Introduction To ClearCase

Document Sample
Introduction To ClearCase Powered By Docstoc
					Rational ClearCase Introduction
   for PeopleSoft DSI UNIX
        Introduction to ClearCase
• What is ClearCase?
   – Rational ClearCase is an advanced multi-platform client-server
     source code management system (i.e. source control system)

• What is Source Control?
   – Source control is effectively change management for the revision
     of source code files in a software product.

• Why do we need Source Control?
   –   To track what a release is comprised of
   –   To enable the company to support released products
   –   To decrease development time and costs
   –   Protect the company’s assets
   –   To enhance the quality of the company’s products
       Introduction to ClearCase
• What does ClearCase do for the Company?
   – Allows developers to see what changes have occurred from version
     to version
   – Allows developers to know which versions comprise a specific
   – Allows the re-creation and support of any given release at any
     given time
   – It is the only way to make your changes visible to the rest of the
     development community
   – Enables parallel development on the same code module by
     multiple developers at the same time without interfering with each
       Introduction to ClearCase
• Which version of the ClearCase configuration management
  software do we use at PeopleSoft?
   – PeopleSoft currently utilizes Rational’s ClearCase 4.2

• What platforms are supported by ClearCase and used at
   – AIX, HP-UX, OSF1, Linux (Red Hat), Solaris, Windows NT 4.0
     & Windows 2000.
   – Other platforms are supported by ClearCase but not currently in
     use at PeopleSoft.
         Introduction to ClearCase
• Who uses ClearCase?
    – All application developers and PeopleTools developers within the
      Products and Technology organization world wide.
    – Any third party organizations contracted by the company to develop
      software for resale by PeopleSoft.
    – All QA automation teams that utilize ‘PSScript’ based automation to
      perform their product tests.

• What sort of support is required for ClearCase?
    – Because ClearCase is utilized by all of development world wide, it must
      be supported in a 24 / 7 on call basis.

• Why does ClearCase require this status?
    – Because the ability to modify and maintain code is critical to the ability of
      the company to develop and support it’s products.
    – Downtime can mean lost revenue and increased development costs. It
      could also prolong the downtime of a customer dependent upon a fix for a
      critical incident.
        Introduction to ClearCase
• Who is responsible for the support of ClearCase?
    – There are two teams that support ClearCase within the company.
        • One team supports only the PeopleTools (PT) organization. That team is
          PeopleTools Build team managed by David Wilson.
        • The other team supports the rest of Application Development. That team is
          the Environments Management (EM) Code and Configuration team managed
          by Kalyan Kalle.
    – Both teams are responsible for the direct support of their respective end
      users. Both teams will utilize the vendor where needed to resolve issues.
    – Both teams may call on DSI for assistance with system and/or
      environment related issues pertaining to ClearCase.
    – End users should open a ‘TPRD’ for assistance on PT related ClearCase
      issues or an ‘Environments Request’ (ER) for EM related ClearCase
      issues. End users should not open DSS requests.

• Why are there two ClearCase groups and two ClearCase installs?
    – PeopleTools desires separation from Application Development. In order to
      meet this requirement a second ClearCase environment was configured.
      The second environment requires a second pre-configured ClearCase
      install pre-set for the PT ClearCase environment.
        Introduction to ClearCase
• Which team should I go to for assistance with a ClearCase issue?
    – Should you have or be assigned an issue regarding ClearCase that you require
      assistance on please determine which ClearCase environment the system is using
      (PT or EM). The system naming standard should be of assistance here as PT- and
      ST- systems should utilize the PT ClearCase environment and all others should
      utilize the EM ClearCase environment. If you do not already have a contact within
      the appropriate team for this issue please add the group’s manager as an ‘Interested
      Party’ to the DSI ticket and request assistance from them. The manager will then
      assign a member of their team to assist you on the issue.
    – You can also get more data about ClearCase from the online help system through
      the use of ‘cleartool man’ and through the DSI UNIX admin web site’s ClearCase

• What responsibilities does the DSI UNIX team have with respect to
    – DSI UNIX system administrators are responsible for the maintenance and
      management of the production ClearCase servers (including on call support) and
      the UNIX based ClearCase clients. They must assist during the process of
      ClearCase patches and upgrades. They are responsible for the installation of
      ClearCase on DSI managed UNIX systems. They must maintain the UNIX NIS
      environment that ClearCase relies upon. They will also be required to assist the
      ClearCase teams on system related issues.

    – On call support will be handled through the current DSI UNIX on call pager
      (also used for ICE support).
         Introduction to ClearCase
• Which UNIX and NetApp systems are the ClearCase servers?
    – ClearCase at PeopleSoft is split in two environments. However, both
      environments share a single Network Appliance for database storage. Each
      environment also has it’s own production and backup server.

    – ST-NFS02 is the production ClearCase Network Appliance. Each environment
      (PT & EM) have their own volume on the unit. PT utilizes ‘vol0’ and EM utilizes
      ‘vol1’. Nothing other than ClearCase utilizes (or should utilize) this filer.

         • Please note: Any and all tickets involving ST-NFS02 need to have both ClearCase
           teams involved (via Interested Parties).

    – PT utilizes ST-SUN20 as it’s production ClearCase server and ST-SUN25 as it’s
      backup ClearCase server. Any and all tickets involving these two servers need to
      have the PT group as an ‘Interested Party’.

    – EM utilizes CC-SUN01 as it’s production ClearCase server and SCB-SUN02 as
      it’s backup ClearCase server. Any and all tickets involving these two servers need
      to have the EM group as an ‘Interested Party’.

    When adding groups as interested parties please include ‘EM-OPS-CC’ for EM and
      ‘PeopleTools Product Build’ for PT
        Introduction to ClearCase
• What does ClearCase depend upon?
      Because ClearCase is a network intensive heterogeneous client-server application it
      requires specific configuration and resources available to it in order for it to run
        • In order for users to appropriately access UNIX based VOBs from NT based
           client systems the users must have matching accounts and groups in both the
           UNIX server(s) and the NT domain. NIS is required because resources like the
           network appliance (ST-NFS02) are shared between the environments and
           cannot have duplicate UID/GIDs used. NIS is also required because other
           UNIX systems are ClearCase clients and the account and group data must be
           accessible by these UNIX systems on every supported client platform.
        • In order for ClearCase to execute effectively certain NFS mount, qtree, rsh and
           CIFS security options (including share security) must be selected on the
           Network Appliance and ClearCase servers.
       Introduction to ClearCase
• What does ClearCase depend upon?

   – Network Appliance (ST-NFS02) Configuration Requirements

       • RSH must be enabled for ‘root’ on production and backup UNIX ClearCase
       • Scheduled snapshots must be disabled.
       • Quota management must be disabled.
       • CIFS share permissions must not be modified once established and the system
         is live
       • CIFS ‘oplocks’ must be disabled
       • The ‘VOB owner’ account (ccadmin for EM and clrcase for PT) must own the
         directories where ClearCase objects will be stored on the ‘filer’
       • CIFS security style must be ‘UNIX’
       • ClearCase volumes should have their ‘qtree’ set to ‘mixed’ mode
       • ClearCase UNIX servers must be able to mount and write on their data
         volume(s) as ‘root’
       • Filer must be configured with a global hot spare for data protection
       • Filer must be series 8xx with Data OnTAP OS 6.01 or higher OR series 7xx
         with Data OnTAP OS 5.3.6 or higher
       • Filer must be a member of the NT ‘CORP’ domain
       • Filer must utilize NIS as it’s authentication source
    Introduction to ClearCase
What does ClearCase depend upon?

– Production UNIX Server Configuration Requirements

    • Must be able to mount the appropriate filer (ST-NFS02) volume and write as
    • Limited ‘sudo’ for ‘ccadmin’ account on all systems designated as EM
      ClearCase Servers (CC-SUN01, SCB-SUN02). ‘clrcase’ account on PT
      ClearCase servers (ST-SUN20, ST-SUN25) should also have limited ‘sudo’
    • ‘root’ must run the nightly cron job to perform the ClearCase backup
    • All accessing server and client systems must use NFS hard mounts for
      ClearCase volumes on the Network Appliance.
    • All accessing clients or servers must be at ClearCase version 4.2 or higher
    • All accessing clients and servers must be able to mount the ‘filer’ in the same
      location in the local filesystem (this requires the Automounter).
        Introduction to ClearCase
• How is ClearCase backed up?

   – Both ClearCase environments (PT and EM) store all critical application data on the
     Network Appliance filer (ST-NFS02) in their respective volumes.
   – Backups are started by a cron job owned by ‘root’.
   – Each production server will copy the latest application specific configuration files
     to the Network Appliance nightly starting at approximately 11PM.
   – Each production server will then lock and checkpoint the ClearCase VOB
     databases in preparation for a Network Appliance Snapshot backup.
   – Once all ClearCase VOB databases are ready, the production server (as root) will
     rsh to the Network Appliance (ST-NFS02) and execute a snapshot backup for the
     defined volume. Once the snapshot is complete, the production server will unlock
     the ClearCase VOB databases re-enabling development. ClearCase backups are
     generally completed before midnight each evening.
   – Each production server manages the Network Appliance snapshot backups for it’s
     volume. Each volume has seven (7) days worth of snapshot backups stored locally
     on the Network Appliance filer (ST-NFS02).
   – The CORP-IT backup team will mount the snapshot each night and will perform a
     full backup of its contents between 12AM and 6AM. CORP-IT on-site backups are
     retained for one (1) month.
   – Once per week, CORP-IT will take a backup and store it off-site. Off-site backups
     are retained for six (6) months.

      Further details on the backup and recovery procedure are available in the
      ClearCase Disaster Recovery Plan maintained by each ClearCase team.

Shared By: