User Tools

Site Tools


faq:pervasive_troubleshooting

This is an old revision of the document!


Pervasive Crash Troubleshooting

Use this document to give you ideas on where to look for when experiencing a crashing Pervasive SQL service. This article is written expecting you know how to run the PCC (Pervasive Control Center) and have a technical background in System Five. If you are unsure what to do, contact Windward Software customer care.

Questions to Ask

  • What is the plan to restore from backup?
    • It can take time to rebuild a new virtual machine and to restore from backup media. Start this while doing any further trouble shooting.
    • Restoring from backup gets the business back into operation. You can trouble shoot a database issue in training or on another server entirely.
  • What has changed?
  • Groom the logs and ask questions to understand service restarts or messages.
  • What date and time was the last known good server state?
  • What date and time was the last known good backup taken (with the PSQL service turned off)?

Known Pervasive Service Crashing Culprits

If the Peravsive SQL database service is crashing here are some things to check.

  1. Unpack a fresh set of DDF's as a corrupt view can cause the PSQL engine to crash without an error in logs
  2. Checking the PVSW Logs. PCC and Help menu. (This can give you an idea of why the service or what file or application is causing an issue)
  3. Is the Pervasive database version patched and up to date. (Verify service pack and version)
  4. If receiving a “Roll forward recovery” or “The log file is corrupt” message in the logs you could try to clear out the Transaction Log
  5. A corrupt database can cause a crash. Restore from backup.
  6. A operating system patch or pending operating system patch can cause odd behavior. (Reboot; apply updates; restore from snapshot or last known good server state)

Trouble Shooting

  • Test network transmission of larger packet size as corrupt packets of data can cause failure
    • At command prompt ping -t -s2000 (Is there packet loss?)
  • Post and search the symptoms on Actian Communities pages.
  • Capture the PVSW log file to be able to show other technicians.

  • Check the server event log for messages that may indicate what the problem is.
  • Using the PCC you can go table by table and view by view to find the offending file that will crash the database service.
  • Consider stopping the relational service which will stop SQL support. This may help isolate the issue to the application causing the problem.
faq/pervasive_troubleshooting.1576279940.txt.gz · Last modified: 2019/12/13 15:32 (4 years ago) by kevin