Please help me explain unexpected "stop-the-world" pause entries that I'm seeing in "/var/log/neo4j/debug.log".
I'm running neo4j-enterprise v4.4.16 on a robust AWS EC2 instance running the current version of Rocky Linux (CentOS 8).
Here are some typical examples:
2023-01-31 15:45:09.244+0000 WARN [o.n.k.i.c.VmPauseMonitorComponent] Detected VM stop-the-world pause: {pauseTime=205, gcTime=0, gcCount=0}
2023-01-31 15:46:01.763+0000 WARN [o.n.k.i.c.VmPauseMonitorComponent] Detected VM stop-the-world pause: {pauseTime=132501, gcTime=0, gcCount=0}
2023-01-31 15:46:13.746+0000 WARN [o.n.k.i.c.VmPauseMonitorComponent] Detected VM stop-the-world pause: {pauseTime=11433, gcTime=0, gcCount=0}
2023-01-31 15:46:25.494+0000 WARN [o.n.k.i.c.VmPauseMonitorComponent] Detected VM stop-the-world pause: {pauseTime=12827, gcTime=0, gcCount=0}
2023-01-31 15:46:25.721+0000 WARN [o.n.k.i.c.VmPauseMonitorComponent] Detected VM stop-the-world pause: {pauseTime=8213, gcTime=0, gcCount=0}
I've confirmed as best I can (using "query.log") that there are no queries running against Neo4J while these occur. I note that the "gcTime" and "gcCount" variables each remain at 0 -- so there are no garbage-collect operations happening.
I'm hoping that somebody here can share more about the specifics of what causes Neo4J to emit these log entries.
I'm investigating an unrelated issue on the system in question -- something that my IDE (a VSCode remote server) is doing causes all connectivity to the system to be lost for periods of at least several minutes. I sometimes see evidence that "sssd_nss" is being terminated -- I don't know if this is a cause or symptom.
If these unexpected pauses are a symptom of some external issue then the timestamp of their occurrence may prove helpful and tracking down the root cause.
I appreciate the attention of this community.
