The Meridian 1 can be set to run diagnostic overlay programs at a specified time, normally in a quiet traffic period (eg, at midnight). The automatic daily running of diagnostic overlays helps to identify and correct faults before users become aware of them.
When the Meridian 1 detects a fault during daily routines, it will automatically run a diagnostic program, overlay program 45, to identify and report the fault condition. A Data Save/Dump (LD 43) is also be performed during the daily routines.
Which Overlays are run
You can check which overlays are set to run on your system, and at what time, by printing the Configuration Record (CFN) in LD 22. This provides a lot of information, but the part pertaining to background and daily routine tasks is shown in the "OVLY" section. You may also print just the overlay data, by entering OVLY at the TYPE prompt.
>LD 22 REQ prt TYPE ovly OVLY SID 0 - System ID number BKGD 044 - Background Overlay task (44: Software Audit) FR44 1 - Frequency control of Overlay 44 in background PBXH X - Hour to perform Message Waiting lamp maintenance, X to disable TODR 00 - Time of Daily Routines (Midnight) DROL 043 045 060 135 137 - Daily Routine Overlays MID_SCPU NO - Midnight Switch CPU Allow or deny (CPP systems only) CY45 00 - Cycles LD 45 can be run whenever a fault is detected MULTI_USER ON - Multi-User Log In
Adding Overlays to the Daily Routines
This is done in LD 17.
REQ chg TYPE ovly SID BKGD PBXH TODR DROL 30 34 - Add overlays separated by space, X to remove MULTI_USER
Valid Background overlays:
30, 33, 34, 35, 36, 37, 38, 40, 41, 43, 44, 45, 46, 53, 135
Valid Daily Routine overlays:
30, 32, 33, 34, 35, 36, 37, 38, 40, 41, 43, 44, 45, 46, 51, 53, 60, 61, 135
To remove an overlay, type X before the overlay number.
- LD 30 (and LD 45) in the DROL will disable any digital sets which are unplugged. In stead of removing LD 30, prevent a set from being disabled by adding CLS FLXA. A set with the class of service FLXA will not disable during the daily routines.
- LD 135 in the DROL will cause a core swap, with
MID_SCPU (NO) YES- Default: YES for Motorola CPs, NO for Pentium CPs.
- You can manualy invoke the DROL at any time (to test) with LD 135, MIDN, and then exit or logout.
It is good practice to always perform a data dump after making any programming changes.
>LD 43 . edd - Perform an Equipment Datadump (data save) DATA DUMP COMPLETE
Midnight CPU Switchovers
As the newer Pentium Call Processors are more 'robust', an automatic daily core swap is no longer required and has been removed from the default daily routine. However, it's still possible to perform a daily core switchover during the midnight routine if desired:
>LD 17 REQ chg TYPE ovly DROL 135 - Add 135 to Daily Routine Overlays MID_SCPU yes - Midnight Switch CPU: Allow (NO will test 3PE only, with LD 135)
- If the customer can tolerate a service interruption of 6 to 10 seconds during the daily routines, and if the system is monitored for critical alarms, then a daily scheduled core switchover is recommended.
- If the customer cannot tolerate a daily scheduled service interruption, then a manual core switch test should be conducted during a scheduled maintenance window at least once a year, and preferable monthly or quarterly. See below.
- Never switching active cores is not recommended, as a faulty inactive core may not be discovered until the active core fails.
These recommendations apply to all dual processor systems.
Manual core switch test: In LD 135, ensure the system is redundant with
STAT CPU, and enter
SCPU to manually switch cores.
A graceful switchover causes a 6 to 10 second service interruption, an ungraceful switchover takes a little longer (excess of 1 min).
Generally, established basic calls survive both the graceful and ungraceful switchovers. Basic calls that are in a transient state (for example, calls that are in the dialing state) survive a graceful switchover but do not survive an ungraceful switchover.
Checking the Daily Routine output
There are two ways you can do this. The simplest way is to leave the maintenance terminal or printer on/connected overnight, and check the output in the morning. PuTTY is useful due to its "scroll-back buffer" which should be set to the maximum size.
Alternatively, you can read the System Event List (SEL) with LD 117. This is also useful for checking other system event messages.
>LD 117 => prt sel 20 - printout length maybe 0-(20)-SELSIZE Printing [4%] (20 of 500) most recent events DROL000 DAILY ROUTINE BEGIN TDS LD34 BEGIN 13:42 8/8/2001
This will print the most recent records in system event list.
SELSIZE is usually 500 (
PRT SELSIZE to find the setting), meaning the
system event log will hold 500 events. The SEL is also searchable using the
FIND XXXX command.
You can also manually invoke the routines, and watch the output.
>ld 135 CCED000 .midn OK .**** - exit the overlay to start DAILY ROUTINES > DROL000 DAILY ROUTINE BEGIN TDS LD34 BEGIN 13:42 8/8/2001
MIDN command (in LD 135) will run the routines after LD 135 is exited.
Normally, no problems should occur. A typical printout will show each overlay in the DROL list starting and exiting normally. Any problems will be shown by the relavent system message, SCH, ERR, etc. Again, these messages will be stored in the system event log.
DROL000 DAILY ROUTINE BEGIN DTIM LD60 BEGIN 00:04 8/8/2001 DROL001 DAILY ROUTINE END DTIM LD60 END 00:04 8/8/2001 DROL000 DAILY ROUTINE BEGIN CCED LD135 BEGIN 00:04 8/8/2001 DROL001 DAILY ROUTINE END CCED LD135 END 00:04 8/8/2001 AUD0000
The 'AUD0000' is the output from LD 44 (Software Audit) and indicates that an error free audit pass was made. No errors detected. System software OK. No action to be taken. Happy, happy!