Indeed, the glusterfs endpoints have simple disappeared from the 2 projects of this cluster.
Discussing of this, I have discovered we had the same issue on a different cluster, with different projects.
We have recreated the endpoints, and had to redeploy all services relying on it.
While they seem to be back to normal, we still have error messages in the console :
TypeError: Cannot read property 'length' of undefined
We also see origin-node eating 20% CPU on each node, which seems a lot to me for only 2 projects.
We haven't switched to 1.1 yet (still using 1.0,8), we'll first do that and see if it continues.
Philippe Lafoucrière - CEO
main : +33 (0) 970 444 643
mobile CA: +1 (581) 986-7540
mobile FR: +33 (0) 6 72 63 75 40