I'm noticing some data differences between my database and Solr. About a week ago my Solr server ran out of disk space, so now I'm observing how the DataImportHandler behaves when Solr runs out of disk space. In a word, I'd say it behaves badly! It looks like out-of-disk-space exceptions are treated like any other document-level exception (so my updates report successful completion). After running out of disk space I see the index shrink, then updates continue until I run out of disk space again, then the index shrinks, etc. I'm running a nightly build from December, 2009. Has this behaviour has changed since then? Is there DIH configuration to fail on certain exceptions?
Thanks, Wojtek -- View this message in context: http://lucene.472066.n3.nabble.com/DataImportHandler-and-running-out-of-disk-space-tp835125p835125.html Sent from the Solr - User mailing list archive at Nabble.com.
