 |
| rameshg87 | lazy_prince: yeah, i think the same is with hp proliant servers as well :-( | 12:55 |
| NobodyCam | or more correctly ... handle power on after power outage | 12:56 |
| lazy_prince | May be its fixed in the new firmware, but the one i have does not let me remove/disable boot entries.. | 12:56 |
| rameshg87 | lazy_prince: you can only reorder, but not remove anything from it | 12:56 |
| NobodyCam | hummm | 12:56 |
| NobodyCam | this will require more thought | 12:57 |
| lazy_prince | Thats what I was taliking about.. So with whole disk images with boot loader, its a different story.. | 12:57 |
| rameshg87 | lazy_prince: yeah, it is :-( | 12:57 |
| NobodyCam | ya | 12:57 |
| lazy_prince | may be we should meet at summit to discuss this at length.. | 12:58 |
| rameshg87 | lazy_prince: do you have other use-cases of whole disk images ? (like windows provisioning or something ? ) | 12:59 |
| NobodyCam | lazy_prince: +++ yes | 13:00 |
| rameshg87 | lazy_prince: NobodyCam: don't know if it's a bad idea | 13:00 |
| NobodyCam | rameshg87: are you going to be here at the summit? | 13:00 |
| rameshg87 | NobodyCam: no :-( | 13:01 |
| NobodyCam | rameshg87: ack... | 13:01 |
| lazy_prince | Yes.. Windows Provisioning is one such use case.. the other is to have a uniform way of booting Business critical machine in case PXE fails. (as pointed out by rameshg87 pointed out, customers having private cloud like to use PXE only for provisioning.) | 13:01 |
| *** ramineni has joined #openstack-ironic | 13:04 | |
| NobodyCam | lazy_prince: Business critical machine is why we do need to support local booting... | 13:04 |
| NobodyCam | yes | 13:05 |
| lazy_prince | agree.. | 13:06 |
| * lazy_prince T time.. | 13:06 | |
| lazy_prince | brb | 13:06 |
| *** romcheg has quit IRC | 13:07 | |
| rameshg87 | NobodyCam: let me think about supporting fs images | 13:11 |
| lazy_prince | back.. | 13:11 |
| NobodyCam | rameshg87: yes.. I am too... we will also talk at the summit about it | 13:11 |
| NobodyCam | wb lazy_prince | 13:11 |
| NobodyCam | oh are we going to have a meeting today? | 13:13 |
| *** foexle has quit IRC | 13:13 | |
| dtantsur | NobodyCam, everyone seems to wonder, if we have a meeting today :) | 13:13 |
| NobodyCam | devananda: and lucas are out... we have the summit in a week. | 13:14 |
| NobodyCam | I am happy to chair a meeting, just not sure how effective we'll be | 13:15 |
| *** jbjohnso has joined #openstack-ironic | 13:16 | |
| NobodyCam | we'll wait for the west costers to wakeup and come online... we'll ask again... I can go either way | 13:17 |
| *** foexle has joined #openstack-ironic | 13:19 | |
| *** datajerk has joined #openstack-ironic | 13:22 | |
| linggao | Morning, NobodyCam | 13:24 |
| NobodyCam | good morning linggao :) | 13:25 |
| rloo | morning NobodyCam, linggao. | 13:26 |
| NobodyCam | good morning rloo :) | 13:26 |
| linggao | my 2 design session propossal are denied and are combined into the Ironic Juno future sessions. Do I need to prepare some slides or it is just free talk. | 13:27 |
| linggao | morning rloo. | 13:27 |
| NobodyCam | linggao: free talk is prob fine.. but remember a picture is worth a 1000 words | 13:27 |
| NobodyCam | linggao: maybe not a full slide deck but some ref slides | 13:28 |
| NobodyCam | might be nice | 13:28 |
| rloo | NobodyCam: I'm going to take today off . Might ping you tomorrow (am hoping to try to get some reviews done tomorrow.) | 13:28 |
| NobodyCam | rloo: ack... I expect this to be a light week.. as many folks will be traveling | 13:29 |
| linggao | where should I put the slides? | 13:29 |
| NobodyCam | rloo: so I can put you in the NO meeting colloum | 13:29 |
| NobodyCam | linggao: thum drive? | 13:29 |
| NobodyCam | :-p | 13:30 |
| rloo | NobodyCam: yes, NO meeting for me today. (I replied to deva's email about that.) I say cancel it ;) | 13:30 |
| NobodyCam | :) | 13:30 |
| NobodyCam | so thats 3 for NO | 13:30 |
| *** matty_dubs|gone is now known as matty_dubs | 13:31 | |
| rloo | NobodyCam: I just looked at the emails. matty_dubs said he'd attend, GheRivero said he's fine skipping it. | 13:31 |
| *** jgrimm has quit IRC | 13:32 | |
| rloo | linggao: I suggest being prepared. bring your own laptop + slides. + something to connect your laptop to projector. | 13:32 |
| NobodyCam | ya... the trend is leaning towards No meeting... But if folks have stuff to chat about I'm happy to chair | 13:32 |
| NobodyCam | rloo: ++ much better then thumb drive | 13:33 |
| rloo | NobodyCam, linggao: also a good idea to put slides on thumb drive. You never know... | 13:33 |
| NobodyCam | though will prob only bring my ipad :-p | 13:33 |
| rloo | NobodyCam: I haven't yet decided what I'll bring.. maybe pen+paper! Ok, I'm outta here. | 13:34 |
| *** rloo has left #openstack-ironic | 13:34 | |
| NobodyCam | have a good day off rloo | 13:34 |
| *** openstackgerrit has joined #openstack-ironic | 13:36 | |
| *** blamar has joined #openstack-ironic | 13:37 | |
| openstackgerrit | Ramakrishnan G proposed a change to openstack/ironic: Add IloDriver and its utils https://review.openstack.org/89500 | 13:38 |
| NobodyCam | bbt...brb | 13:39 |
| *** rameshg87 has quit IRC | 13:39 | |
| *** rameshg87 has joined #openstack-ironic | 13:42 | |
| linggao | Thanks rloo. Enjoy the day off. | 13:43 |
| NobodyCam | anyone worked with Suse? | 13:46 |
| *** vinbs has quit IRC | 13:47 | |
| NobodyCam | I think I am experencing a issue with this line: https://github.com/openstack/ironic/blob/master/ironic/drivers/modules/deploy_utils.py#L133 | 13:48 |
| *** openstackgerrit has quit IRC | 13:51 | |
| jroll | morning ironic :) | 13:53 |
| NobodyCam | good morning jroll | 13:53 |
| *** ams0_ has quit IRC | 13:54 | |
| dtantsur | morning, jroll | 13:54 |
| *** ams0 has joined #openstack-ironic | 13:54 | |
| NobodyCam | jroll: thoughts on a meeting today? | 13:56 |
| *** ams0 has quit IRC | 13:57 | |
| *** ams0 has joined #openstack-ironic | 13:57 | |
| jroll | NobodyCam: JayF_ JoshNang and I should be around, if people want to talk about something | 13:57 |
| *** ams0 has quit IRC | 13:58 | |
| jroll | but our weekly tech talk is during ironic meeting every week so I'm always happy to say no to a meeting :) | 13:58 |
| NobodyCam | jroll: we're trying to see if there is things folks want to chat about... devananda, lucas, and rloo are out today | 13:58 |
| *** ams0 has joined #openstack-ironic | 13:58 | |
| jroll | I don't think I have anything besides status updates | 13:58 |
| NobodyCam | :) | 13:59 |
| jroll | if I have questions I would rather just talk here | 13:59 |
| jroll | I wonder if the other guys in my team would want to talk more about network isolation, but I don't think the blueprint etc is ready yet | 14:00 |
| dtantsur | Folks, my patch celebrated 1-month birthday this weekend: https://review.openstack.org/#/c/85387/. As a birthday gift, could you review it, please? :) | 14:02 |
| *** datajerk has quit IRC | 14:05 | |
| *** datajerk has joined #openstack-ironic | 14:05 | |
| NobodyCam | dtantsur: nit in the commit message. All images older than configurable TTL are deleted! that is a very scarry line | 14:12 |
| dtantsur | NobodyCam, well, it's the 2nd point in algorithm | 14:12 |
| dtantsur | so I assumed here: everything = everything that reached this point | 14:13 |
| NobodyCam | lol I'm looking now | 14:13 |
| rameshg87 | NobodyCam: got a quick question | 14:13 |
| NobodyCam | rameshg87: I have a quick answer: 42 | 14:13 |
| NobodyCam | :-p | 14:13 |
| rameshg87 | NobodyCam: :-) | 14:14 |
| NobodyCam | rameshg87: sure | 14:14 |
| *** Manishanker has quit IRC | 14:14 | |
| * matty_dubs hopes the question was "How many dollars are you going to give matty_dubs?" | 14:14 | |
| rameshg87 | NobodyCam: sometime back we were discussing on a reliable way to avoid the machine getting booted everytime, but able to boot if required | 14:14 |
| rameshg87 | NobodyCam: can we think of installing the bootloader on the partition instead of MBR | 14:14 |
| NobodyCam | matty_dubs: lol :) and good morning | 14:14 |
| matty_dubs | Haha, same to you | 14:14 |
| rameshg87 | NobodyCam: and use chain.c32 module of syslinux to boot the first partition | 14:15 |
| NobodyCam | rameshg87: so it wouldn't be found by a fallback | 14:15 |
| NobodyCam | that could work | 14:16 |
| rameshg87 | NobodyCam: do you mean it wouldn't be found by the BIOS ? If so, yes | 14:16 |
| NobodyCam | ya | 14:16 |
| NobodyCam | rameshg87: :) | 14:16 |
| rameshg87 | NobodyCam: and if the user wanted to boot the machine urgently without ironic, they could use any boot software to boot from that partition instead | 14:16 |
| rameshg87 | NobodyCam: but it would never get booted by default automatically :-) | 14:17 |
| NobodyCam | rameshg87: that makes sense to me :) | 14:17 |
| rameshg87 | NobodyCam: okay, if so pxe could also make use of it assuming that we create the root partition first always on the disk | 14:18 |
| *** rwsu has joined #openstack-ironic | 14:18 | |
| dtantsur | rameshg87, I think we're switching to creating root partition the last, no? | 14:18 |
| * dtantsur brb, reboot | 14:18 | |
| NobodyCam | ya | 14:18 |
| NobodyCam | root is last now so it can be expaned | 14:18 |
| rameshg87 | dtantsur: okay, i see it now .. | 14:19 |
| rameshg87 | the logic is like: if ephemeral partition: then (root partition = 3rd) otherwise (root partition = 1st) | 14:20 |
| rameshg87 | hmm..that can cause trouble :-) | 14:20 |
| *** davidlenwell has quit IRC | 14:21 | |
| *** davidlenwell has joined #openstack-ironic | 14:21 | |
| rameshg87 | NobodyCam: dtantsur: so does that mean that we won't have a fixed partition number for root partition ? | 14:24 |
| dtantsur | rameshg87, no, it will be the last: https://review.openstack.org/#/c/90675/ | 14:24 |
| NobodyCam | dtantsur: question on the cache tests | 14:27 |
| dtantsur | ? | 14:27 |
| NobodyCam | can you add a test for what would happen if the fs system is out of space and unable to pull a image | 14:27 |
| NobodyCam | or did I mass that one | 14:28 |
| dtantsur | NobodyCam, I can, but this patch does change anything with regards to it :) What do you want to verify? | 14:28 |
| NobodyCam | looking to see is thats already covered | 14:29 |
| dtantsur | NobodyCam, how to handle disk space is an interesting point, but as it's used to be a problem before my patch (though on smaller scale), I suggest working on it separately | 14:30 |
| dtantsur | Actually I can take care of it, i.e. purge cache if we have problems with disk space | 14:30 |
| *** jgrimm has joined #openstack-ironic | 14:30 | |
| NobodyCam | dtantsur: I am good with handling in another patch | 14:31 |
| dtantsur | ack, I'll file a bug and start working. Anything else? | 14:31 |
| NobodyCam | but ya I'm not seeing where we check oh look I'm pull a 15 Gib image and i only have 10Gb worth of sapce | 14:32 |
| dtantsur | NobodyCam, we never did, I think | 14:32 |
| NobodyCam | dtantsur: :-p doh | 14:32 |
| NobodyCam | hehehe | 14:32 |
| NobodyCam | ya | 14:32 |
| * NobodyCam also notes he have not run out of space on a production server in many years | 14:33 | |
| NobodyCam | personal laptop now thats another story | 14:33 |
| dtantsur | NobodyCam, https://bugs.launchpad.net/ironic/+bug/1316168 | 14:34 |
| NobodyCam | dtantsur: :) | 14:34 |
| NobodyCam | dtantsur: +2'd :) | 14:37 |
| NobodyCam | though I did not test locally | 14:37 |
| NobodyCam | brb making more coffee | 14:38 |
| dtantsur | thanks :) | 14:38 |
| dtantsur | anyone else mind looking? | 14:38 |
| *** rameshg87 has left #openstack-ironic | 14:40 | |
| *** foexle has quit IRC | 14:45 | |
| *** foexle has joined #openstack-ironic | 14:46 | |
| dtantsur | romcheg1, Haomeng ^^^ ? | 14:47 |
| NobodyCam | just checking sofar we have 4 no meeting votes... anyone have an opposing vote? | 14:47 |
| * romcheg1 is looking | 14:47 | |
| NobodyCam | morning romcheg1 :) | 14:47 |
| romcheg1 | Morning NobodyCam! | 14:47 |
| NobodyCam | how are "things" going in your neck of the woods | 14:48 |
| dtantsur | NobodyCam, I also vote for skipping, meeting at 9pm is exhaustive :) | 14:48 |
| NobodyCam | 5-no / 0-yes / 1-maybe | 14:49 |
| romcheg1 | +1 to skipping. The info I have will be discussed at the design session. | 14:51 |
| romcheg1 | Also 10PM is quite late :) | 14:51 |
| romcheg1 | NobodyCam: Is is 8AM for you when we have a meeting? | 14:53 |
| jroll | our meetings are noon on the west coast | 14:53 |
| jroll | 9am on the east coast | 14:54 |
| romcheg1 | Time zones suck :( | 14:54 |
| jroll | ^ | 14:54 |
| * jroll bbiab | 14:55 | |
| agordeev | romcheg1: 11PM is a bit later than 10 :) | 14:55 |
| dtantsur | ouch | 14:56 |
| matty_dubs | My vote was, "Meh, I'll attend if there is one, but I have nothing to discuss" | 14:57 |
| NobodyCam | gah sorry just lost power | 15:03 |
| NobodyCam | jroll: 3pm on east cost | 15:03 |
| NobodyCam | romcheg1: ya generally noon for me.. but I'm in Atlanta now | 15:04 |
| romcheg1 | NobodyCam: I was just thinking moving the meeting to some other time. But now I see that might be inconvenient for some other guys | 15:05 |
| NobodyCam | :) if we get a vote to have a meeting it would at regualr time :-p | 15:06 |
| NobodyCam | but its looking like no meeting | 15:06 |
| *** pradipta_away is now known as pradipta | 15:06 | |
| NobodyCam | sudo qemu-nbd -c /dev/nbd0 seed.qcow2 | 15:07 |
| NobodyCam | gah WW | 15:07 |
| romcheg1 | NobodyCam: is "gah WW" you password? :-P | 15:09 |
| NobodyCam | nope... Wrong Window | 15:09 |
| NobodyCam | gah is what I thought | 15:09 |
| NobodyCam | :-p | 15:09 |
| matty_dubs | 'sudo qemu-nbd -c /dev/nbd0 seed.qcow2' would be a much better password than 'gah WW' though | 15:11 |
| NobodyCam | lol | 15:12 |
| NobodyCam | oh fresh coffee is ready ...brb | 15:13 |
| dtantsur | romcheg1, any ideas on 85387? :) | 15:13 |
| romcheg1 | dtantsur: Still deploying... | 15:14 |
| dtantsur | ah, I see | 15:14 |
| romcheg1 | Internet is slow today. That's odd because there are only a few people in the office | 15:15 |
| dtantsur | It's weather, I think. What is weather like in your place? | 15:16 |
| romcheg1 | dtantsur: It's unexpectedly cold today | 15:18 |
| dtantsur | romcheg1, that's why bytes run too slowly :) | 15:19 |
| NobodyCam | well when it's cold I move slowly | 15:19 |
| NobodyCam | :-p | 15:19 |
| romcheg1 | Well, cold is wrong word for a +10, but it was about +20 the whole week | 15:19 |
| romcheg1 | dtantsur: Yeah, indeed :D | 15:20 |
| GheRivero | is the option to mark a review as WIP removed from the new gerrit? | 15:21 |
| *** linggao_ has joined #openstack-ironic | 15:21 | |
| NobodyCam | morning GheRivero :) | 15:22 |
| dtantsur | GheRivero, press review, then workflow -1 | 15:22 |
| dtantsur | btw morning, yeah :) | 15:22 |
| NobodyCam | ya | 15:22 |
| GheRivero | thanks @all | 15:23 |
| NobodyCam | GheRivero: your vote for meeting... we are at 6-No / 0-Yes / [1 or 2]-Maybe | 15:24 |
| *** linggao has quit IRC | 15:24 | |
| GheRivero | wow! 6-0 Let's say yes just to make up the final result :P | 15:25 |
| NobodyCam | hehehe | 15:26 |
| NobodyCam | why can I not change https://github.com/openstack/ironic/blob/master/ironic/drivers/modules/deploy_utils.py#L133 in my seed vm :-p | 15:30 |
| *** zdin0bot has joined #openstack-ironic | 15:32 | |
| *** foexle has quit IRC | 15:35 | |
| *** zdin0bot has quit IRC | 15:38 | |
| *** openstackgerrit has joined #openstack-ironic | 15:43 | |
| *** vinbs has joined #openstack-ironic | 15:43 | |
| *** coolsvap|afk is now known as coolsvap | 15:45 | |
| *** vinbs_ has joined #openstack-ironic | 15:46 | |
| NobodyCam | ok that was not my issue :( | 15:48 |
| *** vinbs has quit IRC | 15:49 | |
| *** vinbs_ is now known as vinbs | 15:49 | |
| openstackgerrit | Dirk Mueller proposed a change to openstack/ironic: Remove explicit dependency on amqplib https://review.openstack.org/92153 | 15:50 |
| romcheg1 | dtantsur: Worked for me | 15:58 |
| dtantsur | \o/ | 15:58 |
| *** matty_dubs is now known as matty_dubs|lunch | 16:00 | |
| NobodyCam | brb walkies time | 16:02 |
| *** ams0__ has joined #openstack-ironic | 16:09 | |
| *** ams0 has quit IRC | 16:09 | |
| comstud | can someone give me a WorkFlow+1 on this: | 16:12 |
| comstud | https://review.openstack.org/#/c/84823/ | 16:12 |
| comstud | It's got 2 +2's and i'm tired of rebasing | 16:12 |
| jroll | 08:03:50 NobodyCam | jroll: 3pm on east cost <- err yeah. time zones are hard :P | 16:13 |
| openstackgerrit | Dmitry Tantsur proposed a change to openstack/ironic: Add periodic task to check whether PXE image cache is up-to-date https://review.openstack.org/92115 | 16:16 |
| *** ramineni has quit IRC | 16:19 | |
| NobodyCam | :) | 16:22 |
| NobodyCam | comstud: it shows Review in Progress for me | 16:23 |
| comstud | because it's in progress until someone approves it? | 16:24 |
| NobodyCam | oh so youd like to see it land | 16:24 |
| comstud | correct | 16:24 |
| NobodyCam | you want a +a | 16:24 |
| dtantsur | Also, to be annoying enough today: can someone also approve https://review.openstack.org/#/c/85387/ ? It got 2 +2 :) | 16:24 |
| comstud | yeah, which is aka 'Workflow+1' now. | 16:24 |
| comstud | heh | 16:24 |
| comstud | it has 2 +2, but last person didn't WF+1 it | 16:25 |
| dtantsur | yeah, I thought 2nd person who +2 just gives W+1, isn't it the case? | 16:26 |
| NobodyCam | comstud: is there a bp / bug for the fixme in manager.py | 16:28 |
| comstud | hm, checking | 16:28 |
| comstud | don't remember what the fixme is | 16:29 |
| comstud | yeah, kind of a general bug for this cleanup | 16:29 |
| comstud | sec | 16:29 |
| comstud | NobodyCam: https://bugs.launchpad.net/ironic/+bug/1314732 | 16:30 |
| comstud | same bug as in the commit msg | 16:30 |
| *** mdenny has joined #openstack-ironic | 16:30 | |
| *** ams0__ has quit IRC | 16:31 | |
| *** newell_ has joined #openstack-ironic | 16:31 | |
| *** jistr has quit IRC | 16:31 | |
| *** vinbs has quit IRC | 16:35 | |
| NobodyCam | comstud: looks like 3 is the lucky number | 16:38 |
| *** matty_dubs|lunch is now known as matty_dubs | 16:44 | |
| *** linggao__ has joined #openstack-ironic | 16:46 | |
| *** linggao_ has quit IRC | 16:50 | |
| *** datajerk has quit IRC | 16:50 | |
| *** hemnafk is now known as hemna | 16:52 | |
| *** datajerk has joined #openstack-ironic | 16:54 | |
| openstackgerrit | A change was merged to stackforge/pyghmi: Avoid zombie sessions corrupting living sessions https://review.openstack.org/92131 | 16:56 |
| *** datajerk has quit IRC | 16:58 | |
| *** Mikhail_D_ltp has joined #openstack-ironic | 16:59 | |
| *** JayF_ is now known as JayF | 17:02 | |
| comstud | NobodyCam: ty | 17:07 |
| jroll | InstanceDeployFailure_Remote: RPC do_node_deploy called for b7d102df-ae29-437f-a050-5f9703a29ed9, but provision state is already deleted. | 17:08 |
| jroll | deleted nodes can't be deployed to? | 17:08 |
| jroll | or is provision state supposed to go from deleted to NULL somewhere? | 17:08 |
| *** ifarkas has quit IRC | 17:12 | |
| *** harlowja_away is now known as harlowja | 17:16 | |
| *** rushiagr has quit IRC | 17:19 | |
| *** Madasi has quit IRC | 17:19 | |
| *** openstackgerrit has quit IRC | 17:20 | |
| *** jrist has quit IRC | 17:20 | |
| *** rushiagr has joined #openstack-ironic | 17:22 | |
| *** Madasi has joined #openstack-ironic | 17:22 | |
| NobodyCam | hummm I would expect the latter.. but that may be a bug atm | 17:22 |
| *** jrist has joined #openstack-ironic | 17:22 | |
| *** blamar_ has joined #openstack-ironic | 17:22 | |
| *** blamar has quit IRC | 17:23 | |
| *** blamar_ is now known as blamar | 17:23 | |
| *** lazy_prince has quit IRC | 17:24 | |
| *** openstackgerrit has joined #openstack-ironic | 17:25 | |
| *** morgabra has quit IRC | 17:25 | |
| *** morgabra has joined #openstack-ironic | 17:27 | |
| *** dwalleck has joined #openstack-ironic | 17:29 | |
| *** zdiN0bot has joined #openstack-ironic | 17:32 | |
| *** dwalleck_ has joined #openstack-ironic | 17:32 | |
| *** Nisha has joined #openstack-ironic | 17:33 | |
| Nisha | Hi, Need one help | 17:35 |
| *** dwalleck has quit IRC | 17:35 | |
| Nisha | Getting failure for python 2.6 | 17:35 |
| NobodyCam | Hi Nisha | 17:35 |
| Nisha | AttributeError: _ElementInterface instance has no attribute 'iter' | 17:36 |
| NobodyCam | where, and with what | 17:36 |
| Nisha | NobodyCam: When i replace the function iter with getiterable, it works fine on 2.7 also | 17:37 |
| Nisha | but as per the python document the function is deprecated in python 2.7 | 17:37 |
| *** dwalleck_ has quit IRC | 17:38 | |
| *** datajerk has joined #openstack-ironic | 17:38 | |
| Nisha | Are these errors ignorable while submitting the patch? | 17:40 |
| NobodyCam | if 2.6 dont passs the patch wont land | 17:41 |
| NobodyCam | you can put the patch up and reviewers will comment | 17:41 |
| Nisha | The patch is posted already, https://review.openstack.org/#/c/89500/5 | 17:43 |
| Nisha | We saw the failure there only | 17:44 |
| jbjohnso | wow, I've been quite silly, I can automatically detect the correct serial port and baud rate to use to get serial via SOL | 17:44 |
| jbjohnso | if the vendor does it right... | 17:44 |
| jbjohnso | without a lot of fuss or muss.. | 17:44 |
| NobodyCam | its thats last be I would be worried about. | 17:45 |
| NobodyCam | if the vendor..... | 17:45 |
| Nisha | By substituting iter with getiterable i can make it pass on 2.6 and 2.7 but document declare it deprecated | 17:45 |
| jroll | Nisha: I think using getiterable is fine, since we don't support python 3 yet | 17:46 |
| jroll | there will be no python 2.8 where getiterable() does not exist | 17:47 |
| jroll | might be nice to do something like: `if six.PY2: # use getiterable(); else: # use iter() | 17:47 |
| jroll | in anticipation of python 3 support | 17:48 |
| jroll | Nisha: I think that might be the best solution, do you want me to comment on the review? | 17:48 |
| NobodyCam | yea I think six also has iterkeys() | 17:48 |
| jroll | NobodyCam: this is on an etree Element | 17:48 |
| jroll | https://docs.python.org/2/library/xml.etree.elementtree.html#xml.etree.ElementTree.Element.getiterator | 17:49 |
| NobodyCam | six.iterkeys(dictionary, **kwargs) | 17:49 |
| NobodyCam | Returns an iterator over dictionary‘s keys. This replaces dictionary.iterkeys() on Python 2 and dictionary.keys() on Python 3. kwargs are passed through to the underlying method. | 17:49 |
| jroll | this isn't a dict :) | 17:49 |
| NobodyCam | doh | 17:49 |
| Nisha | its creation of element tree | 17:50 |
| Nisha | Ok, I would try to use "`if six.PY2: # use getiterable(); else: # use iter()" | 17:51 |
| NobodyCam | yep... humm stackoverflow has a intrestring solution: http://stackoverflow.com/questions/7616800/elementtrees-iter-equivalent-in-python2-6 | 17:51 |
| *** dividehex has quit IRC | 17:51 | |
| Nisha | But do you know how could we test on python 2.6 before posting the patch? My devstack has 2.7 installed | 17:51 |
| jroll | you can run: tox -epy26 | 17:52 |
| NobodyCam | jroll: ++ | 17:52 |
| jroll | as far as functional testing, I don't know | 17:52 |
| jroll | :) | 17:52 |
| Nisha | Ok. Thanks for the suggestions. Would do the required changes and post another patch. :) | 17:53 |
| jroll | you're welcome :) | 17:53 |
| openstackgerrit | A change was merged to openstack/ironic: Add create() and destroy() to Node https://review.openstack.org/84823 | 17:54 |
| NobodyCam | comstud: ^^^^ \o/ | 17:54 |
| NobodyCam | :-p | 17:54 |
| *** Tushar_ has joined #openstack-ironic | 17:59 | |
| *** dividehex has joined #openstack-ironic | 17:59 | |
| matty_dubs | NobodyCam: What's the verdict on a meeting? | 18:00 |
| matty_dubs | boris-42: ping; Rally question if you're around | 18:01 |
| NobodyCam | 7 to 0 for no meeting | 18:01 |
| boris-42 | matty_dubs sure I am here | 18:01 |
| boris-42 | matty_dubs always here=) | 18:01 |
| matty_dubs | boris-42: :) | 18:01 |
| matty_dubs | boris-42: So, I've been playing around with Rally, and got some very basic Ironic tests working (yay!) | 18:01 |
| * NobodyCam setp up random ping bot for boris-42 | 18:01 | |
| matty_dubs | boris-42: But, it expects my changes to be in /usr/lib/python2.7/site-packages/rally/, for osclients.py, etc. | 18:02 |
| boris-42 | NobodyCam lol=) | 18:02 |
| NobodyCam | s/setp/sets/ | 18:02 |
| NobodyCam | :-p | 18:02 |
| matty_dubs | Is there a way to make it look in my git checkout instead? | 18:02 |
| boris-42 | matty_dubs yep I know about tat | 18:02 |
| matty_dubs | So I should just copy stuff across for now? | 18:02 |
| boris-42 | matty_dubs https://review.openstack.org/#/c/72679/ | 18:02 |
| matty_dubs | Ah-ha! | 18:03 |
| boris-42 | matty_dubs you will be able to put in ~/.rally/plugins or /etc/rally/plugins/ directories | 18:03 |
| boris-42 | matty_dubs btw why you didn't contribute them?) | 18:03 |
| boris-42 | matty_dubs btw as well you can re-run python setup.py install | 18:03 |
| *** jbjohnso has quit IRC | 18:03 | |
| matty_dubs | Yeah, that's what I had been doing | 18:04 |
| boris-42 | matty_dubs i know it's awful a bit=) | 18:04 |
| matty_dubs | I just wasn't sure if I was missing a flag | 18:04 |
| matty_dubs | Haha | 18:04 |
| matty_dubs | boris-42: I'll submit them, but right now it's kind of rough | 18:04 |
| boris-42 | matty_dubs just submit and you'll get free reviews=) | 18:04 |
| boris-42 | matty_dubs and probably advices=) | 18:04 |
| matty_dubs | Haha, okay | 18:05 |
| *** ifarkas has joined #openstack-ironic | 18:08 | |
| NobodyCam | brb quick walkies.... | 18:09 |
| openstackgerrit | Jay Faulkner proposed a change to openstack/ironic-python-agent: Uninstall more buildtime only deps and clean lists https://review.openstack.org/92178 | 18:09 |
| *** pradipta is now known as pradipta_away | 18:10 | |
| boris-42 | matty_dubs i'll push rally team to merge that patch | 18:11 |
| boris-42 | matty_dubs without docs | 18:11 |
| matty_dubs | boris-42: It's not too big of a deal, really. I was just curious if I was missing a configuration option or anything. | 18:12 |
| *** ndipanov is now known as ndipanov_gone | 18:18 | |
| openstackgerrit | A change was merged to openstack/ironic-python-agent: Uninstall more buildtime only deps and clean lists https://review.openstack.org/92178 | 18:32 |
| jroll | also, exceptions in vendor_passthru methods get swallowed... this isn't exactly ideal :/ | 18:38 |
| jroll | I see how pxe works around it but it's not fun :( | 18:39 |
| boris-42 | matty_dubs it's actually very important patch | 18:42 |
| boris-42 | matty_dubs cause it allows you guys to keep in ironic source tree any benchmarks | 18:43 |
| boris-42 | matty_dubs even if they are not or not be acceped in rally | 18:43 |
| matty_dubs | Oh, interesting. That could be useful. | 18:44 |
| boris-42 | matty_dubs without it it's a bit painfull=0 | 18:46 |
| boris-42 | matty_dubs cause you'll need to wait for every patch reviews in rally | 18:47 |
| openstackgerrit | Josh Gachnang proposed a change to openstack/ironic: Adding swift temp url support https://review.openstack.org/81391 | 18:48 |
| *** jgrimm has quit IRC | 18:50 | |
| matty_dubs | BTW, do we an official theme song? If not, might I propose Alanis Morissette? https://www.youtube.com/watch?v=Jne9t8sHpUc | 18:51 |
| *** gmatefi has joined #openstack-ironic | 18:54 | |
| NobodyCam | this one is better http://jezebel.com/that-alanis-morissette-ironic-song-is-finally-actually-807802604 | 18:56 |
| *** datajerk has quit IRC | 18:57 | |
| NobodyCam | :-p | 18:58 |
| *** datajerk has joined #openstack-ironic | 18:59 | |
| Shrews | NobodyCam: lol | 19:01 |
| NobodyCam | hehehe | 19:01 |
| matty_dubs | NobodyCam: LOL, this is fantastic | 19:01 |
| NobodyCam | its actually Ironic | 19:01 |
| *** Nisha has quit IRC | 19:02 | |
| NobodyCam | I liked the guy dieing from the paper cut | 19:02 |
| NobodyCam | :-p | 19:02 |
| *** zdiN0bot has quit IRC | 19:03 | |
| matty_dubs | Heh, from the comments, someone suggests that "[T]he line "isn't it ironic" is not a statement, but a question. So she lists all of these instances and then asks "Isn't is ironic?" and everyone can answer "No, it is not" and move on." | 19:05 |
| openstackgerrit | linggao proposed a change to openstack/ironic: Add base function for testing xml response https://review.openstack.org/91681 | 19:09 |
| *** linggao__ has quit IRC | 19:09 | |
| *** linggao has joined #openstack-ironic | 19:10 | |
| NobodyCam | lol | 19:10 |
| boris-42 | matty_dubs it's merged | 19:11 |
| matty_dubs | boris-42: Wow, that was fast! | 19:12 |
| boris-42 | matty_dubs you can put now benchmarks scenarios in ~/.rally/plugins/scenarios | 19:13 |
| *** gmatefi has quit IRC | 19:13 | |
| boris-42 | matty_dubs or /etc/rally/plugins/scenarios | 19:13 |
| Shrews | adam_g: Deva asked me to add a tempest test for the new server rebuild functionality. Any thoughts on whether I should add that to the existing test_baremetal_server_ops() scenario test, or should I add a new one? | 19:14 |
| adam_g | Shrews, i think that shoudl go into a new teste_baremetal_advanced_ops test | 19:14 |
| adam_g | Shrews, similar to the basic_server_ops / advanced_server_ops. | 19:14 |
| Shrews | adam_g: ack. thx | 19:15 |
| matty_dubs | boris-42: I think I'm close to having this working; then I'll rebase and pull that in, then send what I've got so far up for review | 19:16 |
| boris-42 | matty_dubs nice =) | 19:16 |
| *** zdiN0bot has joined #openstack-ironic | 19:16 | |
| boris-42 | matty_dubs btw we should add ironic dsvm gates in rally and probably in ironic | 19:17 |
| boris-42 | matty_dubs but it depends on devananda | 19:17 |
| * matty_dubs has to confess to not following that work too closely | 19:18 | |
| *** Tushar_ has left #openstack-ironic | 19:22 | |
| *** coolsvap is now known as coolsvap|afk | 19:24 | |
| * NobodyCam wants this: http://www.ubersignal.com/wilson-460108-mobile-4g-signal-booster-kit-voice-3g-4g-lte-for-all-carriers.html | 19:24 | |
| matty_dubs | boris-42: Did someone already merge Ironic support upstream? I just got a conflict in osclients.py | 19:25 |
| matty_dubs | NobodyCam: You might be in the perfect setup for that. For mine I'd have to run cabling through several floors | 19:25 |
| NobodyCam | :) | 19:25 |
| Shrews | NobodyCam: "As the outside signal gets weaker, you may have to keep your phone within about 18 inches from the inside antenna in order to take advantage of the boosted signal. " 18 inches? Really? | 19:26 |
| matty_dubs | And cheap cable will lose a _lot_ of signal through 30' at 1900 MHz | 19:26 |
| matty_dubs | Oh, the sample video shows it in a car, where 18" is maybe moderately doable | 19:26 |
| NobodyCam | yep. it would basicly be for my hotspot. so 18" is not a prob at all | 19:26 |
| boris-42 | matty_dubs hmm | 19:26 |
| boris-42 | matty_dubs yep romcheg1 add first patch | 19:27 |
| boris-42 | matty_dubs https://github.com/stackforge/rally/commit/8d3d6a117a6feda446559f55e3b0f759db6325b6 | 19:27 |
| matty_dubs | NobodyCam: Could you just get a hotspot with an external antenna, and skip this thing? | 19:27 |
| matty_dubs | boris-42: Ah, okay. That'll cut down on how much I need to send, I guess. | 19:28 |
| NobodyCam | I was in AL a a couple of days ago and nothing I did get me single. I was thinking mount the ant. on the tv ant bracket. so I reaise it up and down | 19:29 |
| NobodyCam | as needed | 19:30 |
| *** datajerk has quit IRC | 19:30 | |
| NobodyCam | s/get/got/ | 19:30 |
| NobodyCam | s/single/signal/ | 19:30 |
| NobodyCam | :-p | 19:30 |
| *** datajerk has joined #openstack-ironic | 19:31 | |
| *** datajerk has quit IRC | 19:32 | |
| *** jbjohnso has joined #openstack-ironic | 19:35 | |
| *** jgrimm has joined #openstack-ironic | 19:37 | |
| *** romcheg has joined #openstack-ironic | 19:37 | |
| *** ifarkas has quit IRC | 19:39 | |
| NobodyCam | steps afk for a few minutes | 19:40 |
| jroll | matty_dubs: JayF has been linking internal people to alanis morissette for months :P | 19:59 |
| matty_dubs | lol | 19:59 |
| matty_dubs | I'm glad I'm not the only one ;) | 20:00 |
| jroll | :) | 20:00 |
| *** ams0 has joined #openstack-ironic | 20:01 | |
| *** zdiN0bot has quit IRC | 20:02 | |
| *** ams0 has quit IRC | 20:11 | |
| *** ams0 has joined #openstack-ironic | 20:28 | |
| openstackgerrit | Josh Gachnang proposed a change to openstack/ironic: Factoring out PXE and TFTP functions https://review.openstack.org/90233 | 20:34 |
| *** zdiN0bot has joined #openstack-ironic | 20:35 | |
| *** Mikhail_D_ltp has quit IRC | 20:35 | |
| matty_dubs | boris-42: Heh, after I pulled out the code I'd added that overlapped with romcheg's, I'm left with just this: http://fpaste.org/99352/32223813/ | 20:38 |
| matty_dubs | Though I'm looking now at expanding that | 20:38 |
| boris-42 | matty_dubs I think it's better to make something like | 20:39 |
| boris-42 | matty_dubs node.create() node.list() | 20:39 |
| boris-42 | matty_dubs in single benchmark | 20:39 |
| boris-42 | matty_dubs that allows you to track how this operations scales depending on amount of nodes | 20:39 |
| romcheg | matty_dubs boris-42: Do you guys ever sleep? ^) | 20:39 |
| boris-42 | romcheg matty_dubs nope | 20:39 |
| matty_dubs | lol, it's only 4:30pm here... I can't speak for boris-42 though ;) | 20:40 |
| boris-42 | matty_dubs here it's 00:40 | 20:40 |
| boris-42 | a.m. | 20:40 |
| matty_dubs | boris-42: Ah, I was looking at some existing benchmarks to see how they handle that. I'll give that a try. | 20:40 |
| matty_dubs | Yikes! | 20:40 |
| boris-42 | matty_dubs https://github.com/stackforge/rally/blob/master/rally/benchmark/scenarios/neutron/network.py | 20:41 |
| boris-42 | matty_dubs here is the sample ^ | 20:41 |
| *** datajerk has joined #openstack-ironic | 20:41 | |
| matty_dubs | Perfect | 20:42 |
| matty_dubs | Ah, and that pulls in a utils.py for Neutron. I was just looking at that pattern. | 20:42 |
| romcheg | matty_dubs: I was going to go, so I will ask boris-42 about this discussion. | 20:44 |
| romcheg | *tomorrow | 20:44 |
| matty_dubs | Sure. Get some sleep! ;) | 20:46 |
| NobodyCam | no sleep for romcheg | 20:46 |
| NobodyCam | :) | 20:46 |
| romcheg | NobodyCam: :-P | 20:46 |
| matty_dubs | No sleep 'til Atlanta! | 20:46 |
| NobodyCam | romcheg: everyting ok there in the home land | 20:47 |
| NobodyCam | matty_dubs: I'm in Atlanta | 20:47 |
| romcheg | NobodyCam: No, it's not unfortunately :( | 20:47 |
| NobodyCam | :( | 20:47 |
| jroll | NobodyCam: then time to sleep! | 20:48 |
| NobodyCam | jroll: its almost 5 here | 20:49 |
| jroll | :) | 20:49 |
| boris-42 | matty_dubs yep we have that pattern | 20:51 |
| boris-42 | matty_dubs in utils we are putting atomic actiosn | 20:51 |
| boris-42 | matty_dubs that are used to build benchmarks | 20:52 |
| *** ams0 has quit IRC | 20:54 | |
| boris-42 | matty_dubs btw http://junodesignsummit.sched.org/event/82bbddb45555dc39fa6029ee56e7dc4e#.U2f6oa2Syig | 20:55 |
| matty_dubs | Ah, I missed that | 20:56 |
| * matty_dubs adds to his agenda | 20:56 | |
| matty_dubs | Oh, though that's concurrent with "Hardware Multitenancy Risk Mitigation" :-\ | 20:57 |
| boris-42 | matty_dubs yep =) | 21:02 |
| boris-42 | matty_dubs risk is everywhere=) | 21:02 |
| matty_dubs | Haha | 21:04 |
| *** zdiN0bot has quit IRC | 21:07 | |
| *** linggao has quit IRC | 21:07 | |
| *** ams0 has joined #openstack-ironic | 21:13 | |
| *** boris-42 has quit IRC | 21:15 | |
| *** boris-42 has joined #openstack-ironic | 21:16 | |
| *** mrda-away is now known as mrda | 21:18 | |
| *** ams0 has quit IRC | 21:24 | |
| *** ams0 has joined #openstack-ironic | 21:24 | |
| *** matty_dubs is now known as matty_dubs|gone | 21:25 | |
| *** ams0_ has joined #openstack-ironic | 21:30 | |
| *** ams0 has quit IRC | 21:30 | |
| *** ams0_ has quit IRC | 21:35 | |
| *** datajerk has quit IRC | 21:48 | |
| *** ams0 has joined #openstack-ironic | 21:55 | |
| *** ams0 has quit IRC | 21:58 | |
| *** ams0 has joined #openstack-ironic | 21:58 | |
| *** ams0 has quit IRC | 22:03 | |
| *** krtaylor has quit IRC | 22:10 | |
| *** ams0 has joined #openstack-ironic | 22:26 | |
| openstackgerrit | Jarrod Johnson proposed a change to stackforge/pyghmi: Implement function to check SOL payload as keepalive https://review.openstack.org/92220 | 22:30 |
| openstackgerrit | Jarrod Johnson proposed a change to stackforge/pyghmi: Implement function to check SOL payload as keepalive https://review.openstack.org/92220 | 22:32 |
| openstackgerrit | A change was merged to stackforge/pyghmi: Implement function to check SOL payload as keepalive https://review.openstack.org/92220 | 22:36 |
| *** zdiN0bot has joined #openstack-ironic | 22:45 | |
| *** zdiN0bot has left #openstack-ironic | 22:54 | |
| *** radsy has joined #openstack-ironic | 23:00 | |
| *** radsy has joined #openstack-ironic | 23:00 | |
| *** romcheg has quit IRC | 23:05 | |
| *** mdenny has quit IRC | 23:13 | |
| *** ams0_ has joined #openstack-ironic | 23:13 | |
| *** ams0_ has quit IRC | 23:15 | |
| *** ams0_ has joined #openstack-ironic | 23:15 | |
| *** ams0 has quit IRC | 23:17 | |
| *** ams0_ has quit IRC | 23:20 | |
| *** hemna is now known as hemna_ | 23:36 | |
| *** newell_ has quit IRC | 23:43 | |
| *** neillc has joined #openstack-ironic | 23:51 | |
| openstackgerrit | Josh Gachnang proposed a change to openstack/ironic: Factoring out PXE and TFTP functions https://review.openstack.org/90233 | 23:51 |
| openstackgerrit | Josh Gachnang proposed a change to openstack/ironic: Adding a reference driver for the agent https://review.openstack.org/84795 | 23:51 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!