Bioinformatics.org
[Regeneron]
Not logged in
  • Log in
  • Bioinformatics.org
    Membership (41430+) Group hosting [?] Wiki
    Franklin Award
    Sponsorships

    Careers
    About bioinformatics
    Bioinformatics training
    Bioinformatics jobs

    Research
    All information groups
    Online databases Online analysis tools Online education tools More tools

    Development
    All software groups
    FTP repository
    SVN & CVS repositories [?]
    Mailing lists

    Forums
    News & Commentary
  • Submit
  • Archives
  • Subscribe

  • Jobs Forum
    (Career Center)
  • Submit
  • Archives
  • Subscribe
  • BIRCH: Comprehensive bioinfo. system - Support tickets

    Submit | Open tickets | Closed tickets

    [ Ticket #1185 ] Need more robust way to recover if BIRCH download fails
    Date:
    08/13/13 21:09
    Submitted by:
    B_Fristensky
    Assigned to:
    Alvare
    Category:
    Getbirch
    Priority:
    7
    Ticket group:
    Bug
    Resolution:
    Resolved
    Summary:
    Need more robust way to recover if BIRCH download fails
    Original submission:
    If a BIRCH download fails during an update, you are left with missing BIRCH directories, because getbirch refuses to try to install if a partial copy of BIRCH exists. This happens because the first thing we do before downloading is to run UNINSTALL-birch.py.

    This would probably also be a problem if we were doing a new install.

    We need to rethink how getbirch determines what to do if a previous install attempt has failed.
    Please log in to add comments and receive followups via email.
    Followups
    Comment Date By
    Several fixes:

    1. Getbirch now runs UNINSTALL-birch within the download method, after downloading and before untarring.

    2. UNINSTALL-birch.py now has a -d <dirname> option that is called to ensure that it finds the copy of BIRCH that needs to be deleted.

    3. To make UNINSTALL-birch independent of whether or not the rest of BIRCH is present, code from various Python modules in birchlib have been hard-coded into UNINSTALL-birch.py, making it a standalone script.

    4. If there is not a copy of UNINSTALL-birch.py, getbirch downloads a copy so it can do the uninstall.
    08/25/13 17:03 B_Fristensky
    Decision: Let's run UNINSTALL-birch AFTER the downloads are complete. The only advantage to running before is that it saves disk space, and we can usually assume that most systems today will have enough extra disk space.
    08/17/13 15:23 B_Fristensky
    Install: Don't need to worry about install. Getbirch gives the framework and bin files unique names each time a download is done. These are Java temporary files, meaning that JVM automatically deletes them when done. So a failed download during a new install should not leave behind a partial version of BIRCH.
    08/17/13 13:59 B_Fristensky
    No results for "Dependent on ticket"
    No results for "Dependent on Task"
    No other tickets are dependent on this ticket
    Ticket change history
    Field Old value Date By
    status_id Open 08/25/13 17:03 B_Fristensky
    resolution_id Not Resolved 08/25/13 17:03 B_Fristensky
    close_date 12/31/69 19:00 08/25/13 17:03 B_Fristensky
    status_id Pending 08/13/13 21:10 B_Fristensky
    priority 5 08/13/13 21:10 B_Fristensky
    assigned_to unset 08/13/13 21:10 B_Fristensky
    resolution_id Unset 08/13/13 21:10 B_Fristensky

     

    Copyright © 2019 · Scilico, LLC