Develop a Maintenance Plan
The primary goal of a maintenance plan is to maintain integrity of the data collected. If you are successful in this regard, you will always be able to recover from a service interruption and continue operation with minimal or no loss of data. Since you can never ensure 100% system uptime, you must frequently and regularly back up current data and configuration files, and maintain non-current archive files in a read-only state. following guidelines for backup and routine maintenance.
Daily Maintenance
- Use Historian Administrator to back up the current archive and most recent .IHA archived data file. This preserves data collected up to this moment in time. You do not need to back up any read-only archive files after they have been backed up once.
If you are doing more than a nightly backup, or backing up more than the last two archives, use the Microsoft Volume Shadow Copy Service (VSS) to preserve memory. Refer to the Back Up an Archive using Volume Shadow Copy Service section for more information.
- Use Windows Explorer to back up the .IHC file if it has been modified, unless it is backed up automatically. This file contains all current configuration information (tag configuration, archive configuration, and collector configuration). Using this file, you can restore the system configuration after an unplanned shutdown.
Routine Maintenance
On a regular schedule, examine and analyze the system performance indicators displayed on the System Statistics page of Historian Administrator as follows.
| Field | Recommended Action |
|---|---|
| Est. Days to Full | If time is growing short, make sure that the server has sufficient unused storage capacity to open a new archive when the active one fills up. Verify that the Create New Archives Automatically function is enabled. If it is disabled, you must manually create a new archive before the active archive fills up. Note: If you do not have enough unused storage capacity, you may have to enable the Overwrite Old Archives feature. Since this feature overwrites existing data, exercise caution in using it. |
| Consumption Rate of Archive Storage | If the rate is excessively high, reduce the rate at which data flows into the system or increase the filtering applied to the data to lower the rate of archiving. To reduce the collection rate, slow the polling rate on some or all tags. To increase filtering, enable compression at the collector and/or archiver and widen the compression deadbands. |
| Failed Writes | If the display shows a significant number of failed writes, investigate the cause and take corrective action to eliminate the malfunctions. Refer to the DataArchiver-XX.log file or query the message database to determine the tags for which failed writes occurred. For example, trying to write values to a deleted archive causes failed writes. Trying to archive data with a timestamp that precedes the start time of the first archive, trying to write to a read-only archive, or trying to write a value with a timestamp more than 15 minutes ahead of the current time on an archiver will produce a failed write. |
| System Alerts | Examine the messages and alerts and take appropriate action to correct any problems. |
On a regular schedule, examine and analyze the performance indicators displayed in the Performance section
| Field | Recommended Actions |
|---|---|
| Avg. Event Rate Chart | Is the rate at a normal level? Does the chart exhibit an acceptable trend line? If not, determine why. Balance polling schedules, adjust scan frequencies (collection intervals), and modify compression deadbands to lighten load. |
| Compression Chart | Is compression effectiveness acceptable? If not, verify that compression is enabled and then widen the deadbands to increase the effect of compression. |
| Overruns Chart | If the value is anything other than zero, determine the severity and cause of the problem and take corrective action. |