Hello,
I have 3 mongos and 2 shards, each shard have three members. From one shard primary member went down and when we restarted it, we see huge connection on this node.
Error message in mongos log:
I ASIO [NetworkInterfaceASIO-TaskExecutorPool-26-0] Connecting to *****:
I ASIO [NetworkInterfaceASIO-TaskExecutorPool-1-0] Connecting to ******:
I ASIO [NetworkInterfaceASIO-TaskExecutorPool-44-0] Connecting to *****:
I ASIO [NetworkInterfaceASIO-TaskExecutorPool-46-0] Failed to connect to ***: - NetworkInterfaceExceededTimeLimit: Operation timed out
I ASIO [NetworkInterfaceASIO-TaskExecutorPool-2-0] Failed to connect to ***: - NetworkInterfaceExceededTimeLimit: Operation timed out
I ASIO [NetworkInterfaceASIO-TaskExecutorPool-11-0] Failed to connect to ***: - NetworkInterfaceExceededTimeLimit: Operation timed out
When we started this secondary node we see drastic connection connect (42/min → 245722/min → 3435669/min to this node in mongos log.
Using below connection pooling parameter in mongos:
–setParameter connPoolMaxShardedConnsPerHost=1400 --setParameter connPoolMaxConnsPerHost=1400 --setParameter ShardingTaskExecutorPoolMinSize=12
Using below connection parameter in application connection string.
wtimeoutMS=0&retryWrites=true&w=2