Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2025-10-03
15:04:35 sean-k-mooney much better thne before you started im juust wonderign if it will explode at say 24_24 or can we push to 32+
15:06:27 gibi sean-k-mooney: yeah 21 is the magic number from the customer. I can add more cases.
15:06:39 gibi please leave a comment in the review so I won't forget
15:06:54 sean-k-mooney a standard 2U server can typicaly hold 24 2.5" ssd's in the front which woould also need 96 pci lanes to support at full speed. given many chips top out at 96-128 lanes with some going ot the low 160s range 32 device is proably close to the max we woudl expct to see
15:07:35 sean-k-mooney sure i can im ok with the current test uing the reproted worse case i just was wondering if you had playied aroudn with it
15:07:44 sean-k-mooney ill pull it locally and try it
15:08:22 sean-k-mooney but ya 120+ second ot 6.9 is a very nice impovement
15:08:33 gibi sean-k-mooney: if it is two slow for 32/32 then try to decrease the max alocation candidates from 1000 to smaller
15:08:39 gibi s/two/too/
15:08:44 sean-k-mooney but 120+ to less then one in the non worse case is very promising
15:09:07 gibi I have drop for the weekend in minutes so I won't be able to test it myself today
15:09:18 sean-k-mooney gibi: oh ya that a good poitn if we can provide soem guiance on tuning the values for larger systems than that at least a workaroudn
15:09:26 gibi btw up until 8/8 I tested the logic in devstack as well and the numbers are pretty similar
15:09:53 sean-k-mooney cool ya i was ment to finish an hour ago so ill loop back to this on monday
15:09:59 sean-k-mooney gibi: enjoy your weekend
15:10:00 gibi cool thanks
15:10:03 gibi same to you
15:35:04 opendevreview Nisha Brahmankar proposed openstack/nova master: Add graceful shutdown feature for nova-compute https://review.opendev.org/c/openstack/nova/+/962975
15:42:23 tkajinam I wonder if anyone disagrees with closing all xen related bugs by setting their status to Won't fix ?
15:44:37 sean-k-mooney well we deleted the driver and the libvirt xen suspprot a few years ago at this point
15:45:24 sean-k-mooney so i dont see why we woudl not set it to wont fix at this point or invlaid
16:02:01 nisha04 This is Nisha from LY Corporation, Japan. We got a green signal in the nova PTG 2025.2 to implement the graceful shutdown feature for nova-compute. (https://etherpad.opendev.org/p/nova-2025.2-ptg#L1088)
16:02:01 nisha04 We have the basic PoC code ready here: https://review.opendev.org/c/openstack/nova/+/962975
16:02:01 nisha04 Hi nova core members,
16:02:03 nisha04 I want to discuss the design choices and spec requirements. I will be attending the OpenInfra Summit in Paris this month. It would be great if we could discuss them in person during the summit. Are you attending the summit? If so, could you share preferable slots, if any, for a discussion.
16:14:10 sean-k-mooney nisha04: some of the core will be tere i belive Uggla stephenfin and bauzas are trvelign/have talks
16:14:43 sean-k-mooney in general the full core team will not but there shoudl be an opteruntiy to have some level of dicusssion the PTG is the week after as well so there is also that operunity
16:20:07 nisha04 sean-k-mooney: Thanks for the info. I just wanted to have a quick discussion and receive informal feedback. I will try to contact Uggla stephenfin and bauzas if they have any available time slot.
16:20:36 Uggla But I’d also like to have this as a topic at the PTG, since I know @gibi will be available and deeply involved.
16:20:36 Uggla I can try to accommodate the timeslot as much as possible with your timezone.
16:20:36 Uggla @nisha04 I think we could find a slot at the OpenInfra Summit to do some kind of pre-check.
16:21:13 sean-k-mooney nisha04: bauzas is doing a contibuting to nova talk
16:21:42 sean-k-mooney i.e. how to contibute so you might be able to find them in the halway after
16:21:59 sean-k-mooney but ya ptg dicussion after woudl be good to allow all to partipate
16:22:33 Uggla I'll be there too.
16:24:18 nisha04 I will also try to find you in the hallway during the summit and hope to have a quick chat sometime.
16:24:18 nisha04 Thanks a lot! I definitely want to bring this topic in the PTG for collective feedback since we are actively working on it now.
16:30:07 Uggla nisha04, pretty sure we will find a slot to discuss.
17:28:22 opendevreview Merged openstack/placement master: Remove excessive logging from GET a_c https://review.opendev.org/c/openstack/placement/+/962775
18:33:14 dansmith melwitt: finally got my devstack unfscked.. but now I've deployed without barbican
18:33:36 dansmith I have several things commented out in my local.conf... is enable_service barbican enough or do I need more?
18:49:01 opendevreview sean mooney proposed openstack/placement master: [WIP] Fix copy_arr_if_needed() and add observability https://review.opendev.org/c/openstack/placement/+/963006
18:53:27 opendevreview sean mooney proposed openstack/placement master: [WIP] Fix copy_arr_if_needed() and add observability https://review.opendev.org/c/openstack/placement/+/963006
#openstack-nova - 2025-10-04
11:02:31 opendevreview Balazs Gibizer proposed openstack/nova master: DNM:Test with optimize_for_wide_provider_trees https://review.opendev.org/c/openstack/nova/+/963036
11:02:31 opendevreview Balazs Gibizer proposed openstack/nova master: DNM: test with placement fix https://review.opendev.org/c/openstack/nova/+/963035
14:46:34 opendevreview Balazs Gibizer proposed openstack/placement master: Inline _consolidate_allocation_requests https://review.opendev.org/c/openstack/placement/+/963052
14:48:15 opendevreview Balazs Gibizer proposed openstack/nova master: DNM:Test with optimize_for_wide_provider_trees https://review.opendev.org/c/openstack/nova/+/963036
14:48:15 opendevreview Balazs Gibizer proposed openstack/nova master: DNM: test with placement fix https://review.opendev.org/c/openstack/nova/+/963035
15:53:57 opendevreview Balazs Gibizer proposed openstack/placement master: Inline exceeds_capacity https://review.opendev.org/c/openstack/placement/+/963053
18:45:34 opendevreview Balazs Gibizer proposed openstack/placement master: Inline exceeds_capacity https://review.opendev.org/c/openstack/placement/+/963053
#openstack-nova - 2025-10-05
21:20:21 opendevreview Steve Baker proposed openstack/nova master: Add VNC console support for the Ironic driver https://review.opendev.org/c/openstack/nova/+/942528
#openstack-nova - 2025-10-06
07:46:31 opendevreview Julien LE JEUNE proposed openstack/nova master: Update start_service() function in test https://review.opendev.org/c/openstack/nova/+/963156
07:49:22 opendevreview Julien LE JEUNE proposed openstack/nova master: Update start_service() function in test https://review.opendev.org/c/openstack/nova/+/963156
07:56:38 nicolairuckel can set them for the NVRAM file as well. I asked this in the libvirt channel and was told that this might be a better place for this question.
07:56:38 nicolairuckel I'm working on this patch to persist the NVRAM: https://review.opendev.org/c/openstack/nova/+/959682 So far it works for all migrations but I habve to manually add read permissions to `instance-0000003_VARS.fd`. The file for `disk` already has those. In the libvirt IRC they told me that not having global read permissions is intentional and I'm wondering why the disk in Nova has those permissions and how I
08:48:46 opendevreview Julien LE JEUNE proposed openstack/nova master: Update start_service() function in test https://review.opendev.org/c/openstack/nova/+/963156
09:37:23 opendevreview Julien Le Jeune proposed openstack/nova master: Update start_service() function in test https://review.opendev.org/c/openstack/nova/+/963156
10:30:00 sean-k-mooney gibi: i dont know if you read this comment https://review.opendev.org/c/openstack/placement/+/962776/comment/f68ed8e5_0803b616/ but when i was running the tests locally i noticed that changing the request_limit paramter had no perforncem imporment
10:30:04 sean-k-mooney """ min(conf.max_allocation_candidates, requst_limit) """
10:30:15 sean-k-mooney i think we are missing ^ somewhere
11:16:26 gibi_ sean-k-mooney: we intentionally not changed the meaning of the limit parameter when max_allocation_candidates was introduced due to
11:16:29 gibi_ Placement generates these candidates fully before applying the limit parameter provided in the allocation candidate query to be able do a random sampling if [placement]randomize_allocation_candidates is True.
11:16:52 gibi_ the limit parameter is applied last
11:17:01 gibi_ if we change that then that changes how randomize_allocation_candidates behave
11:17:10 gibi_ so we did not changed limit
11:35:43 sean-k-mooney hum i see
11:35:48 sean-k-mooney that is unfortunate
11:36:04 sean-k-mooney but i guess that the legacy we are stuck with
11:37:03 sean-k-mooney there are two cases i guess to consdier request_limit < max_allocation_candidates
11:37:31 sean-k-mooney i was hoping in this case we coudl treadt it as if max_allocation_candidates == request_limit
11:37:55 sean-k-mooney in the other case its already going to be enforced
11:39:09 sean-k-mooney i wonder if we coudl implemetn the randomize_allocation_candiates diffently
11:39:24 sean-k-mooney we currently have depthfirst or breathfirst search
11:39:48 sean-k-mooney but couldl we instead do a random search for that usecase.
11:40:49 sean-k-mooney it would not be exactuly the same behavior but the randomisation was really there because of the orginial depth first approch to generating candiates
11:41:07 sean-k-mooney if we are using breathfirst its much less required
11:41:49 sean-k-mooney that a sepreate poblem then the one that your trying to fix but i wonder fi we should conitnue to supprot that if we cant apply optimisation based on teh request limit
11:42:51 sean-k-mooney in otherwords i think we could provide randomness in the generation fo the candiate without needign to generate all of them and doing a random choice at the end
11:43:20 sean-k-mooney that would fulfil the same usecase but with a lot less wasted effort
11:43:40 sean-k-mooney maybe another topic for the ptg dicussion?
11:46:27 sean-k-mooney gibi: basically im not sure that genrating all possible candiate is part of the api contract fo that parmater so we shoudl be free to evolve its meaning over tiem to be more effeictent provided it still has the same overall effect
11:57:21 gibi I'm OK to discuss this on the PTG. But for me this is a small issue that limit and max_allocation_candidates has two different meaning today
11:58:44 sean-k-mooney ya its minor but it means the clinet cant contol the perfoamce of the request by asking for less cnadiates in the responce
11:59:15 sean-k-mooney to me it undermines why the request_limit parmater exists in the api in the first place
11:59:36 gibi that parameter from the start only limited the size of the response nothing lese
11:59:39 gibi else
12:00:51 sean-k-mooney sure but i woudl expect it to do more then that as an end user and it prevent nova for example form being smart and say (i knwo this query might generate a lot of candiate so im going to use a lower limit dynicmicly)
12:01:01 sean-k-mooney not that nova has the intelegents to know when it shoudl do that
12:01:20 sean-k-mooney so its not a blocker but i just find it surprising
12:02:06 sean-k-mooney gibi: i noticed you commented on your inlining patches that they dont seam to provide a benifit
12:02:27 sean-k-mooney gibi: do you think that is a dead end or are you plannign to do more expermients in that direction
12:03:22 gibi unfortunately the randomize_allocation_candidates option defined as
12:03:23 gibi If True, when limiting allocation candidate results, the results will be a random sampling of the full result set.
12:03:34 gibi *full result set* :/
12:03:59 gibi sean-k-mooney: the first inlining helps, the second not much
12:04:05 sean-k-mooney right but it does not define what the full result set is
12:04:36 gibi this helps If True, when limiting allocation candidate results, the results will be a random sampling of the full result set.
12:04:39 gibi sorry
12:04:41 gibi wrong buffer
12:04:47 gibi this helps https://review.opendev.org/c/openstack/placement/+/963052/1/placement/tests/functional/test_allocation_candidates.py#265

Earlier   Later