Permalink
Commits on Mar 13, 2012
  1. @iNecas
  2. @iNecas

    798376 - handle change in value returned by repo.sync method

    In wail_for_tasks it was expected there is always a pulp task given. After this
    change also delayed job task is accepted.
    iNecas committed Mar 13, 2012
Commits on Mar 12, 2012
  1. @weissjeffm

    Revert "802346 - wait until PostgreSQL accepts connections"

    This reverts commit 08a13f4.
    
    This commit causes katello-configure to fail with:
    'for i in {1..6}; do /usr/bin/psql -U postgres -h localhost -c 'select count(*) from pg_tables' >/dev/null 2>&1 || sleep 5; done' is not qualified and no path was specified. Please qualify the command or specify a path.
    weissjeffm committed Mar 12, 2012
  2. @knowncitizen
  3. @pitr-ch
  4. @lzap
  5. @lzap
  6. @lzap
  7. @lzap
  8. @lzap
  9. @lzap
  10. @iNecas

    798772 - fix conversion to local timezone

    This code was used to convert UTC to local timezone
    
        self.sync_date.to_time.in_time_zone(Time.now.zone).to_datetime
    
    Two problems are with it:
    
       1. current timezone is used (which might be different from the time the sync
          plan was created)
    
       2. for some timezones (e.g. EDT) this line fails with
            undefined method `period_for_utc' for nil:NilClass
    
    Replacing with:
    
        self.sync_date.localtime.to_datetime
    iNecas committed Mar 12, 2012
  11. @pitr-ch

    pchalupa's public key

    pitr-ch committed Mar 12, 2012
  12. @lzap
  13. @lzap
  14. @lzap
  15. @lzap
  16. @iNecas

    798376 - fix problem with discovery process

    Setting PulpTaskStatus#error_details to [] instead of 0
    iNecas committed Mar 12, 2012
Commits on Mar 9, 2012
  1. @bbuckingham

    790063 - search - few more mods for consistency

    This commit contains a few more small changes for improved behavior in
    search.
    
    The changes are primarily in errata search to combine id and title as
    a single searchable field.  The reason for doing this is that when the
    user views the list of errata, they see id and title as a single string;
    therefore, this allows them to search based on how they view the data.
    Chose to leave this query using an ngram, since that provides a nice behavior
    for the user.
    
    Also, added the ':' to the 'input_filter' and updated where this method
    is called to ensure that the filter is always applied before we
    add a term to the query (e.g. prepend query with 'name:').
    bbuckingham committed Mar 9, 2012
  2. @mccun934
  3. @mccun934
  4. @mccun934
  5. @mccun934
  6. @ehelms

    798376 - Sync management page reworked to generate error and success

    notices once upon sync completion.
    
    This bug fixes makes a number of changes to how the most recent sync
    status is determined and stored in the database.  Whenever a sync is
    started an object will be created in the databse to track the pulp task
    and be updated as the sync status is updated from pulp.  Further, upon
    completion of a sync, a delayed job will kickoff that will generate a
    success or error notice and if there are errors dump the errors to the
    log.  This change also reduces the number of calls to pulp when
    calculating the overall size of a product in the UI.798376 - Sync
    management page reworked to generate error and success notices once upon
    sync completion.
    ehelms committed Mar 9, 2012
  7. @jlsherrill
  8. @iNecas
  9. @jlsherrill
  10. @jlsherrill
  11. @iNecas

    801107 - allow wildcards in image factory names mappings

    Allows handling both "Red Hat Enterprise Linux" and "Red Hat Enterprise Linux
    Server" with a single configuration line.
    iNecas committed Mar 9, 2012
Commits on Mar 8, 2012
  1. @knowncitizen

    799523 - Fix for new environment full page load.

    Also greatly simplified the JS and HAML for this page so that the overall experience is better and faster.
    knowncitizen committed Mar 8, 2012
  2. @bbuckingham

    786572 - elasticsearch - reduce config to use 3 shards vs default (5)

    Elasticsearch (by default) is configured to use 5 shards and 1 replica.
    That allows the server to grow scaling across up to 10 servers.  While
    that is nice, it does come at the cost of some memory.  We could reduce
    the number of shards to 1, given that our initial deployment configuration
    is only 1 node; however, that would not facilitate future growth/scalability.
    In order to strike a balance between scalability and minimize impact to memory,
    we are reducing the configuration to use 3 shards.
    
    Note: from some basic testing with 5 vs 3 shards, observed the following:
    - 3 shards : VIRT(~1650m), RES(~395m)
    - 5 shards : VIRT(~1720m), RES(~452m)
    
    Scenario run in both configurations before arriving at the above was:
    - define 2 providers (1 w/ rhel6 & 1 w/ rhel6.1)
    - sync both 2 times
    - promote both 2 times
    bbuckingham committed Mar 8, 2012
  3. @mbacovsky

    801549 - After katello-configure thumbslug fails to start with Unable…

    … to load the ssl keystore
    
    - fixed thumbslug.conf to pint to /etc/pki/katello/keystore
    - /etc/pki/katello/keystore is owned by root.katello
    - user thumbslug was added to katello group
    mbacovsky committed Mar 8, 2012
  4. @bbuckingham

    790063 - search - changes for consistency/behavior

    This commit makes minor changes to the usage of elasticsearch
    within Katello.  The primary reason for these changes is to provide
    additional consistency across the search queries as well as
    provide slightly different behavior, in hopes that it'll help
    users.
    
    First thing users should be aware of:
    
    - They query syntax used on all pages (with a couple exceptions) is based
      entirely off of Lucene query syntax.
      - The couple of exceptions are on System Templates where a user may
        (from the right sliding tree) type in text to locate a repo or
        package name.  Since those provide 'auto-complete', they don't
        really align with all of the other search queries in the UI.
    
    The following are a couple of key changes:
    
    1. Provide a 'default_field' for queries that the user provides, but
       does not provide a term  (e.g. "searchingtext" vs "name:searchtext").
       The approach used for selecting the default_field is based on the
       context of where the query is performed.  Generally, this is based
       on the data the user sees (e.g. provider names, user names...etc).
    
       There are several reasons for this approach; however, the key reasons
       were: 1. it should improve search performance (by not searching
       across all indexed fields by default and 2. we do not currently show
       the user all fields that are indexed; therefore, it could cause
       confusion to show matches on fields the user is not aware of.
    
    2. Use the 'keyword' analyzer on the primary field (e.g name).  This
       change should provide an improved behavior when data contains
       delimiters such as - or space.  The keyword analyzer will essentially treat
       an entire name as a single token vs breaking it in to multiple
       tokens.
    bbuckingham committed Mar 8, 2012
  5. @jlsherrill