[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: OS 1.0.7 eating memory like candies after halloween



Check the number of events you have in etcd.  If it's freakishly large, look for examples of them.

On Nov 3, 2015, at 3:03 PM, Philippe Lafoucrière <philippe lafoucriere tech-angels com> wrote:

Hi,

We don't know why yet, but our master is eating memory at an impressive speed rate.
8Gb in ~ 1 minute, then all the swap, and then restart.
We're trying to figure out why, but something is really wrong with this issue, something is stacking to death.
We don't have any clue from the logs, except:

node [replace here with node 1 2 and 3] hasn't been updated for 1m30.538112963s. Last ready condition is: {Type:Ready Status:Unknown LastHeartbeatTime:2015-11-03 20:46:40 +0100 CET LastTransitionTime:2015-11-03 20:48:11 +0100 CET Reason:NodeStatusUnknown Message:Kubelet stopped posting node status.}

for the 3 nodes. We'll keep this thread updated, and post an issue on github if we have relevant information.

Thanks

-- 
Philippe Lafoucrière - CEO
main : +33 (0) 970 444 643
mobile CA: +1 (581) 986-7540
mobile FR: +33 (0) 6 72 63 75 40

_______________________________________________
users mailing list
users lists openshift redhat com
http://lists.openshift.redhat.com/openshiftmm/listinfo/users

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]