Re: Warning TcpCommunicationSpi: Failed to allocate shared memory segment
Please properly subscribe to the user list (this way we will not have to manually approve your emails). All you need to do is send an email to ì email@example.comî and follow simple instructions in the reply.
In regards to your question, yes, Ignite automatically use the shared memory capabilities on Unix machines if there are nodes that are running on the same physical host. In fact this exception is harmless it says that a process won't be able to allocate shared memory segment for its needs and the process switched back to TCP/IP based communication layer. To answer precisely what happened please share all the logs from all the nodes.
I have started several Ignite nodes (each running in a separate JVM) in the same machine. Then I use cache.invoke in these nodes for updating a cache value. Cache entry is updated successfully.
I have inspected the logs and there are entries like:
16/06/09 23:27:04 WARN TcpCommunicationSpi: Failed to allocate shared memory segment (switching to TCP, may be slower).
Is it the case Ignite uses shared memory for communicate Ignite nodes running in the same computer? Any advise on how to avoid the "Failed to allocate shared memory segment" problem?