Jyotirmoy Sinha created HDDS-11189:
--------------------------------------
Summary: OM Disk getting filled up on snapshots which leads to OM
service outage
Key: HDDS-11189
URL: https://issues.apache.org/jira/browse/HDDS-11189
Project: Apache Ozone
Issue Type: Bug
Components: OM, Snapshot
Reporter: Jyotirmoy Sinha
OM Disk getting filled up on snapshots which leads to OM service outage
Scenario :
* Generate data over parallel threads over various volume/buckets
* Perform parallel snapshot create/delete/list operations over above buckets
* Perform parallel snapdiff operations over each bucket
* Perform parallel read operations of snapshot contents
* Introduce OM and cluster restarts in between along with DN decommissioning
and balancer restarts.
OM Leader disk is getting filled by snapshot data upto 100% -
{code:java}
/dev/sda3 1.1T 1.1T 20K 100% /var {code}
OM Data disk sizes -
{code:java}
1.1T /var/lib/hadoop-ozone/om/data
175M /var/lib/hadoop-ozone/om/ozone-metadata
2.1G /var/lib/hadoop-ozone/om/ratis {code}
Other OM nodes are also filled up upto 100% -
{code:java}
/dev/sda3 1.1T 1.1T 18M 100% /var {code}
{code:java}
/dev/sda3 1.1T 1.1T 55M 100% /var {code}
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
