Recently, I switched from using the default column based delta
approach for my model indexing and instead began using the timestamp
based delta. In the development environment, everything seemed fine,
but when we moved to production, we started experiencing exceptions
anytime a search was timed to coincide with when the indexer finished
a delta index run and sent SIGHUP to searchd.

This search timing would trigger a Riddle::ResponseError with the
message: "No response from searchd (status: , version: )".

I tried to get around this error by catching this particular exception
and retrying the search, but then I received a
ThinkingSphinx::ConnectionError complaining that searchd could not be
found. Even when I caught this exception as well and allowed 5 retry
attempts, each time after the first ResponseError, I would receive a
ConnectionError.

However, if did not attempt to catch the error and simply refreshed
the page after triggering an exception as described above, the
refreshed page would work fine.

Is this a known issue that the Sphinx searchd briefly becomes
unavailable during the final stage of reindexing? And is there a way
around this problem?

I'm using Sphinx 0.9.8.1-release (r1533) for PostgreSQL 8.3.6 with the
ThinkingSphinx plugin from Github as of 2009-03-11.

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"Thinking Sphinx" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to 
[email protected]
For more options, visit this group at 
http://groups.google.com/group/thinking-sphinx?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to