Note
Access to this page requires authorization. You can try signing in or changing directories.
Access to this page requires authorization. You can try changing directories.
Certain Azure HDInsight 5.1 cluster versions contain an issue where temporary OpenSSL-related directories accumulate under /tmp/tmp-*openssl. These directories are not automatically cleaned up by the system as intended. Over time, this leads to a disk usage spike to 100% on the headnodes, causing cluster instability and headnode unresponsiveness.
Symptoms
- Headnode becomes unreachable or slow.
- YARN/DFS operations fail due to lack of disk space.
- Health probes report headnode unhealthy.
- Logs show No space left on device errors.
- SSH sessions may fail or commands hang due to full /tmp.
Impact
Affects HDInsight clusters as below:
- Causes /tmp to reach 100% utilization.
- Headnodes enter unresponsive/unhealthy state.
- Can affect job submission, Ambari access, and essential HDInsight control-plane operations.
Root cause
A regression in the latest HDInsight 5.1 release introduced an issue where OpenSSL temporary directories created in /tmp/tmp-*openssl do not get cleaned up automatically as expected.
Recommended steps
To resolve this issue, run the following script action on headnodes
https://hdiconfigactions.blob.core.windows.net/openssl-patch/openssltmpclean.sh
Important
Automated cleanup included in the code path does not execute as expected. The cron-based cleanup is the recommended and supported mitigation until a patched HDInsight image is released.