Open GoogleCodeExporter opened 8 years ago
Do you have 2 processes running on the same server reading/writing the same
index file?
Original comment by simone.chiaretta
on 22 Sep 2010 at 8:33
Not due to my settings... unless there is something about GoDaddy that can
cause two instances to run at once, but I wouldn't think so. That would make
sense for the error, but not really for my setup. It happens all the time.
Posting fails, logging in to Admin fails, and my only fix is to recycle the app
pool and hope I can delete that file fast! Until it locks again...
Original comment by ariank...@gmail.com
on 23 Sep 2010 at 1:22
Another dump... When I post from Windows Live Writer, I get the following error
so WLW think it failed, though it actually does post:
Server Error 0 Occurred
Lock obtain timed out:
NativeFSLock@D:\Hosting\6568906\html\App_Data\write.lock:
System.IO.IOException: The process cannot access the file
'D:\Hosting\6568906\html\App_Data\write.lock' because it is being used by
another process.
at System.IO.__Error.WinIOError(Int32 errorCode, String maybeFullPath)
at System.IO.FileStream.Init(String path, FileMode mode, FileAccess access, Int32 rights, Boolean useRights, FileShare share, Int32 bufferSize, FileOptions options, SECURITY_ATTRIBUTES secAttrs, String msgPath, Boolean bFromProxy)
at System.IO.FileStream..ctor(String path, FileMode mode, FileAccess access)
at Lucene.Net.Store.NativeFSLock.Obtain() at Lucene.Net.Store.Lock.Obtain(Int64 lockWaitTimeout)
at Lucene.Net.Index.IndexWriter.Init(Directory d, Analyzer a, Boolean create, Boolean closeDir, IndexDeletionPolicy deletionPolicy, Boolean autoCommit, Int32 maxFieldLength, IndexingChain indexingChain, IndexCommit commit)
at Lucene.Net.Index.IndexWriter.Init(Directory d, Analyzer a, Boolean closeDir, IndexDeletionPolicy deletionPolicy, Boolean autoCommit, Int32 maxFieldLength, IndexingChain indexingChain, IndexCommit commit)
at Lucene.Net.Index.IndexWriter..ctor(Directory d, Analyzer a, MaxFieldLength mfl)
at Subtext.Framework.Services.SearchEngine.SearchEngineService.EnsureIndexWriter()
at Subtext.Framework.Services.SearchEngine.SearchEngineService.DoWriterAction(Action`1 action)
at Subtext.Framework.Services.SearchEngine.SearchEngineService.ExecuteRemovePost(Int32 entryId)
at Subtext.Framework.Services.SearchEngine.SearchEngineService.AddPosts(IEnumerable`1 posts, Boolean optimize)
at Subtext.Framework.Services.SearchEngine.SearchEngineService.AddPost(SearchEngineEntry post)
at Subtext.Framework.Services.SearchEngine.IndexingService.ExecuteAddPost(SearchEngineEntry entry)
at Subtext.Framework.Services.SearchEngine.IndexingService.AddPost(Entry entry, IList`1 tags)
at Subtext.Framework.Services.EntryPublisher.Publish(Entry entry)
at Subtext.Framework.XmlRpc.MetaWeblog.PostContent(String username, String password, Post& post, Boolean publish, PostType postType)
Original comment by ariank...@gmail.com
on 24 Sep 2010 at 6:28
Yes, I think it's always the same problem: something is locking your write.lock
file.
Original comment by simone.chiaretta
on 24 Sep 2010 at 7:29
Any thoughts here? What would be locking it? It's a shared hosting server at
GoDaddy. I don't have any processes running -- just a directory full of files
hosted in IIS. Am I the only person to see this?
Original comment by ariank...@gmail.com
on 27 Sep 2010 at 2:43
I'm also getting this behaviour. I'm running two blogs on shared hosting. At
first I was running using two processes which always locked me out of admin for
one of the weblogs. After about a week of to/fro with support, I'm quite sure
we now have it configured as a single process. The locking error is occuring
only intermittently.
I get this trying to login as Admin:
[IOException: The process cannot access the file
'\\boswinfs04\home\Users\Web\b2432\whl.ianwi\subtext25120\App_Data\write.lock'
because it is being used by another process.]
System.IO.__Error.WinIOError(Int32 errorCode, String maybeFullPath) +7715632
System.IO.FileStream.Init(String path, FileMode mode, FileAccess access, Int32 rights, Boolean useRights, FileShare share, Int32 bufferSize, FileOptions options, SECURITY_ATTRIBUTES secAttrs, String msgPath, Boolean bFromProxy) +1162
System.IO.FileStream..ctor(String path, FileMode mode, FileAccess access) +67
Lucene.Net.Store.NativeFSLock.Obtain() +428
[LockObtainFailedException: Lock obtain timed out:
NativeFSLock@\\boswinfs04\home\Users\Web\b2432\whl.ianwi\subtext25120\App_Data\w
rite.lock: System.IO.IOException: The process cannot access the file
'\\boswinfs04\home\Users\Web\b2432\whl.ianwi\subtext25120\App_Data\write.lock'
because it is being used by another process.
at System.IO.__Error.WinIOError(Int32 errorCode, String maybeFullPath)
at System.IO.FileStream.Init(String path, FileMode mode, FileAccess access, Int32 rights, Boolean useRights, FileShare share, Int32 bufferSize, FileOptions options, SECURITY_ATTRIBUTES secAttrs, String msgPath, Boolean bFromProxy)
at System.IO.FileStream..ctor(String path, FileMode mode, FileAccess access)
at Lucene.Net.Store.NativeFSLock.Obtain()]
Lucene.Net.Store.Lock.Obtain(Int64 lockWaitTimeout) +574
Original comment by ian.wije...@gmail.com
on 4 Oct 2010 at 4:46
Sorry to make a pest of myself, but I don't know what else to do. My Subtext
log is full of the above lock error and errors from
Subtext.Framework.Services.SearchEngine.SearchEngineService stating "Something
left a lock in the index folder: deleting it." This is constant. It happens
over and over again. I called GoDaddy and they say that the application pool
keeps crashing. I don't have any custom modules running though. What do I
need to do next?
Original comment by ariank...@gmail.com
on 5 Nov 2010 at 7:05
But is your application running?
The log message you are receiving is because the search engine is trying to
recover from the locking error deleting the file.
If you don't need the search you can disable it... I think the web.config
should have the info on how do to it
Original comment by simone.chiaretta
on 6 Nov 2010 at 1:22
Not sure what you mean about the app running. I can hit the site just fine. I
get the lock exception if I try to post content and sometimes in the admin area
(won't always even let me in). GoDaddy says the app domain keeps crashing, so
presumably the lock file is left open from last time. Again and again. I'll
just disable then. Not sure why I didn't think of that. It's a shame that I
can't figure out the actual problem though.
Original comment by ariank...@gmail.com
on 6 Nov 2010 at 3:43
I checked the web.confi gand there's no reference to Lucene, but there is a
block concerning search:
<FullTextSearchEngineSettings
type="Subtext.Framework.Configuration.FullTextSearchEngineSettings,
Subtext.Framework">
<Parameters>
<MinimumDocumentFrequency>10</MinimumDocumentFrequency>
</Parameters>
</FullTextSearchEngineSettings>
The admin tool just gives me an option to rebuild the index. What do I do to
actually disable it?
Original comment by ariank...@gmail.com
on 6 Nov 2010 at 6:54
I have the same problem -- I'm hosting on my own IIS 7 server, just one domain,
but with multiple blog support. I do seem to remember some option for full text
support when I installed this the first time, but now all I see is what the
previous poster mentioned.
Original comment by tpuj...@gmail.com
on 22 Nov 2010 at 11:40
Did anyone ever figure out how to disable this search feature to get rid of the
locking problem?
Original comment by ian.wije...@gmail.com
on 27 Nov 2010 at 1:53
Sorry, forgot to check:
just add IsEnabled="False" in the FullTextSearchEngineSettings tag
Simone
Original comment by simone.chiaretta
on 27 Nov 2010 at 5:38
Original comment by simone.chiaretta
on 30 Nov 2010 at 8:49
I logged a second error about this and just received this error again.
Original comment by billgraz...@gmail.com
on 19 Feb 2011 at 1:02
Haven't been able to reproduce this. Can you try deploying the .pdb files?
Original comment by haac...@gmail.com
on 16 May 2011 at 3:50
Issue 296 has been merged into this issue.
Original comment by haac...@gmail.com
on 16 May 2011 at 4:12
I no longer use Subtext. I'm using Orchard now. It appears that this was
caused by GoDaddy (and likely other shared hosts) using pools of worker threads
which would each attempt to lock the same index file. I'm not sure what a good
fix for this would be, but I was never able to resolve it.
Original comment by ariank...@gmail.com
on 16 May 2011 at 5:36
Original issue reported on code.google.com by
ariank...@gmail.com
on 22 Sep 2010 at 4:09