We are running MQ connector in multinode (4 nodes - Conductor/Compute Configuration). The purpose of this job is read data from Source Queue, move to work queue and transform the data from source queue. In case of any failures the data from work queue will be moved to source queue and job will be triggered again.
We observed there is a delay for one of the node to get disconnected from queue manager, quite surprising it took around 2 hours and 4 minutes in one occasion. The time it takes for disconnecting from queue manager changes for each run. The wait time in MQ connector is set to 1 sec and job is scheduled to run every run.
Can someone throw some light the reason for the delay in getting the message 'Disconnecting from the Queue Manager' and 'Disconnected Sucessfully'.
Config file:
Code: Select all
{
node "conductor"
{
fastname "xxxxx1"
pools "conductor"
resource disk "/etl/IS/Datasets" {pools "buffer"}
resource scratchdisk "/etl/IS/Scratch" {pools "buffer"}
}
node "node1"
{
fastname "xxxxx2"
pools ""
resource disk "/etl_s1/IS/Datasets50" {pools ""}
resource scratchdisk "/etl_s1/IS/Scratch50" {pools ""}
}
node "node2"
{
fastname "xxxxx2"
pools ""
resource disk "/etl_s1/IS/Datasets51" {pools ""}
resource scratchdisk "/etl_s1/IS/Scratch51" {pools ""}
}
node "node3"
{
fastname "xxxxx2"
pools ""
resource disk "/etl_s1/IS/Datasets52" {pools ""}
resource scratchdisk "/etl_s1/IS/Scratch52" {pools ""}
}
node "node4"
{
fastname "xxxxx2"
pools ""
resource disk "/etl_s1/IS/Datasets53" {pools ""}
resource scratchdisk "/etl_s1/IS/Scratch55" {pools ""}
}
}