VMware to GCP issue

[kkbaum]

I have a working MALCOLM image from VMWare Workstation that our network engineer converted and uploaded to our Google Cloud platform. When starting MALCOLM I get the following errors constantly till I kill the script:

pcap-monitor-1 | 2024-11-22 18:49:50 WARNING: GET http://opensearch:9200/_cluster/health [status:N/A request:4.345s]
pcap-monitor-1 | ConnectionRefusedError: [Errno 111] Connection refused
pcap-monitor-1 |
pcap-monitor-1 | The above exception was the direct cause of the following exception:
pcap-monitor-1 |
pcap-monitor-1 | urllib3.exceptions.NewConnectionError: <urllib3.connection.HTTPConnection object at 0x7f6062989f50>: Failed to establish a new connection: [Errno 111] Connection refused

I know Google makes changes to accommodate the image but I have yet in three days discovered what could be causing the issue. I have rebuilt MALCOLM from the Git files, verified network settings, turned off any firewalls and even changed the MTU of the Docker0 interface to match the local adapter.

[mmguero]

It's normal to see that maybe once or twice during startup, but eventually once opensearch starts it should resolve. The most likely issues is a lack of resources for the OpenSearch container. So, a couple of questions for you:

1. What are the specs for the VM as far as memory/CPU goes?
2. What's the output of grep JAVA_OPTS config/*.env when run from your Malcolm installation directory?

[kkbaum]

1. Specs are an e2-medium which consists of 2 single-core vCPUs and 4gb RAM.
2. Output from files:
   config/logstash.env:LS_JAVA_OPTS=-server -Xmx2g -Xms2g -Xss1536k -XX:-HeapDumpOnOutOfMemoryError -Djava.security.egd=file:/dev/./urandom -Dlog4j.formatMsgNoLookups=true
   config/opensearch.env:OPENSEARCH_JAVA_OPTS=-server -Xmx3500m -Xms3500m -Xss256k -XX:-HeapDumpOnOutOfMemoryError -Djava.security.egd=file:/dev/./urandom -Dlog4j.formatMsgNoLookups=true

[mmguero]

That's your issue, then. Malcolm requires at a minimum 16GB of ram and 8 logical CPUs.