Skip to content
Commits on May 6, 2013
  1. @nwgh

    Handle smtp errors gracefully

    nwgh committed
Commits on Mar 6, 2013
  1. @nwgh
  2. @nwgh
  3. @nwgh

    Move StreamLogger to stoneridge module

    nwgh committed
    It's now used in multiple places, so we need to move it somewhere
    shareable.
  4. @nwgh
Commits on Mar 1, 2013
  1. @nwgh

    Add runner support for page load tests

    nwgh committed
    This lets the runner choose which process (and argument set) to use
    based on the extension of the test. An extension of .js corresponds to
    xpcshell tests, while an extension of .page corresponds to page load
    tests.
Commits on Feb 26, 2013
  1. @nwgh

    Switch subprocess.Popen to stoneridge.Process

    nwgh committed
    This is an override of subprocess.Popen that provides sensible defaults
    for stone ridge programs that want to call subprocesses, namely:
     * stdout is redirected to a pipe by default
     * stderr is redirected to stdout by default
     * universal_newlines (translate '\r' or '\r\n' to '\n') is always on
  2. @nwgh

    Use a timeout for xpcshell

    nwgh committed
    This way we don't get test runs that hang for TOO long. Also, shove the
    stdout from both xpcshell and tcpdump to a file instead of a pipe, to
    avoid possible deadlocks (which may have been the cause of xpcshell
    hanging to begin with!)
Commits on Feb 25, 2013
  1. @nwgh

    Merge branch 'master' into pep8

    nwgh committed
    Conflicts:
    	srcloner.py
    	stoneridge.py
    	tools/srenqueuer.py
  2. @nwgh

    Email on all failures

    nwgh committed
    Joy!
Commits on Feb 21, 2013
  1. @nwgh

    More pep8 fixes found by flake8

    nwgh committed
Commits on Feb 20, 2013
  1. @nwgh

    Fix lint warnings in module

    nwgh committed
Commits on Feb 19, 2013
  1. @nwgh

    Attempt to make queue listeners recover from connection failures

    nwgh committed
    This may or may not work (according to pika's github account), but it's
    worth having around.
  2. @nwgh

    Merge branch 'master' into debugging_fixes

    nwgh committed
    Conflicts:
    	srworker.py
  3. @nwgh

    Forget RPC for now

    nwgh committed
    For some reason, things get hung up returning the response, with no
    debugging info available. This is crap, and the RPC is less important
    for the first push (though I still want to find a way to do it)
Commits on Feb 16, 2013
  1. @nwgh

    Make cleaner a daemon

    nwgh committed
    This means things will stick around for a while after a run is
    completed, which will make it easier to debug things!
  2. @nwgh

    Place xpcshell output in the sr output directory

    nwgh committed
    Do this directly, instead of via the collator after xpcshell has been
    run. This makes things easier to find when debugging problems.
Commits on Feb 15, 2013
  1. @nwgh

    Fix scheduler on production infrastructure

    nwgh committed
    Apparently the long waits cause timeouts in the rabbitmq connections,
    and pika doesn't handle reconnecting for us (laaaaaaaame!). Instead of
    using a long-lived connection that will hit TCP timeouts, use a
    short-lived one that will pretty much constantly have data going across
    it, to avoid the timeouts, and the related socket issues. Win!
Commits on Feb 12, 2013
  1. @nwgh
  2. @nwgh

    Quiet down pika logging

    nwgh committed
    I don't need to see all of pika's debug messages in the stoneridge logs,
    so let's keep it to errors and worse only, please!
  3. @nwgh

    Move a couple imports

    nwgh committed
    Windows never will actually call stoneridge.daemonize, which requires a
    couple unix-only imports. Move those imports into stoneridge.daemonize
    so windows programs can import the stoneridge module.
Commits on Feb 8, 2013
  1. @nwgh

    Fix mis-named variable

    nwgh committed
  2. @nwgh
Commits on Feb 2, 2013
  1. @nwgh

    Make help work in stoneridge programs again

    nwgh committed
    Automagically parsing out the log information had screwed up help output
    from programs when -h or --help was passed in (but not any other way)
    because the ArgumentParser for getting log filename was intercepting
    that. By having it not generate its own help, we fix that issue.
  2. @nwgh
Commits on Feb 1, 2013
  1. @nwgh

    Fix running xpcshell

    nwgh committed
    Had missed a reference to an old global variable that now needs to be
    determined at runtime in run_spcshell (specifically, the xpcshell
    variable). This fixes that, as well as adds some logging of the
    xpcshell command line being run.
  2. @nwgh

    Modernize more except clauses

    nwgh committed
  3. @nwgh

    Add support for running as a unit test

    nwgh committed
    This lets things like the worker know not to call any subprocesses
Commits on Jan 31, 2013
  1. @nwgh

    Add more logging in RPC calls

    nwgh committed
  2. @nwgh
  3. @nwgh

    Make enqueuing a new test run work

    nwgh committed
    * Was missing a couple arguments to the function
    * Also need to handle operating systems/netconfigs properly
  4. @nwgh

    Make config filename available

    nwgh committed
    This is for programs that need to pass it to sub-programs (master,
    cloner, worker).
Commits on Jan 30, 2013
  1. @nwgh
  2. @nwgh
Commits on Jan 29, 2013
  1. @nwgh
Something went wrong with that request. Please try again.