Re: Review Request 54288: Make leader elections resilient to ZK disconnections.

> On Dec. 2, 2016, 3:58 p.m., Joshua Cohen wrote: > > src/test/java/org/apache/aurora/scheduler/discovery/CuratorSingletonServiceTest.java, > > lines 215-218 > > > > > > Should we have an escape hatch for the case

Re: Review Request 54288: Make leader elections resilient to ZK disconnections.

> On Dec. 2, 2016, 10:37 p.m., Zameer Manji wrote: > > Can someone else validate that the tests pass for them locally? I can't > > reproduce the jenkins failure. I wonder if this is a timing issue in the tests ? `expectGroupEvent(PathChildrenCacheEvent.Type.CHILD_ADDED)` returns as soon as

Re: Review Request 54288: Make leader elections resilient to ZK disconnections.

--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/54288/#review157847 --- Can someone else validate that the tests pass for them locally? I

Re: Review Request 54288: Make leader elections resilient to ZK disconnections.

--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/54288/#review157844 --- Master (4bc5246) is red with this patch.

Re: Review Request 54288: Make leader elections resilient to ZK disconnections.

--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/54288/#review157840 --- @ReviewBot retry - Zameer Manji On Dec. 1, 2016, 7:19 p.m.,

Re: Review Request 54269: Improve scheduling throughput via logging changes.

--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/54269/#review157838 --- Master (3ea0331) is green with this patch.

Re: Review Request 54269: Improve scheduling throughput via logging changes.

> On Dec. 1, 2016, 5:41 p.m., Stephan Erb wrote: > > src/main/java/org/apache/aurora/scheduler/state/TaskStateMachine.java, line > > 474 > > > > > > Please use the logging builtin formatting rather than doing the

Re: Review Request 54269: Improve scheduling throughput via logging changes.

> On Dec. 2, 2016, 11:29 a.m., Joshua Cohen wrote: > > src/main/resources/logback.xml, line 26 > > > > > > Can you add a comment here explaining why we're not using `%class` and > > `%line`? They're definitely

Re: Review Request 54269: Improve scheduling throughput via logging changes.

--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/54269/ --- (Updated Dec. 2, 2016, 1:51 p.m.) Review request for Aurora, David McLaughlin,

Re: Review Request 54299: Extend warm-up time by `max_consecutive_failures` attempts.

--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/54299/#review157764 --- src/main/python/apache/aurora/executor/common/health_checker.py

Re: Review Request 54299: Extend warm-up time by `max_consecutive_failures` attempts.

--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/54299/#review157807 --- Ship it! It took me a long time to understand this after

Re: Review Request 54269: Improve scheduling throughput via logging changes.

--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/54269/#review157799 --- Ship it! src/main/resources/logback.xml (line 26)

Re: Review Request 54288: Make leader elections resilient to ZK disconnections.

--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/54288/#review157759 --- Thanks for picking this up! This is a basic question, but I just

Re: Review Request 54299: Extend warm-up time by `max_consecutive_failures` attempts.

--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/54299/#review157722 --- Ship it! Master (3ea0331) is green with this patch.

Review Request 54299: Extend warm-up time by `max_consecutive_failures` attempts.

--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/54299/ --- Review request for Aurora, David McLaughlin, Joshua Cohen, Stephan Erb, and