Difference between revisions of "Meetings/Swift"

Revision as of 21:37, 19 January 2016

Meeting Time	Every Wednesday at 21:00 UTC in #openstack-meeting
Meeting Logs	http://eavesdrop.openstack.org/meetings/swift/2016/
Chair	notmyname

Agenda: January 20, 2016 2100 UTC

Swift release [notmyname]
- Update eventlet https://review.openstack.org/#/c/269811/
Hackathon [notmyname]

Patches
- Change schedule priority of daemon/server in config - https://review.openstack.org/#/c/238799/ [peterlisak? onovy?]
  - 2 big questions in the review comments
    - Does setting ionice actually improve anything? Swift's processes are rather interconnected, so it's very difficult to isolate a logical API operation by process. eg nice'ing the container server would impact object PUTs since the container is updated as part of the object write.
    - Should we even be doing this in Swift itself instead of distro init scripts. On the one hand, the ionice priority "code" has to be rewritten for every distro. On the other, ops know ionice and how/when to use it and does that actually belong in Swift? If it does, why not other ops/management tools?
    - Real tests result
- Auditor Watchers patches [Zyric]
  - Object Auditor Watcher patch - https://review.openstack.org/#/c/212824/
  - Account Auditors patch - https://review.openstack.org/#/c/268830/
- I would like to discuss about below patches with following summary - http://paste.openstack.org/show/484089/ [ho]
  - Fix posting accounts behavior when half of account servers downed - https://review.openstack.org/#/c/266190/
  - Fix posting containers behavior when half of container servers downed - https://review.openstack.org/#/c/266193/
  - Fix deleting containers behavior when half of container servers downed - https://review.openstack.org/#/c/266199/
  - Fix 404 in container PUT when the majority of the accounts are missing
Specs
- tempurls with a container-level scope - https://review.openstack.org/#/c/199607/ [cbartz]

When adding an item, please include your IRC nickname with it.

@@ Line 19: / Line 19: @@
 **** Does setting ionice actually improve anything? Swift's processes are rather interconnected, so it's very difficult to isolate a logical API operation by process. eg nice'ing the container server would impact object PUTs since the container is updated as part of the object write.
 **** Should we even be doing this in Swift itself instead of distro init scripts. On the one hand, the ionice priority "code" has to be rewritten for every distro. On the other, ops know ionice and how/when to use it and does that actually belong in Swift? If it does, why not other ops/management tools?
+**** [https://docs.google.com/spreadsheets/d/1dzcMutNwRYcDiSMNOhKLM1tk8QX3uqFg9BJszJ6plr4/edit?usp=sharing Real tests result]
 ** Auditor Watchers patches [Zyric]
 *** Object Auditor Watcher patch - https://review.openstack.org/#/c/212824/