Changes between Version 12 and Version 13 of Meeting29April2010


Ignore:
Timestamp:
Jan 4, 2017, 2:24:10 AM (21 months ago)
Author:
rutsky
Comment:

update "slave" with "worker"

Legend:

Unmodified
Added
Removed
Modified
  • Meeting29April2010

    v12 v13  
    2121 * Build status in DB
    2222   * better / faster access to build history
    23  * Expand Latent-Slave support
     23 * Expand Latent-Worker support
    2424   * KVM/VMWare/Xen/qemu via libvirt?
    2525 * Project / Repository
     
    3838 * Logging
    3939   * Compression of logs on the wire
    40    * Slave-side logging
     40   * Worker-side logging
    4141 * Graceful Shutdown of Master
    42  * Slave Administrative help
    43    * Graceful don't-schedule-stuff-here option (lower slave priority) (with some kind of scheduling)
    44    * communication with slave admin
     42 * Worker Administrative help
     43   * Graceful don't-schedule-stuff-here option (lower worker priority) (with some kind of scheduling)
     44   * communication with worker admin
    4545   * remote shell
    4646
     
    7777Mozilla will be working on this project, so we'll see how it turns out soon enough.
    7878
    79 == Latent-Slave Support ==
     79== Latent-Worker Support ==
    8080
    81 Nobody objected to this idea.  Chris Atlee mentioned wanting the ability to do generate an arbitrary number of slaves without naming them all individually: currently every EC2 slave must have a distinct AMI that knows, at a minimum, its unique slave name.
     81Nobody objected to this idea.  Chris Atlee mentioned wanting the ability to do generate an arbitrary number of workers without naming them all individually: currently every EC2 worker must have a distinct AMI that knows, at a minimum, its unique worker name.
    8282
    8383== Multi-project / Multi-repository support ==
     
    9797== Windows Compatibility ==
    9898
    99 Buildbot needs a Windows coordinator - someone who can judge the sanity of a patch, test it out locally, and so on.  We also need some Windows buildslaves.  I heard a volunteer for this who turns out not to be the person I thought -- who was that?  Hopefully Mozilla will be supplying a dozen or so buildslaves, so we'll be able to run Windows on some of those in a controlled fashion.
     99Buildbot needs a Windows coordinator - someone who can judge the sanity of a patch, test it out locally, and so on.  We also need some Windows workers.  I heard a volunteer for this who turns out not to be the person I thought -- who was that?  Hopefully Mozilla will be supplying a dozen or so workers, so we'll be able to run Windows on some of those in a controlled fashion.
    100100
    101101== Source Step Mode Cleanup ==
     
    103103I think that everyone wants to see this happen, but nobody has yet stepped up to the challenge.
    104104
    105 An interesting tangent came up here: moving the VC smarts from the slave to the master.  There are several possible approaches here: continue to use buildbot on the slave side, with the master telling the slave to run 'git this' and 'svn that'; or replace the PB connection with a simpler, non-Python-specific protocol that could be implemented with minimal slave-side requirements (e.g., http://github.com/djmitche/remsh/).  It would be possible to implement the latter in a "transitional" fashion, so that remsh slaves could act just like regular buildbot slaves, without running buildbot/twisted/python on the slave side.
     105An interesting tangent came up here: moving the VC smarts from the worker to the master.  There are several possible approaches here: continue to use buildbot on the worker side, with the master telling the worker to run 'git this' and 'svn that'; or replace the PB connection with a simpler, non-Python-specific protocol that could be implemented with minimal worker-side requirements (e.g., http://github.com/djmitche/remsh/).  It would be possible to implement the latter in a "transitional" fashion, so that remsh workers could act just like regular buildbot workers, without running buildbot/twisted/python on the worker side.
    106106
    107 We briefly touched on the idea of "splitting" buildbot into a master (requiring Jinja, Sqlite, etc.) and a slave (with minimal requirements).
     107We briefly touched on the idea of "splitting" buildbot into a master (requiring Jinja, Sqlite, etc.) and a worker (with minimal requirements).
    108108
    109109== Expanded Notification Framework ==
     
    115115Chris has already added a good deal of optimization to the log-handling code, and while some were surprised and excited to find out about this, I didn't hear any suggestions of new features to be added here.
    116116
    117 == Community Supplied Slaves ==
     117== Community Supplied Workers ==
    118118
    119 A few ideas were suggested to make it easier for a buildmaster admin to manage a set of community-supplied slaves (see above).  Nobody spoke up to work on these ideas, though.  As I understand it, most of the organizations using Buildbot supply their own slaves.
     119A few ideas were suggested to make it easier for a buildmaster admin to manage a set of community-supplied workers (see above).  Nobody spoke up to work on these ideas, though.  As I understand it, most of the organizations using Buildbot supply their own workers.
    120120
    121121== Tests ==
     
    131131 * build coordination {34} - zmanda
    132132 * multi-project {35} - zmanda
    133  * windows {17} - ?? + mozilla slaves
     133 * windows {17} - ?? + mozilla workers
    134134 * tests - dustin
    135135