Bug 1370654
| Summary: | Auto Tune Cassandra Memory Usage Based on Limit | ||
|---|---|---|---|
| Product: | OpenShift Container Platform | Reporter: | Dan McPherson <dmcphers> |
| Component: | Hawkular | Assignee: | Matt Wringe <mwringe> |
| Status: | CLOSED ERRATA | QA Contact: | chunchen <chunchen> |
| Severity: | medium | Docs Contact: | |
| Priority: | medium | ||
| Version: | 3.3.0 | CC: | aos-bugs, penli, tdawson, twiest, whearn, wsun |
| Target Milestone: | --- | ||
| Target Release: | --- | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | If docs needed, set a value | |
| Doc Text: |
The Cassandra instance will now automatically configure its memory based on the pods memory and cpu limits. Previously this was configured via environment variables and these can still be used to override the memory configuration if desired.
|
Story Points: | --- |
| Clone Of: | Environment: | ||
| Last Closed: | 2016-09-27 09:46:28 UTC | Type: | Bug |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
"A typical approach is to tune the heap size to 60-70% of the limit value to allow room for native memory from the JVM. You may want to leave even more space for the Cassandra file cache." Cassandra has its own mechanism and recommended approaches for its memory configuration what we will be using: http://docs.datastax.com/en/cassandra/2.2/cassandra/operations/opsTuneJVM.html blocked by #1371578, will verify it once the bug is fixed. FYI: has #1371578 been closed as NOTABUG Verified with Image ID: docker://sha256:c951c67e1d8a47815d4dadc03ab487009f5f51084641cddad59218030ccaa17a After deploy metrics, the MAX_HEAP_SIZE and HEAP_NEWSIZE could be calculated correctly. In my test env, file cassandra-env.sh, MAX_HEAP_SIZE and HEAP_NEWSIZE is substitude to 1024M and 200M based on pod memory limits(3975499776 which is between 2 and 4G) and CPU_LIMIT(2). Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2016:1933 |
You can accomplish this in a couple of ways: 1) Read from the cgroup value directly CONTAINER_MEMORY_IN_BYTES=`cat /sys/fs/cgroup/memory/memory.limit_in_bytes` or 2) Use the downward api env: - name: MEM_LIMIT valueFrom: resourceFieldRef: resource: limits.memory A typical approach is to tune the heap size to 60-70% of the limit value to allow room for native memory from the JVM. You may want to leave even more space for the Cassandra file cache.