Has anyone ever experienced an empty Internal Index set and a forever queued rebuild of the Internal Index? Even after removing the set completely, restarting the application and forcing a rebuild, the index rebuild never starts. I've checked the Umbraco logs for any messages regarding errors or rebuilds and nothing seems to get logged out. Here's my current config:
Looking at your indexpath it says san-data i take it this is in load balanced environment with a san? If so then could it be permissions problem domain account which needs to be on all boxes in the cluster? Also what version of umbraco are you using?
Thanks for your reply Ismail. You are correct. The Indexes are living on a SAN. I'm fairly certain permissions are correct as deleting the Internal index folder results in it being recreated when the application restarts. The index does eventually become unqueued but only contains 2 documents compared to Umbraco which contains a couple of thousand content pages. Its actually quite an old version of Umbraco, version 4.7.2 to be precise.
Is the umbraco.config xml cache file shared on san or does each site have its own? If they have own are the sizes the same? It may be if they are different then one is using smaller file?
Are you doing anything with gatheringnode data on that index? Also how much content have you got? I recall someone else having issues with a large site / xml config file.
So after a little more further investigation, it turns out that the issue is related to content. For example, the same code base and configuration works with the site's stage database but not with the site's production database.
Empty Internal Index set and forever queued
Has anyone ever experienced an empty Internal Index set and a forever queued rebuild of the Internal Index? Even after removing the set completely, restarting the application and forcing a rebuild, the index rebuild never starts. I've checked the Umbraco logs for any messages regarding errors or rebuilds and nothing seems to get logged out. Here's my current config:
<?xml version="1.0"?> <ExamineLuceneIndexSets> <IndexSet SetName="InternalIndexSet" IndexPath="~/App_Data/SAN-DATA/ExamineIndexes/Internal/"> <IndexAttributeFields> <add Name="id" /> <add Name="nodeName" /> <add Name="updateDate" /> <add Name="writerName" /> <add Name="path" /> <add Name="nodeTypeAlias" /> <add Name="parentID" /> </IndexAttributeFields> <IndexUserFields /> <IncludeNodeTypes/> <ExcludeNodeTypes /> </IndexSet> </ExamineLuceneIndexSets><?xml version="1.0"?> <Examine> <ExamineIndexProviders> <providers> <add name="InternalIndexer" type="UmbracoExamine.UmbracoContentIndexer, UmbracoExamine" supportUnpublished="true" supportProtected="true" interval="10" analyzer="Lucene.Net.Analysis.WhitespaceAnalyzer, Lucene.Net"/> </providers> </ExamineIndexProviders> <ExamineSearchProviders defaultProvider="InternalSearcher"> <providers> <add name="InternalSearcher" type="UmbracoExamine.UmbracoExamineSearcher, UmbracoExamine" analyzer="Lucene.Net.Analysis.WhitespaceAnalyzer, Lucene.Net"/> </providers> </ExamineSearchProviders> </Examine>Dan,
Looking at your indexpath it says san-data i take it this is in load balanced environment with a san? If so then could it be permissions problem domain account which needs to be on all boxes in the cluster? Also what version of umbraco are you using?
Regards
Ismail
Thanks for your reply Ismail. You are correct. The Indexes are living on a SAN. I'm fairly certain permissions are correct as deleting the Internal index folder results in it being recreated when the application restarts. The index does eventually become unqueued but only contains 2 documents compared to Umbraco which contains a couple of thousand content pages. Its actually quite an old version of Umbraco, version 4.7.2 to be precise.
Dan,
Is the umbraco.config xml cache file shared on san or does each site have its own? If they have own are the sizes the same? It may be if they are different then one is using smaller file?
Regards
Ismail
The umbraco.config cache file is shared on the same SAN and currently has around 18MB's worth of content in it. Thanks Ismail.
Dan,
Are you doing anything with gatheringnode data on that index? Also how much content have you got? I recall someone else having issues with a large site / xml config file.
Regards
Ismail
Hi Ismail,
Not that I know of. I believe there are at least 10,000+ nodes with Umbraco so maybe that has something to do with the index not being rebuilt.
Thanks,
Dan.
So after a little more further investigation, it turns out that the issue is related to content. For example, the same code base and configuration works with the site's stage database but not with the site's production database.
is working on a reply...