Hi Mark,
Thanks for the reply. The issue happened because the server kill/restart
process failed which meant the old instance was still running, and it had
the data related to old files loaded into memory. And when commit was
issued, it was trying to locate those old files and failing. Once this w
How are you doing the backup? You have to coordinate with Solr - files may be
changing when you try and copy it, leaving to an inconsistent index. If you
want to do a live backup, you have to use the backup feature of the replication
handler.
- Mark
On Feb 23, 2013, at 3:54 AM, Prakhar Birla
