Learn ceph architecture and try to deploy a ceph cluster.

Looking for hackers with the skills:

storage cluster filesystem

This project is part of:

Hack Week 11

Activity

  • about 11 years ago: a_faerber liked this project.
  • about 11 years ago: LarsMB joined this project.
  • about 11 years ago: LarsMB liked this project.
  • about 11 years ago: qakapil joined this project.
  • about 11 years ago: osynge liked this project.
  • about 11 years ago: wanghaisu liked this project.
  • about 11 years ago: wanglh added keyword "storage" to this project.
  • about 11 years ago: wanglh added keyword "cluster" to this project.
  • about 11 years ago: wanglh added keyword "filesystem" to this project.
  • about 11 years ago: lisaseraph joined this project.
  • about 11 years ago: zhonglidong started this project.
  • about 11 years ago: wanglh originated this project.

  • Comments

    • zhonglidong
      about 11 years ago by zhonglidong | Reply

      Hope we could setup a ceph environment :)

    • qakapil
      about 11 years ago by qakapil | Reply

      I am working in the ceph storage team. Let me know if you need any info...

    • osynge
      about 11 years ago by osynge | Reply

      As far as I know no one has made a crush-map to graphviz, this would be a very cool idea, provide some understanding of ceph, and I'm sure if you ask about (and I can help here) we can get a few crush-maps to visualise.

    • osynge
      about 11 years ago by osynge | Reply

      I just asked if anyone had made a nice crushmap visualisation tool on ceph irc and a guy with email address eric.mourgaya@arkea.com offered to help if you want to do this.

    • osynge
      about 11 years ago by osynge | Reply

      Kapil created this project https://hackweek.suse.com/projects/574 you might want to look :)

    • a_faerber
      about 11 years ago by a_faerber | Reply

      I would be interested in enabling aarch64 builds of ceph.

    Similar Projects

    Hacking a SUSE MLS 7.9 Cluster by roseswe

    Description

    SUSE MLS (Multi-Linux Support) - A subscription where SUSE provides technical support and updates for Red Hat Enterprise Linux (RHEL) and CentOS servers

    The most significant operational difference between SUSE MLS 7 and the standard SUSE Linux Enterprise Server High Availability Extension (SLES HAE) lies in the administrative toolchain. While both distributions rely on the same underlying Pacemaker resource manager and Corosync messaging layer, MLS 7 preserves the native Red Hat Enterprise Linux 7 user space. Consequently, MLS 7 administrators must utilize the Pacemaker Configuration System (pcs), a monolithic and imperative tool. The pcs utility abstracts the entire stack, controlling Corosync networking, cluster bootstrapping, and resource management through single-line commands that automatically generate the necessary configuration files. In contrast, SLES HAE employs the Cluster Resource Management Shell (crmsh). The crm utility operates as a declarative shell that focuses primarily on the Cluster Information Base (CIB). Unlike the command-driven nature of pcs, crmsh allows administrators to enter a configuration context to define the desired state of the cluster using syntax that maps closely to the underlying XML structure. This makes SLES HAE more flexible for complex edits but requires a different syntax knowledge base compared to the rigid, command-execution workflow of MLS 7.

    Scope is here MLS 7.9

    Goals

    • Get more familiar with MLS7.9 HA toolchain and Graphical User Interface and Daemons
    • Create a two node MLS cluster with SBD
    • Check different use cases
    • Create a "SUSE Best Practices" presentation slide set suitable for Consulting Customers

    Resources

    • You need MLS7.9 (Qcow2) installed + subscription
    • KVM server with 2 KVMs, 2 SBD
    • RHEL7 and HA skills


    Work on kqlite (Lightweight remote SQLite with high availability and auto failover). by epenchev

    Description

    Continue the work on kqlite (Lightweight remote SQLite with high availability and auto failover).
    It's a solution for applications that require High Availability but don't need all the features of a complete RDBMS and can fit SQLite in their use case.
    Also kqlite can be considered to be used as a lightweight storage backend for K8s (https://docs.k3s.io/datastore) and the Edge, and allowing to have only 2 Nodes for HA.

    Goals

    Push kqlite to a beta version.
    kqlite as library for Go programs.

    Resources

    https://github.com/kqlite/kqlite