Based on continued research, we seem to have fixed the problem. For us, the clue was "Resource temporarily unavailable, will not index. Try --rotate option." in our indexer log file.
The solution: I renamed the ".new" index files to replace the older index files, and stopped and restarted searchd. Then, I ran the indexer directly with the --all and --rotate options. (Pointing to the TS generated config file, of course.) Everything appears to be connected again. Now, I've also updated our cron task to run the indexer directly as well. Hopefully that solves it and/or helps someone else. On Monday, April 14, 2014 8:45:40 AM UTC-5, Garrett Dimon wrote: > > This may be a Sphinx issue, but since we're not touching Sphinx directly, > I wanted to check here first. > > We recently updated Sphinx on our journey to updating TS and ultimately > using RT indexing. We currently running delayed deltas with 'rake ts:in' > via cron every 12 hours. > > We've done this for years, but lately, seemingly around the 2.1.6 upgrade, > it seems that while the core.new indexes are being created, they aren't > being rotated into service any longer. > > Looking at the searchd logs, I'm seeing "WARNING: nothing to rotate after > SIGHUP ( in queue=0 )". > > Are there any common mistakes that might lead to this or is it possible > that 2.1.6 introduced a change to how indexes are rotated? > > Thanks, > Garrett > -- You received this message because you are subscribed to the Google Groups "Thinking Sphinx" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/thinking-sphinx. For more options, visit https://groups.google.com/d/optout.
