Re: Nodes marked dead…. leap second?

2012-07-02 Thread Henrik Schröder
Bug: https://lkml.org/lkml/2012/6/30/122 Simple fix to reset the leap second flag: date; date `date +"%m%d%H%M%C%y.%S"`; date; /Henrik On Mon, Jul 2, 2012 at 1:56 PM, Jean Paul Adant wrote: > Hi, > > I did have the same problem with cassandra 1.1.1 on Ubuntu 11.10 > I had to reboot all nodes >

Re: Nodes marked dead…. leap second?

2012-07-02 Thread Jean Paul Adant
Hi, I did have the same problem with cassandra 1.1.1 on Ubuntu 11.10 I had to reboot all nodes I'm interested in any information about this. Thanks Jean Paul 2012/7/2 Filippo Diotalevi > Hi, > we had some really weird issues during the weekend, with our cassandra > nodes starting marking as

Nodes marked dead…. leap second?

2012-07-02 Thread Filippo Diotalevi
Hi, we had some really weird issues during the weekend, with our cassandra nodes starting marking as dead other (working) nodes in the cluster. That happened all Sunday, and it's still happening. Node are marked dead and up all the time…. Some example logs: INFO [GossipTasks:1] 2012-07-02 06: