Error changing memory.limit_in_bytes resource limit on existing gears

Solution Unverified - Updated -

Issue

We're altering the resource_limits.conf on a pair of nodes within the OpenShift Enterprise environment.
The resources we're altering are:

memory_limit_in_bytes=12884901888
memory_memsw_limit_in_bytes=13153337344

The bounds we're setting are perfectly within the capacity of the node host to work with. However, when we run the following commands:

service ruby193-mcollective restart; oo-cgroup-enable --with-all-containers; oo-pam-enable --with-all-containers; oo-admin-ctl-tc restart

The cgroup portion fails with:

CGroups parameter cannot be set to value: memory.limit_in_bytes = 12884901888

All the other commands complete successfully.

It's worth noting that subsequent gears are created with the correct (i.e. new) resource profiles - it's just legacy gears that seem to hang on to the old resource profile. Also, there are no pending operations in the gears, and it is possible to move them from one node to another in the district.

Is this a known issue? Is there a workaround? Or is there a step we're missing here?

Environment

  • OpenShift Enterprise 2.0
  • OpenShift Enterprise 2.1

Subscriber exclusive content

A Red Hat subscription provides unlimited access to our knowledgebase of over 48,000 articles and solutions.

Current Customers and Partners

Log in for full access

Log In