You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@lucenenet.apache.org by Simone Chiaretta <si...@gmail.com> on 2010/01/04 22:16:44 UTC
"Lock obtain timed out: SimpleFSLock" appearing sometimes
Hi all,
since it's my first time here I'd like to introduce myself.
My name is Simone Chiaretta, and I'm member of Subtext blogging engine team,
and I'm a ASP.NET MVC fan.
I also have a blog at http://codeclimber.net.nz, where I also wrote a few
posts about Lucene.net
But going back to the main topic of this email: I'm implementing a
Lucene.net based search engine for Subtext and I'm writing because I've a
strange problem that happens from times to times:
I get the exception of the subject "Lock obtain timed out: SimpleFSLock..."
every now and then when the ApplicationPool recycles.
I know only one IndexWriter can exist at a time, and in fact the writer is a
static variable inside the search engine service.
The problem seems that the Dispose of the search engine, which closes the
writer is not enough for releasing the lock.
public void Dispose()
{
var searcher = _searcher;
if (searcher != null)
{
searcher.Close();
}
var writer = _writer;
if (writer != null)
{
writer.Close();
}
var directory = _directory;
if(directory != null) {
directory.Close();
}
writer = _writer;
if (writer != null)
{
_writer = null;
}
}
For the moment I made 2 workaround that I hope will solve the problem:
I'm checking weather the index is locked or not before creating a new one
if(_writer==null)
{
if(IndexReader.IsLocked(_directory))
{
__log.Error("Something left a lock in the index folder: deleting
it");
IndexReader.Unlock(_directory);
__log.Info("Lock Deleted... can proceed");
}
_writer = new IndexWriter(_directory, _analyzer);
_writer.Optimize();
}
and I'm explicitly disposing the search engine in the Application_End event
of the WebApplication
searchEngine.Dispose();
But I'd like to know if maybe I'm making some mistake somewhere that is
causing the problem, or if there is a more elegant way to fix this problem.
For a complete reference here is the link to the latest version of the
SearchEngineService class that encapsulate all the fulltext searching logic
of Subtext
http://code.google.com/p/subtext/source/browse/trunk/src/Subtext.Framework/Services/SearchEngine/SearchEngineService.cs
Oh... one last thing.. I'm using Lucene.net 2.3.2
Thank you
Simone
--
Simone Chiaretta
Microsoft MVP ASP.NET - ASPInsider
Blog: http://codeclimber.net.nz
RSS: http://feeds2.feedburner.com/codeclimber
twitter: @simonech
Any sufficiently advanced technology is indistinguishable from magic
"Life is short, play hard"
Re: "Lock obtain timed out: SimpleFSLock" appearing sometimes
Posted by Simone Chiaretta <si...@gmail.com>.
Yes... it appears to have disappeared.
I was expecting that the Dispose method of a IDisposable would have been
called anyway by the garbage collector when the app pool was recycled.
I might add a finalizer just to play on the safe side :)
Thx
Simone
On Wed, Jan 6, 2010 at 4:59 PM, Franklin Simmons <
fsimmons@sccmediaserver.com> wrote:
> Simone,
>
> Are you still experiencing locks after adding the call to Dispose? I ask
> because I see as part of your workaround that you are "explicitly disposing
> the search engine in the Application_End event". This isn't so much a
> workaround as a correction - it is correct to call Dispose directly. It
> sounds like an expectation exists that the Dispose method would be invoked
> otherwise? To my knowledge aside from using statements, the only other way
> Dispose might be invoked implicitly is via a class finalizer. You should
> consider adding a finalizer that invokes Dispose as a safeguard (not a
> solution) to ensure Lucene resources are released.
>
>
> -----Original Message-----
> From: Simone Chiaretta [mailto:simone.chiaretta@gmail.com]
> Sent: Monday, January 04, 2010 7:02 PM
> To: lucene-net-user@lucene.apache.org
> Subject: Re: "Lock obtain timed out: SimpleFSLock" appearing sometimes
>
> Thank you.
> I'm not doing any optimize other then the one at startup and after adding a
> new post.
> And I'm 100% sure nobody added a post between startup and when, after a few
> hours, the apppool was restarted.
> Simo
>
> On Tue, Jan 5, 2010 at 12:03 AM, Franklin Simmons <
> fsimmons@sccmediaserver.com> wrote:
>
> > Simone,
> >
> > Is it possible that an optimize is exceeding any of the application
> pool's
> > time out settings? Optimize can be a lengthy process (for large
> indexes).
> > I am not intimate with the behavior of IIS's process termination, but
> > summarily killing a process in the midst of an optimize can be
> problematic.
> >
> >
> > -----Original Message-----
> > From: Simone Chiaretta [mailto:simone.chiaretta@gmail.com]
> > Sent: Monday, January 04, 2010 4:17 PM
> > To: lucene-net-user
> > Subject: "Lock obtain timed out: SimpleFSLock" appearing sometimes
> >
> > Hi all,
> > since it's my first time here I'd like to introduce myself.
> > My name is Simone Chiaretta, and I'm member of Subtext blogging engine
> > team,
> > and I'm a ASP.NET MVC fan.
> > I also have a blog at http://codeclimber.net.nz, where I also wrote a
> few
> > posts about Lucene.net
> >
> > But going back to the main topic of this email: I'm implementing a
> > Lucene.net based search engine for Subtext and I'm writing because I've a
> > strange problem that happens from times to times:
> > I get the exception of the subject "Lock obtain timed out:
> SimpleFSLock..."
> > every now and then when the ApplicationPool recycles.
> >
> > I know only one IndexWriter can exist at a time, and in fact the writer
> is
> > a
> > static variable inside the search engine service.
> >
> > The problem seems that the Dispose of the search engine, which closes the
> > writer is not enough for releasing the lock.
> >
> > public void Dispose()
> > {
> > var searcher = _searcher;
> > if (searcher != null)
> > {
> > searcher.Close();
> > }
> >
> > var writer = _writer;
> > if (writer != null)
> > {
> > writer.Close();
> > }
> >
> > var directory = _directory;
> > if(directory != null) {
> > directory.Close();
> > }
> >
> > writer = _writer;
> > if (writer != null)
> > {
> > _writer = null;
> > }
> > }
> >
> > For the moment I made 2 workaround that I hope will solve the problem:
> >
> > I'm checking weather the index is locked or not before creating a new one
> >
> > if(_writer==null)
> > {
> > if(IndexReader.IsLocked(_directory))
> > {
> > __log.Error("Something left a lock in the index folder: deleting
> > it");
> > IndexReader.Unlock(_directory);
> > __log.Info("Lock Deleted... can proceed");
> > }
> > _writer = new IndexWriter(_directory, _analyzer);
> > _writer.Optimize();
> > }
> >
> > and I'm explicitly disposing the search engine in the Application_End
> event
> > of the WebApplication
> >
> > searchEngine.Dispose();
> >
> > But I'd like to know if maybe I'm making some mistake somewhere that is
> > causing the problem, or if there is a more elegant way to fix this
> problem.
> >
> > For a complete reference here is the link to the latest version of the
> > SearchEngineService class that encapsulate all the fulltext searching
> logic
> > of Subtext
> >
> >
> http://code.google.com/p/subtext/source/browse/trunk/src/Subtext.Framework/Services/SearchEngine/SearchEngineService.cs
> >
> > Oh... one last thing.. I'm using Lucene.net 2.3.2
> >
> > Thank you
> > Simone
> >
> >
> > --
> > Simone Chiaretta
> > Microsoft MVP ASP.NET - ASPInsider
> > Blog: http://codeclimber.net.nz
> > RSS: http://feeds2.feedburner.com/codeclimber
> > twitter: @simonech
> >
> > Any sufficiently advanced technology is indistinguishable from magic
> > "Life is short, play hard"
> >
>
>
>
> --
> Simone Chiaretta
> Microsoft MVP ASP.NET - ASPInsider
> Blog: http://codeclimber.net.nz
> RSS: http://feeds2.feedburner.com/codeclimber
> twitter: @simonech
>
> Any sufficiently advanced technology is indistinguishable from magic
> "Life is short, play hard"
>
--
Simone Chiaretta
Microsoft MVP ASP.NET - ASPInsider
Blog: http://codeclimber.net.nz
RSS: http://feeds2.feedburner.com/codeclimber
twitter: @simonech
Any sufficiently advanced technology is indistinguishable from magic
"Life is short, play hard"
RE: "Lock obtain timed out: SimpleFSLock" appearing sometimes
Posted by Franklin Simmons <fs...@sccmediaserver.com>.
Simone,
Are you still experiencing locks after adding the call to Dispose? I ask because I see as part of your workaround that you are "explicitly disposing the search engine in the Application_End event". This isn't so much a workaround as a correction - it is correct to call Dispose directly. It sounds like an expectation exists that the Dispose method would be invoked otherwise? To my knowledge aside from using statements, the only other way Dispose might be invoked implicitly is via a class finalizer. You should consider adding a finalizer that invokes Dispose as a safeguard (not a solution) to ensure Lucene resources are released.
-----Original Message-----
From: Simone Chiaretta [mailto:simone.chiaretta@gmail.com]
Sent: Monday, January 04, 2010 7:02 PM
To: lucene-net-user@lucene.apache.org
Subject: Re: "Lock obtain timed out: SimpleFSLock" appearing sometimes
Thank you.
I'm not doing any optimize other then the one at startup and after adding a
new post.
And I'm 100% sure nobody added a post between startup and when, after a few
hours, the apppool was restarted.
Simo
On Tue, Jan 5, 2010 at 12:03 AM, Franklin Simmons <
fsimmons@sccmediaserver.com> wrote:
> Simone,
>
> Is it possible that an optimize is exceeding any of the application pool's
> time out settings? Optimize can be a lengthy process (for large indexes).
> I am not intimate with the behavior of IIS's process termination, but
> summarily killing a process in the midst of an optimize can be problematic.
>
>
> -----Original Message-----
> From: Simone Chiaretta [mailto:simone.chiaretta@gmail.com]
> Sent: Monday, January 04, 2010 4:17 PM
> To: lucene-net-user
> Subject: "Lock obtain timed out: SimpleFSLock" appearing sometimes
>
> Hi all,
> since it's my first time here I'd like to introduce myself.
> My name is Simone Chiaretta, and I'm member of Subtext blogging engine
> team,
> and I'm a ASP.NET MVC fan.
> I also have a blog at http://codeclimber.net.nz, where I also wrote a few
> posts about Lucene.net
>
> But going back to the main topic of this email: I'm implementing a
> Lucene.net based search engine for Subtext and I'm writing because I've a
> strange problem that happens from times to times:
> I get the exception of the subject "Lock obtain timed out: SimpleFSLock..."
> every now and then when the ApplicationPool recycles.
>
> I know only one IndexWriter can exist at a time, and in fact the writer is
> a
> static variable inside the search engine service.
>
> The problem seems that the Dispose of the search engine, which closes the
> writer is not enough for releasing the lock.
>
> public void Dispose()
> {
> var searcher = _searcher;
> if (searcher != null)
> {
> searcher.Close();
> }
>
> var writer = _writer;
> if (writer != null)
> {
> writer.Close();
> }
>
> var directory = _directory;
> if(directory != null) {
> directory.Close();
> }
>
> writer = _writer;
> if (writer != null)
> {
> _writer = null;
> }
> }
>
> For the moment I made 2 workaround that I hope will solve the problem:
>
> I'm checking weather the index is locked or not before creating a new one
>
> if(_writer==null)
> {
> if(IndexReader.IsLocked(_directory))
> {
> __log.Error("Something left a lock in the index folder: deleting
> it");
> IndexReader.Unlock(_directory);
> __log.Info("Lock Deleted... can proceed");
> }
> _writer = new IndexWriter(_directory, _analyzer);
> _writer.Optimize();
> }
>
> and I'm explicitly disposing the search engine in the Application_End event
> of the WebApplication
>
> searchEngine.Dispose();
>
> But I'd like to know if maybe I'm making some mistake somewhere that is
> causing the problem, or if there is a more elegant way to fix this problem.
>
> For a complete reference here is the link to the latest version of the
> SearchEngineService class that encapsulate all the fulltext searching logic
> of Subtext
>
> http://code.google.com/p/subtext/source/browse/trunk/src/Subtext.Framework/Services/SearchEngine/SearchEngineService.cs
>
> Oh... one last thing.. I'm using Lucene.net 2.3.2
>
> Thank you
> Simone
>
>
> --
> Simone Chiaretta
> Microsoft MVP ASP.NET - ASPInsider
> Blog: http://codeclimber.net.nz
> RSS: http://feeds2.feedburner.com/codeclimber
> twitter: @simonech
>
> Any sufficiently advanced technology is indistinguishable from magic
> "Life is short, play hard"
>
--
Simone Chiaretta
Microsoft MVP ASP.NET - ASPInsider
Blog: http://codeclimber.net.nz
RSS: http://feeds2.feedburner.com/codeclimber
twitter: @simonech
Any sufficiently advanced technology is indistinguishable from magic
"Life is short, play hard"
Re: "Lock obtain timed out: SimpleFSLock" appearing sometimes
Posted by Simone Chiaretta <si...@gmail.com>.
Thank you.
I'm not doing any optimize other then the one at startup and after adding a
new post.
And I'm 100% sure nobody added a post between startup and when, after a few
hours, the apppool was restarted.
Simo
On Tue, Jan 5, 2010 at 12:03 AM, Franklin Simmons <
fsimmons@sccmediaserver.com> wrote:
> Simone,
>
> Is it possible that an optimize is exceeding any of the application pool's
> time out settings? Optimize can be a lengthy process (for large indexes).
> I am not intimate with the behavior of IIS's process termination, but
> summarily killing a process in the midst of an optimize can be problematic.
>
>
> -----Original Message-----
> From: Simone Chiaretta [mailto:simone.chiaretta@gmail.com]
> Sent: Monday, January 04, 2010 4:17 PM
> To: lucene-net-user
> Subject: "Lock obtain timed out: SimpleFSLock" appearing sometimes
>
> Hi all,
> since it's my first time here I'd like to introduce myself.
> My name is Simone Chiaretta, and I'm member of Subtext blogging engine
> team,
> and I'm a ASP.NET MVC fan.
> I also have a blog at http://codeclimber.net.nz, where I also wrote a few
> posts about Lucene.net
>
> But going back to the main topic of this email: I'm implementing a
> Lucene.net based search engine for Subtext and I'm writing because I've a
> strange problem that happens from times to times:
> I get the exception of the subject "Lock obtain timed out: SimpleFSLock..."
> every now and then when the ApplicationPool recycles.
>
> I know only one IndexWriter can exist at a time, and in fact the writer is
> a
> static variable inside the search engine service.
>
> The problem seems that the Dispose of the search engine, which closes the
> writer is not enough for releasing the lock.
>
> public void Dispose()
> {
> var searcher = _searcher;
> if (searcher != null)
> {
> searcher.Close();
> }
>
> var writer = _writer;
> if (writer != null)
> {
> writer.Close();
> }
>
> var directory = _directory;
> if(directory != null) {
> directory.Close();
> }
>
> writer = _writer;
> if (writer != null)
> {
> _writer = null;
> }
> }
>
> For the moment I made 2 workaround that I hope will solve the problem:
>
> I'm checking weather the index is locked or not before creating a new one
>
> if(_writer==null)
> {
> if(IndexReader.IsLocked(_directory))
> {
> __log.Error("Something left a lock in the index folder: deleting
> it");
> IndexReader.Unlock(_directory);
> __log.Info("Lock Deleted... can proceed");
> }
> _writer = new IndexWriter(_directory, _analyzer);
> _writer.Optimize();
> }
>
> and I'm explicitly disposing the search engine in the Application_End event
> of the WebApplication
>
> searchEngine.Dispose();
>
> But I'd like to know if maybe I'm making some mistake somewhere that is
> causing the problem, or if there is a more elegant way to fix this problem.
>
> For a complete reference here is the link to the latest version of the
> SearchEngineService class that encapsulate all the fulltext searching logic
> of Subtext
>
> http://code.google.com/p/subtext/source/browse/trunk/src/Subtext.Framework/Services/SearchEngine/SearchEngineService.cs
>
> Oh... one last thing.. I'm using Lucene.net 2.3.2
>
> Thank you
> Simone
>
>
> --
> Simone Chiaretta
> Microsoft MVP ASP.NET - ASPInsider
> Blog: http://codeclimber.net.nz
> RSS: http://feeds2.feedburner.com/codeclimber
> twitter: @simonech
>
> Any sufficiently advanced technology is indistinguishable from magic
> "Life is short, play hard"
>
--
Simone Chiaretta
Microsoft MVP ASP.NET - ASPInsider
Blog: http://codeclimber.net.nz
RSS: http://feeds2.feedburner.com/codeclimber
twitter: @simonech
Any sufficiently advanced technology is indistinguishable from magic
"Life is short, play hard"
RE: "Lock obtain timed out: SimpleFSLock" appearing sometimes
Posted by Franklin Simmons <fs...@sccmediaserver.com>.
Simone,
Is it possible that an optimize is exceeding any of the application pool's time out settings? Optimize can be a lengthy process (for large indexes). I am not intimate with the behavior of IIS's process termination, but summarily killing a process in the midst of an optimize can be problematic.
-----Original Message-----
From: Simone Chiaretta [mailto:simone.chiaretta@gmail.com]
Sent: Monday, January 04, 2010 4:17 PM
To: lucene-net-user
Subject: "Lock obtain timed out: SimpleFSLock" appearing sometimes
Hi all,
since it's my first time here I'd like to introduce myself.
My name is Simone Chiaretta, and I'm member of Subtext blogging engine team,
and I'm a ASP.NET MVC fan.
I also have a blog at http://codeclimber.net.nz, where I also wrote a few
posts about Lucene.net
But going back to the main topic of this email: I'm implementing a
Lucene.net based search engine for Subtext and I'm writing because I've a
strange problem that happens from times to times:
I get the exception of the subject "Lock obtain timed out: SimpleFSLock..."
every now and then when the ApplicationPool recycles.
I know only one IndexWriter can exist at a time, and in fact the writer is a
static variable inside the search engine service.
The problem seems that the Dispose of the search engine, which closes the
writer is not enough for releasing the lock.
public void Dispose()
{
var searcher = _searcher;
if (searcher != null)
{
searcher.Close();
}
var writer = _writer;
if (writer != null)
{
writer.Close();
}
var directory = _directory;
if(directory != null) {
directory.Close();
}
writer = _writer;
if (writer != null)
{
_writer = null;
}
}
For the moment I made 2 workaround that I hope will solve the problem:
I'm checking weather the index is locked or not before creating a new one
if(_writer==null)
{
if(IndexReader.IsLocked(_directory))
{
__log.Error("Something left a lock in the index folder: deleting
it");
IndexReader.Unlock(_directory);
__log.Info("Lock Deleted... can proceed");
}
_writer = new IndexWriter(_directory, _analyzer);
_writer.Optimize();
}
and I'm explicitly disposing the search engine in the Application_End event
of the WebApplication
searchEngine.Dispose();
But I'd like to know if maybe I'm making some mistake somewhere that is
causing the problem, or if there is a more elegant way to fix this problem.
For a complete reference here is the link to the latest version of the
SearchEngineService class that encapsulate all the fulltext searching logic
of Subtext
http://code.google.com/p/subtext/source/browse/trunk/src/Subtext.Framework/Services/SearchEngine/SearchEngineService.cs
Oh... one last thing.. I'm using Lucene.net 2.3.2
Thank you
Simone
--
Simone Chiaretta
Microsoft MVP ASP.NET - ASPInsider
Blog: http://codeclimber.net.nz
RSS: http://feeds2.feedburner.com/codeclimber
twitter: @simonech
Any sufficiently advanced technology is indistinguishable from magic
"Life is short, play hard"