It sounds like you should first run CheckIndex to see if there's corruption if both indexes... and then run again with -fix to repair the corruption.
That repair simply removes any segments with corruption. So after that you'll have to re-index whatever docs are missing... Mike On Thu, Mar 18, 2010 at 7:45 PM, Andrew Bruno <[email protected]> wrote: > Hi all, > > I am using Compass 1.1 with Lucene 2 > > Our product offers a 2 lucene sub-indexes per customer, A and B, where > B extends A. > In size, A & B are about 15G, where A =10G > > We had a server failure a few days ago related to corrupt RAM, and > when the server came back up, and services restarted OK (or so it > appears from logs). > > A few days later we have uncovered that index A does not have any > documents indexed prior to server failure, only new documents after > server failure. > Index B is OK. > > Index A has millions of documents indexed against it. Is there a way > that I could try and repair it, even offline? > > Thanks > Andrew > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [email protected] > For additional commands, e-mail: [email protected] > > --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
