Sender | Message | Time |
---|---|---|
13 Oct 2021 | ||
Krishna Durai | Yes, the idea is to submit many inferences at one go and chech the results asynchronously | 03:40:49 |
Dan Sun | @_slack_kubeflow_U02AF95BSUF:matrix.org Do we support the migration path from 0.5? | 03:40:53 |
Dan Sun | We do not support batch inference yet but the idea is that we can run a spark job or something like that to take a batch file and send the requests to the inference services, collect back the response and write to a resulting file. | 03:55:03 |
Jagadeesh J joined the room. | 04:32:29 | |
_slack_kubeflow_U02HL87RBDG joined the room. | 07:11:29 | |
Krishna Durai | Thanks for your reply Dan! | 07:15:17 |
_slack_kubeflow_U02HLB9PNMQ joined the room. | 07:44:41 | |
Suresh Nakkiran joined the room. | 13:16:29 | |
Suresh Nakkiran | Dan Sun the kserve migration job does not work from kfserving v0.5.0 because there are knative and istiod version/config differences between v0.5.0 and v0.6.0.. | 13:16:30 |
Marcin Jimenez joined the room. | 14:57:30 | |
Marcin Jimenez | what is the recommended strategy on upgrades from 0.5 to 0.6 and beyond? | 14:57:30 |
Dan Sun | There will be no community meeting today, we have Kubecon talk for KServe today at 2:30pm PDT Serving Machine Learning Models at Scale Using KServe - Animesh Singh, IBM | 15:20:08 |
theofpa joined the room. | 15:34:53 | |
Benjamin Tan joined the room. | 15:40:36 | |
Benjamin Tan | I must be doing something wrong but I still haven't figured this out:
I have Kubeflow set up on a private GKE (so https everything), but I want to have my Knative models not have to go through Dex. I've tried setting up an additional Istio Ingress but keep hitting into various problems (IngressNotConfigured , not being able to curl with the new domain etc) | 15:42:19 |
Benjamin Tan | Is there an easier way to bypass Auth for Knative? | 15:42:39 |
Benjamin Tan | And if I bypass it, does it still go through https? | 15:45:42 |
Nick joined the room. | 15:51:18 | |
iamlovingit joined the room. | 15:55:14 | |
Benjamin Tan | Otherwise, I would be interested in how other foiks handle this | 15:56:02 |
Dan Sun | In reply to@_slack_kubeflow_UM56LA7N3:matrix.orgBenjamin Tan if you have describe your setup in more details I can help out to trouble shooting, we have setup this before so pretty sure it should work | 16:26:32 |
Nick | In reply to@_slack_kubeflow_UFVUV2UFP:matrix.orgDan Sun actually I think it’s 2.30pm tomorrow https://kccncna2021.sched.com/event/lV2s/serving-machine-learning-models-at-scale-using-kserve-animesh-singh-ibm?iframe=no | 18:11:35 |
Peilun Li joined the room. | 18:44:29 | |
Peilun Li | So I guess we have to: 1. First upgrade our manifest from 0.5 to 0.6 (without changing existing services -- still on 0.5) 2. Then install 0.7 manifest to the cluster 3. Meanwhile for new service deployment we can deploy with 0.7 directly, while the old 0.5 deployments are still controlled by 0.6. 4. Finally use the migration job to migrate existing 0.5 services (controlled by 0.6) to 0.7's api group and controller. Would that be a working path? | 18:44:29 |
Peilun Li | In reply toundefined(edited) ... controller. Would ... => ... controller. 5. Delete the 0.6 manifest afterwards. Would ... | 18:49:12 |
Dan Sun | In reply to@_slack_kubeflow_U0127AUTPMH:matrix.orgoops sorry my bad | 19:12:04 |
Dan Sun | In reply toundefined(edited) ... KServe today at ... => ... KServe tomorrow at ... | 19:12:12 |
Dan Sun | In reply toundefined(edited) ... tomorrow at ... => ... tomorrow 10/14 at ... | 19:12:34 |
14 Oct 2021 | ||
Riley Green | Download image.png | 02:35:49 |
Riley Green | Hi, I have installed Kubeflow 1.4 from manifests and I tried to deploy the sklearn-iris predictor sample and I am getting the following error: Any ideas on a way forward? Many thanks 🙏 | 02:35:49 |