Closed rajarshd closed 6 years ago
Thanks @rajarshd for the report. The Neo4j Browser is loading for me, but it's providing the following error message:
Database access not available. Please use
:server connect
to establish connection.
Here's a screenshot:
I'll investigate.
Here are the digital ocean usage graphs. Looks like there was some large computation on 2018-10-12 that likely triggered the outage:
I've copied neo4j's debug.log
below, which appears to capture the error:
Of interest are the lines:
2018-10-12 01:35:26.099+0000 ERROR [o.n.b.t.SocketTransportHandler] Fatal error occurred when handling a client connection: Connection reset by peer Connection reset by peer
2018-10-12 01:37:12.695+0000 ERROR [o.n.b.t.SocketTransportHandler] Fatal error occurred when handling a client connection: Operation timed out Operation timed out
2018-10-12 05:46:45.136+0000 WARN [io.netty.channel.DefaultChannelPipeline] An exceptionCaught() event was fired, and it reached at the tail of the pipeline. It usually means the last handler in the pipeline did not handle the exception. GC overhead limit exceeded
I am not sure if the first two errors are meaningful, but the third indicates that the available memory was exhausted, which likely led to the database crashing. We've had memory issues before: see https://github.com/dhimmel/hetionet/issues/9 and https://github.com/neo4j/neo4j/issues/9209. I don't remember the browser still being accessible however for previous failures.
Okay I just rebooted the Digital Ocean droplet, so the Hetionet Browser is back up and working properly. I would love a solution to these crashes due to memory use, but don't realize have the ability/time to find a solution. Or at least a better way to have the neo4j instance restart upon crashing.
@rajarshd thanks for the report and please open a new issue if you notice anymore downtime!
Thanks for building the great resource. The Neo4J browser, however, has not been accessible for the last couple of days.
Thanks in advance for looking into it.