We have seen the following error repeatedly in our 'event log':
'InnerException:Lock obtain timed out:
Here is some background on our server setup
- Four servers, each with the same DNN install, behind a load balancer
- Our main Server has 'File Replication' set on it to copy files to the same location in the 'mirror' sites.
- Deleting the Lucene/index files located at \dnn\App_Data\Search
- Reindexing the site using '/Admin/Search-Admin'
- Setting 'Schedule enabled', then clicking 'run now' for the 'Search: Site Crawler' task in 'Host > Schedule'
However, this does not cause the lucene files to be updated in '\dnn\App_Data\Search' and the event log fills with these three errors:
- 'Scheduler Event Failure':Lock obtain timed out: [email protected]:\inetpub\wwwroot\live\App_Data\Search\write.lock
- 'Scheduler Exception': Lock obtain timed out: [email protected]:\inetpub\wwwroot\live\App_Data\Search\write.lock
- 'General Exception': Lock obtain timed out:
From looking at this error on the 'Lucene.Net' site, it appears that this error is telling us that 'this exception is thrown when write.lock could nt be acquired. This happens when a writer tries to open an index that another writer already has open.'
In the 'Host > Scheduler' we have set up the 'Search Site Crawler' as a task only on the first server, with file replication copying the index files to the '\dnn\App_Data\Search' directories of the other servers after a reindex.
This is not working and we're at a loss as to how to fix this. All settings in DNN appear to be correct.
Can you offer a solution to this problem?