Difference between revisions of "Meetings/Swift"

Latest revision as of 20:23, 1 May 2024

2100 UTC Meeting
Next meeting: May 1, 2024
Location: #openstack-swift on OFTC
Chair: timburke
Utils refactor - expect merge conflicts! https://review.opendev.org/c/openstack/swift/+/914029 Probe test time outs https://zuul.opendev.org/t/openstack/builds?job_name=swift-probetests-centos-9-stream&job_name=swift-probetests-centos-8-stream&project=openstack%2Fswift&result=TIMED_OUT&skip=0&limit=100 Some of these make sense -- some patchesets break every probe test and our retry-failed-tests logic causes the test time to essentially double Others just seem to hang for an hour or more liberasurecode release https://review.opendev.org/c/openstack/liberasurecode/+/917784 Nothing major, but it's been a couple years pyeclib manylinux wheels https://review.opendev.org/c/openstack/pyeclib/+/817498 - Add Dockerfile to build manylinux wheels https://review.opendev.org/c/openstack/pyeclib/+/917857 - Add job to build wheels expirer work https://review.opendev.org/c/openstack/swift/+/914713 - expirer: new options to control task iteration https://review.opendev.org/c/openstack/swift/+/916026 - distributed parallel task container iteration https://review.opendev.org/c/openstack/swift/+/912496 - add bytes of expiring objects to queue entry broker row insertion order py2/py3 behavior difference unreleased swiftclient regression https://review.opendev.org/c/openstack/python-swiftclient/+/916135

Meeting Logs	http://eavesdrop.openstack.org/meetings/swift/2024/
Useful Commands	#link #info #agreed #topic and #startmeeting

PTG action items
- (timburke) interop feedback on https://review.opendev.org/c/openinfra/interop/+/811049/
- better defaults (or at least, use better defaults in install/deployment guides)
  - (mattoliverau) recommend servers-per-port
  - (mattoliverau) make etherpad to collect more defaults that need updating
  - Etherpad: https://etherpad.opendev.org/p/swift-better-defaults
- (acoles) dark data audit watcher patch: https://review.opendev.org/c/openstack/swift/+/787656
- (timburke) abandon old patches
- (mattoliverau) bug triage/squash
  - Etherpad: https://etherpad.opendev.org/p/swift-bug-triage-a-thon
  - Ethercalc: https://ethercalc.openstack.org/mf3yro7018m0
- (acoles) drop logging translations - https://bugs.launchpad.net/swift/+bug/1674543
- (timburke) look into how to translate docs so seongsoocho can propose Korean translations
- check continued usefulness of the ops runbook
- consolidate various admin guides
- write a "why swift?" doc(/README?) section

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

@@ Line 3: / Line 3: @@
 ! '''2100 UTC Meeting'''
 |-
-| '''Next meeting:''' Nov 30th, 2022
+| '''Next meeting:''' May 1, 2024
 |-
 | '''Location:''' #openstack-swift on OFTC
@@ Line 11: / Line 11: @@
 |
 <!-- List 2100UTC meeting items below here -->
-* CPython broke our func tests
+* Utils refactor - expect merge conflicts!
-** https://github.com/python/cpython/issues/99220 | BaseHTTPRequestHandler.parse_request() loses client-provided information
+** https://review.opendev.org/c/openstack/swift/+/914029
-*** Still no response :-/
+* Probe test time outs
-** DSVM job is currently pinned to focal: https://review.opendev.org/c/openstack/swift/+/864946
+** https://zuul.opendev.org/t/openstack/builds?job_name=swift-probetests-centos-9-stream&job_name=swift-probetests-centos-8-stream&project=openstack%2Fswift&result=TIMED_OUT&skip=0&limit=100
-** https://review.opendev.org/c/openstack/swift/+/863441 | Preserve client-provided paths that start '//'
+** Some of these make sense -- some patchesets break every probe test and our retry-failed-tests logic causes the test time to essentially double
-*** Has one +2, any objections?
+** Others just seem to hang for an hour or more
-* Testing under py310
+* liberasurecode release
-** https://review.opendev.org/c/openstack/requirements/+/863581/ | Add new dependency: pytest-cov
+** https://review.opendev.org/c/openstack/liberasurecode/+/917784
-*** Still current blocker; just pinged to see if we can move forward on it
+** Nothing major, but it's been a couple years
-* We helped find a cpython 3.11.0 bug!
+* pyeclib manylinux wheels
-** zigo first discovered it: https://meetings.opendev.org/irclogs/%23openstack-swift/%23openstack-swift.2022-11-24.log.html
+** https://review.opendev.org/c/openstack/pyeclib/+/817498 - Add Dockerfile to build manylinux wheels
-** timburke eventually tracked it down and wrote up https://github.com/python/cpython/issues/99886
+** https://review.opendev.org/c/openstack/pyeclib/+/917857 - Add job to build wheels
-** Looks like there will probably be a fix for 3.11.1: https://github.com/python/cpython/pull/99902
+* expirer work
-** Is it worth us making changes in swift to avoid the segfault? https://review.opendev.org/c/openstack/swift/+/866051 seems to cover it, but if it only impacts one cpython release...
+** https://review.opendev.org/c/openstack/swift/+/914713 - expirer: new options to control task iteration
-* ring v2
+** https://review.opendev.org/c/openstack/swift/+/916026 - distributed parallel task container iteration
-** https://review.opendev.org/c/openstack/swift/+/865066 | WIP: Rings as zipfiles
+** https://review.opendev.org/c/openstack/swift/+/912496 - add bytes of expiring objects to queue entry
-** https://review.opendev.org/c/openstack/swift/+/866082 | Add swift-ring-packer utility
+* broker row insertion order py2/py3 behavior difference
-<!--
+* unreleased swiftclient regression
-** assorted other patches that seemed related to the investigation -- are they still relevant? please abandon if not
+** https://review.opendev.org/c/openstack/python-swiftclient/+/916135
-*** https://review.opendev.org/c/openstack/swift/+/849453 - container: broaden is_deleted method by using empty
-*** https://review.opendev.org/c/openstack/swift/+/848930 - Remove :memory: from DatabaseBrokers and unittests
-*** https://review.opendev.org/c/openstack/swift/+/848962 - ContainerBroker: modify determination of object_count
-*** https://review.opendev.org/c/openstack/swift/+/848909 - Make in-memory databases worse
-*** https://review.opendev.org/c/openstack/swift/+/844553 - sharding: shard deleted containers that have shards with objects
--->
 <!-- End 2100UTC meeting items -->
 |}
@@ Line 41: / Line 35: @@
 {| class="wikitable"
 |-
-| '''Meeting Logs''' || http://eavesdrop.openstack.org/meetings/swift/2022/
+| '''Meeting Logs''' || http://eavesdrop.openstack.org/meetings/swift/2024/
 |-
 |'''Useful Commands'''|| #link #info #agreed #topic and #startmeeting