oracle probe alert (Profile timed out, the probe may be hanging!)

Products

• DX Unified Infrastructure Management (Nimsoft / UIM)

Issue/Introduction

We getting a huge number of alerts (Profile xxxxxl8xxxxps_xxxxxxxx timed out, the probe may be hanging!) from the oracle probe .

1) Why this type of alert is generating.
2) How to solve this issue

Environment

• Release: 9.2.0 or higher
• Component: UIM - ORACLE

Resolution

You have a few options:

1 - Talk to your Oracle Database administrator to check the ENQUEUE_RESOURCES and DML_LOCKS

Some hints:

Number of table and row locks (acquired and converted) that timed out before they could complete. Consider to increase parameter ENQUEUE_RESOURCES or DML_LOCKS.

dml_locks
processes are Oracle initialization parameters.) The number of processes defaults to 50. So the default value for dml_locks = 4 * (1.1 * 50) = 220. However, depending on number of users, this may not be large enough for your system. Not having enough dml_locks and/or enqueue_resources may cause latch contention.

enqueue_resources
the number of resources that can be locked by the lock manager. The value of enqueue_resources should be no greater than dml_locks + 20.

2 - You can just disable the alarm to no about the enqueue_timeouts

   <enqueue_timeouts>
      type = 18
      active = yes
      qos = yes
      alarm = no
      qos_list = yes
      description = Monitors number of table and row locks that timed out
      clear_msg = rac_enqueue_timeouts_1
      clear_sev = clear
      scheduling = rules
      msg_variables = $check.x;$profile.x;$instance.x;$INSTANCE_NAME.x;$ENQUEUE_TIMEOUTS.n;$interval_value.n;
      condition = >=
      column = ENQUEUE_TIMEOUTS
      key = $INSTANCE_NAME
      samples = 1
      interval = 
      <qos_lists>
         <0>
            qos_name = enqueue_timeouts
            qos_desc = Oracle Enqueue Timeouts
            qos_unit = Timeouts
            qos_abbr = Count
            qos_max = 0
            qos_value = ENQUEUE_TIMEOUTS
            qos_key = $INSTANCE_NAME
         </0>
      </qos_lists>
   </enqueue_timeouts>

3 - Create a Pre-Processing rule in the Auto-Operators in the nas probe to exclude these alarms.

The Pre-Processing Rules Tab (nas probe documentation)

4 - Increase the profile time value:

Additional Information

The oracle probe has a limit of up to 50 connections. If you go above that limit you will normally encounter performance and scalability issues.

Also note that if you have hundreds of profiles, the probe may be overloaded. Examine the "Check Interval" and make sure it isn't set to 30 seconds or less. Try 5-minute intervals.