Show TOC

_ESP_Clients_MonitorLocate this document in the navigation structure

Contains information about the performance of all currently active gateway client connections and a copy of data from the _ESP_Clients stream. Monitoring data is available only if the time-granularity option in the project configuration (CCR) file is set to greater than 0. The frequency of updates corresponds to the value of the time-granularity option. For example, if set to 1, an update is published every second, if set to 30, an update is published every 30 seconds, and if set to 0, reporting is disabled.

Column Type Description
Handle long A unique integer ID of the connection.
user_name string The user name provided by the client during connection establishment. Shown once the user is authenticated.
IP string The IP address of the client machine, as a string.
host string The symbolic host name of the client machine, if available. If not available, host is the IP address of the client machine.
port integer The TCP port number from which the connection originates.
login_time msdate The time the server accepts (but does not authenticate) the connection.
conn_tag string The user-set symbolic connection tag name. If not set by the user, conn_tag is NULL.
cpu_pct float Total CPU usage for the client thread, as a percentage of a single CPU core.
last_update seconddate The time of the current update.
subscribed integer The status of a subscription to a stream: 1 if subscribed; 0 if not subscribed, indicating a publisher.
sub_trans_per_sec float The client's performance, in transactions per second, received by the client since the last update.
sub_rows_per_sec float The client's performance, in data rows per second, received by the client since the last update.
sub_inc_trans long The number of transactions, envelopes, or messages received by the client since the last update.
sub_inc_rows long The number of data rows received by the client since the last update.
sub_total_trans long The total number of transactions, envelopes, or messages received by the client. Transactions, envelopes, and messages still in the queue are not counted.
sub_total_rows long The total number of data rows received by the client. Rows still in the queue are not counted.
sub_dropped_rows long The total number of data rows dropped in the gateway because they were not read quickly enough by the client. For lossy subscriptions.
sub_accum_size integer The current number of rows collected in the accumulator to be sent in the next pulse. For pulsed subscriptions.
sub_queue integer The number of rows queued for transmission to the client.
sub_queue_fill_pct float The current sub_queue, as a percentage, relative to the queue size limit. If sub_queue_fill_pct reaches 100 percent, any future attempts to post data to this client are blocked, propagating the flow control back to the source of the post.
sub_work_queue integer The number of rows for transmission to the client that are being transferred from the proper queue to the socket buffer. The rows can be regrouped by envelopes.
pub_trans_per_sec float The client's performance, in transactions per second, sent by the client since the last update. Envelopes and any service messages count as transactions.
pub_rows_per_sec float The client's performance, in data rows per second, sent by the client since the last update.
pub_inc_trans long The number of transactions, envelopes, or messages sent by the client since the last update.
pub_inc_rows long The number of data rows sent by the client since the last update.
pub_total_trans long The total number of transactions, envelopes, or messages sent by the client. Does not include the transactions/envelopes/messages in the queue.
pub_total_rows long The total number of data rows sent by the client. Does not include the rows in the queue.
pub_stream_id long The numeric ID of the stream to which the client is trying to currently publish data. Typically, pub_stream_id is -1, meaning that data is not currently being published to the stream. Since the server publishes data in very short bursts, the monitor will usually not report any value other than -1.
node_cpu_pct float Total CPU usage for the client, as a percentage of all CPU cores on the machine. Total CPU usage equals system CPU usage plus user CPU usage.
node_cpu_pct_system float System CPU usage for the client, as a percentage of all CPU cores on the machine.
node_cpu_pct_user float User CPU usage for the client, as a percentage of all CPU cores on the machine.
cpu_time interval Total CPU time since the creation of the client, in microseconds. Total CPU time equals system CPU time plus user CPU time.
cpu_time_system interval Total system CPU time, in microseconds, since the creation of the client thread.
cpu_time_user interval Total user CPU time, in microseconds, since the creation of the client thread.
time_since_start interval Duration of lapsed real time since the creation of the client thread.