Skip to content

Fix v0.9#3

Open
Tutj wants to merge 6 commits into
guimou:mainfrom
Tutj:fix_v0.9
Open

Fix v0.9#3
Tutj wants to merge 6 commits into
guimou:mainfrom
Tutj:fix_v0.9

Conversation

@Tutj
Copy link
Copy Markdown

@Tutj Tutj commented Dec 24, 2020

Some fixes are added. Also a new folder of ODH containing Kfdef of ODH v0.9

Tutj added 6 commits December 24, 2020 18:03
After deploying, Ceph RGW should be connected by using 'rook-ceph-rgw-s3a'
Service name of MySQL should be corrected with 'achdbservice' based on information inserted when deploying MariaDB
Kafka cluster is deployed on 'ach' namespace based on steps of deploying kafka.
Therfore, url of bootstrapServer of kafka should be corresponded with the right namespace
This file contains necessary information for deploying ODH 0.9
@guimou
Copy link
Copy Markdown
Owner

guimou commented Dec 27, 2020

@Tutj Thanks for all the corrections. Files were from a time were I deployed the Database and Kafka in another folder, which is not necessary.
But before I merge, quick remark/question: do you think it's really useful to deploy all the ODH components like Seldon, AI library,...? It just consumes more resources and may "disorient" the user who does not know what to do with all those pods, deployment configs, routes,...

@Tutj
Copy link
Copy Markdown
Author

Tutj commented Dec 28, 2020

Agree, we can remove unnecessary components of ODH that can consume more resources.

@Tutj
Copy link
Copy Markdown
Author

Tutj commented Dec 28, 2020

I also want to remove "rook-ceph-rgw-ocs-storagecluster-cephobjectstore" service when deploying rgw for OCS and modify route to point to default service of OCS. I will update the PR later

@guimou
Copy link
Copy Markdown
Owner

guimou commented Dec 28, 2020 via email

@Tutj
Copy link
Copy Markdown
Author

Tutj commented Dec 29, 2020

As by default, Noobar will be the object storage and we will have pods running under the control of noobar. However, i dont see that pod running since we deploy the demo on AWS? (actually i dont know why OCS running on AWS has no rgw :s). That is why we need to deploy rgw by hand then in this case "rook-ceph-rgw-ocs-storagecluster-cephobjectstore" service has no meaning, neither noobar?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants