Mid Server restarts frequently


Description

Symptoms

Agent Log shows the below error:

/Downloads/log1/agent0.log.0:23443: 2019/05/14 16:57:24 (109) StartupSequencer WARNING *** WARNING *** Encountered error: [An active MID Server with a duplicate name detected.] in ensuring agent record on the instance. Retry...

Wrapper logs show the below error:

2019/05/14 16:57:24 | [error occurred during error reporting (null), id 0xc0000005]
2019/05/14 16:57:24 | 
2019/05/14 16:57:24 | #
2019/05/14 16:57:24 | # There is insufficient memory for the Java Runtime Environment to continue.
2019/05/14 16:57:24 | # Native memory allocation (malloc) failed to allocate 32744 bytes for ChunkPool::allocate
2019/05/14 16:57:24 | # An error report file with more information is saved as:
2019/05/14 16:57:24 | # C:\ServiceNow\agent\hs_err_pid23672.log
2019/05/14 16:57:24 | #
2019/05/14 16:57:24 | # Compiler replay data is saved as:
2019/05/14 16:57:24 | # C:\ServiceNow\agent\replay_pid23672.log
2019/05/14 16:57:24 | JVM exited unexpectedly.

Release

  • All versions¬†

Cause

By default, the mid server has wrapper.java.maxmemory=1024 and max concurrent threads set to 25, which may not be sufficient to carry out the transactions in the environment and once it reaches the max memory, the mid server goes down.

Resolution

Validate if the mid server meets the minimum system requirements to carry out the transactions in your environment and increase the JVM memory or adjust the maximum threads to meet the requirements.

Additional Information

  1. Set the MID Server JVM memory size
  2. Set MID Server thread use
  3. MID Server system requirements