Failure to parse redo should try restarting the task
Currently, for an Oracle source if Replicate has a parsing failure it appears to immediately try to read it a second time. If it fails again the task will just stop attempting to read any redo. The Parsing Failures are considered warnings rather than errors and no additional error or warnings will be sent. I have a separate Idea posted about treating these Parsing Failure as errors rather than warnings.
QLIK will continue periodically writing latency values to the log that are randomly spread around the latency that was seen the last time it read redo. So, over time the reported latencies become extremely erroneous, i.e. a few seconds when the real latency is many hours.
QLIK should instead restart the task and retry reading the redo. In our case we have always been able to read the redo from Archive Log when we finally see the problem. I wonder if it might be due to the read attempt happening very soon after the online redo is switched and perhaps nothing has been written to the new redo yet. If so, a time delay on the restart might be necessary.
NOTE: Upon clicking this link 2 tabs may open - please feel free to close the one with a login page. If you only see 1 tab with the login page, please try clicking this link first: Authenticate me! then try the link above again. Ensure pop-up blocker is off.
