Forums

Hi,

I had the indexing working well on dotnetnuke.worrells.net.au. I shifted the license across to www.worrells.net.au as we went live this morning with our new site.

I am now receiving errors when forcing a run of the indexing scheduled task. There are still permission problems trying to upload files so I will place "log.txt" file contents below

__________________________________________________________________________________________________________
2/1/2012 11:56:51 AM: psSiteSearch indexing site http://www.worrells.net.au
2/1/2012 11:56:51 AM: Directories to index for PortalID 0: 0
2/1/2012 11:56:51 AM: BuildDocumentList failed 0: http://
2/1/2012 11:56:51 AM: Crawl() exception caught. http://www.worrells.net.au:System.NullReferenceException: Object reference not set to an instance of an object.
   at PureSystems.Modules.psSiteSearch.psSiteIndex.Crawl(String strSiteURL, Int32 intEnableWinAuth, String strWinDomain, String strWinUsername, String strWinPassword, String strDNNUser, String strDefaultDoc, Int32 intDebug, String strTempPath, Int32 intIgnoreLinkText, Int32 _intPortalID, String _strPortalAlias, Int32 _siteID, String _strCanonicalLinkRegEx)
   at PureSystems.Modules.psSiteSearch.psSiteIndex.StartCrawl()
__________________________________________________________________________________________________________ 

I wonder whether the BuildDocumentList failed task and the empty "http://" after it are a clue to what is going wrong

Thanks,
Angus.

Hi,

I have tried that. I even deleted the site that was present in psSiteSearch admin and re-created it all over again.

I got the same error message.

Thanks,
Angus.

It turns out that the problem was a link on the front page which was incorrect. The link was to www..worrells.net.au (note the two full stops or periods after the www). This broke indexing completely.

After fixing the link, indexing started working successfully so it looks like an invalid domain crashes the indexing completely.

Thanks,
Angus.