Category:CSC: Difference between revisions
No edit summary |
Add YouTube video summaries and fix broken video embeds (HTML5video -> EmbedVideo) |
||
| (One intermediate revision by one other user not shown) | |||
| Line 2: | Line 2: | ||
[[category:CSC]] | [[category:CSC]] | ||
==Training on CSC | ==Training on CSC PROJECT PART 1. Dated 24 OCT 2024 - Tetra Support Staff - Manish Singh== | ||
{{#ev:youtube|QWcps6hsWOw|640}} | |||
'''Video summary:''' Overview of the CSC client's Apache CloudStack private-cloud environment, demoed first on an Ubuntu test setup then walked through against the real production environment (two locations, Noida and Pune). Covers the CloudStack dashboard/architecture: the management node (controls everything via API/UI, backed by a MariaDB database), Compute hosts (physical hypervisors, each running multiple KVM guest VMs that can be live-migrated between hosts, rebooted, or put into maintenance mode from the UI), and primary vs. secondary storage (primary for live VM/management data, secondary -- originally NFS, since migrated to Ceph for resilience -- for ISOs and templates used to build new VMs). The main project work: upgrading the CloudStack version itself, upgrading each management node's underlying OS off end-of-life CentOS onto Rocky Linux, and setting up a Galera cluster so the two management nodes replicate their database and can fail over to each other -- validated first in the Ubuntu test environment, already live in Pune, still pending client downtime approval in Noida. Also covers the general host-replacement pattern used across the client's compute fleet: migrate all VMs off a host, reformat it onto Rocky Linux, and re-add it to the cluster. | |||
Latest revision as of 02:53, 25 July 2026
Training on CSC PROJECT PART 1. Dated 24 OCT 2024 - Tetra Support Staff - Manish Singh[edit]
Video summary: Overview of the CSC client's Apache CloudStack private-cloud environment, demoed first on an Ubuntu test setup then walked through against the real production environment (two locations, Noida and Pune). Covers the CloudStack dashboard/architecture: the management node (controls everything via API/UI, backed by a MariaDB database), Compute hosts (physical hypervisors, each running multiple KVM guest VMs that can be live-migrated between hosts, rebooted, or put into maintenance mode from the UI), and primary vs. secondary storage (primary for live VM/management data, secondary -- originally NFS, since migrated to Ceph for resilience -- for ISOs and templates used to build new VMs). The main project work: upgrading the CloudStack version itself, upgrading each management node's underlying OS off end-of-life CentOS onto Rocky Linux, and setting up a Galera cluster so the two management nodes replicate their database and can fail over to each other -- validated first in the Ubuntu test environment, already live in Pune, still pending client downtime approval in Noida. Also covers the general host-replacement pattern used across the client's compute fleet: migrate all VMs off a host, reformat it onto Rocky Linux, and re-add it to the cluster.
Pages in category "CSC"
This category contains only the following page.