----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/11128/#review20721 -----------------------------------------------------------
Ship it! src/master/master.cpp <https://reviews.apache.org/r/11128/#comment42788> Just to be sure, this should only be called if the slave is checkpointing. The fact that you have seen this bug, tells me that you are running slaves with "--checkpoint"! Please note that this feature is still experimental! - Vinod Kone On May 14, 2013, 12:21 a.m., Brenden Matthews wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/11128/ > ----------------------------------------------------------- > > (Updated May 14, 2013, 12:21 a.m.) > > > Review request for mesos. > > > Description > ------- > > From 2fa3b02c3a0042f8f76ed9055d2e74ee010dc635 Mon Sep 17 00:00:00 2001 > From: Brenden Matthews <[email protected]> > Date: Mon, 6 May 2013 09:54:03 -0700 > Subject: [PATCH 21/24] Terminate correct tasks when a slave disconnects. > > Previously, when a slave disconnected all tasks for that framework would > be removed and it would result in a bad state for a given framework. In > the case of Hadoop, it would result in a bunch of zombie tasks running > on the slaves which never terminate. > > Review: https://reviews.apache.org/r/11128 > --- > src/master/master.cpp | 47 +++++++++++++++++++++++++---------------------- > 1 file changed, 25 insertions(+), 22 deletions(-) > > > Diffs > ----- > > src/master/master.cpp 3207157961e772ae543a236536ae1559acfe110b > > Diff: https://reviews.apache.org/r/11128/diff/ > > > Testing > ------- > > Used in production at airbnb. > > > Thanks, > > Brenden Matthews > >
