Enabling logging in UndoDB

It is possible to tell UndoDB to generate extra logging messages which help us better to understand any issues that you find.  Collecting this data before sending us the issue can help us to resolve your issues more quickly.  Enabling additional logging can be done in a number of ways:

  1. Start UndoDB with the UNDODB_debug_level_internal environment variable set to 2.  If UndoDB is normally started with a command like undodb-gdb <application>, you would instead use UNDODB_debug_level_internal=2 undodb-gdb <application>.
  2. Logging can also be enabled during runtime. At the (undodb-gdb) prompt, run maint-undodb-set-log-levels -1 2. This can only be done after the debug session has started.

These options cause additional information to be logged internally, which can then be dumped to files with the maint-undodb-dump-internal-log command.  The log files will have names like undodb_log.<pid>.

There are a few other commands that can provide useful diagnostics and are typed at the (undodb-gdb) prompt:

  • info proc mappings
    • Lists all the memory maps in use, including any maps created by UndoDB for its own use.
  • maint-undodb-show-exec-summary
    • Show some statistics relating to UndoDB's code analysis.
  • maint-undodb-show-snapshots
    • Shows all active snapshots.

If you can send us the log files and the output of these commands, it will help us to identify or rule out any issues with UndoDB.

Have more questions? Submit a request

0 Comments

Please sign in to leave a comment.
Powered by Zendesk