0911 - Dynamic Database is Inconsistent

The dynamic database (DDB) audit (whether running automatically or on-demand) has detected that checksums are inconsistent. This means that one or more cards do not concur with the current network configuration. (The UIM lists a maximum of four affected cards.)

The dynamic database audit looks for checksum errors in dynamic database tables on LIM and Service Module cards. The fixed database, which is entered via the OAM card, includes all provisionable tables and options. The dynamic database has information about the state of those static-database entries (for example, which links are actually available). Each LIM and Service Module card has a copy of the fixed database and the dynamic database. The entries in the tables in the dynamic database on each card change as network conditions change. Cards calculate and record dynamic database checksums in real time as updates are applied.

If the dynamic database is being updated during the dynamic database audit, then the inconsistent checksums might not indicate a true problem. To ensure that all dynamic database changes have been received and applied, the OAM processes audit replies after a required quiet period.

Note:

You can use the “DDBAUDTIMER” SS7OPTS option to enable and disable the background audit.

Typically, when the dynamic database copies are consistent among all the cards except one, then there is a problem on that one card. It is possible, however, for one card to have the correct information when the other cards do not (for example, a card might have a link on it that is actually available when the other cards incorrectly show the link as unavailable). When this alarm occurs, further troubleshooting is necessary to determine the actual status of the routes, links, linksets, and subsystems.

Example

RLGHNCXA21W 00-02-07 11:02:30 EST  EAGLE 41.0
**  0100.0911 ** SYSTEM                  Dynamic database is inconsistent
                 Card 1101, 1102, 1107, 1108, ... (3 others)

Alarm Level: Major

Recovery

The Eagle cannot automatically determine which cards are inconsistent. Do the following to address a dynamic-database inconsistency alarm:

1. Run the rept-stat-ddb command to display the last dynamic database audit report.
2. Initialize the affected cards identified in the dynamic database audit report:
   1. If a small number of cards are reported inconsistent, initialize them using the init-card:loc= command. You should initialize the affected cards immediately.
   2. If all cards of the same type report inconsistencies, initialize them using the init-card:appl=command.

      Caution:

      This command could affect service as it will boot all the appl cards at the same time (for example, the command could cause loss of SCCP if the init-card:appl=VSCCPcommand is run).
3. Run the rept-stat-card:mode=full:loc= command on each card identified as inconsistent to verify the card is IS-NR.
4. If a large number of cards of various types report inconsistent, you might need to issue the init-network or the init-sys command to clear the inconsistency.

   Caution:

   These commandswill cause a nodal outage.
5. After the cards are initialized, run the aud-data:type=DDB command. If the report shows no inconsistency, then the problem is fixed.
6. Next, you can use the dbg-ddb command to troubleshoot the problem. The parameters to use in the dbg-ddb command depend on the type of dynamic database update that was missed (route/link/linkset).
   When a system reports dynamic database inconsistencies, first check the IMT buses statistics to make sure they are clean. This is important because multicast updates resulting from network activity transit via the IMT buses and any outstanding issue on this part of the system may lead some cards to miss the updates and OAM to report dynamic database inconsistencies. When a checksum is identified to be incorrect and is updated by a wild write audit, you may want to know that a dynamic database inconsistency reported on a card was due to a wild write (rather than any other cause, such as a missed multicast). In the scenario where one or a group of cards miss a dynamic database update related to a network state change, then the counter collecting the total number of update misses is incremented. Use the dbg-ddb command to retrieve this counter.
7. Contact My Oracle Support (MOS).