Skip to content
This repository has been archived by the owner on Jan 30, 2020. It is now read-only.

fleet start fails arbitrarily on cluster with more than 500 nodes #1579

Open
Vishant0031 opened this issue May 4, 2016 · 2 comments
Open

Comments

@Vishant0031
Copy link

Test environment:
coreos version 835.11
fleetd version 0.11.5
etcd Version: 2.2.0

Fleet nodes:
700

Issue:
I am trying to start a fleet service on my cluster. I use command fleetctl start (diskinstall.service).
fleetctl list-machines shows 700 nodes.
and fleet service is a Global=true service and suppose to be launched on all the nodes.
When I start the service (fleetctl start diskinstall.service), service is started only on <700 nodes.
Sometimes the service is launched on ~400 nodes or ~500 and its very random.

At all the time fleetctl list-machines outputs 700 nodes.
etcd2 is also healthy all the time.
I have to repeat the fleetctl start command to overcome this issue.

Thanks.

@Vishant0031
Copy link
Author

@waddles
Copy link

waddles commented Jul 26, 2016

Does specifying an etcd endpoint as in #1650 make it better for you?

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Projects
None yet
Development

No branches or pull requests

3 participants