Migration from FAST ESP to Lucene Solr

Download the presentation and see the video.

Michael McIntosh, Vice President of Enterprise Search Technologies at TNR, spoke at the Lucene Revolution conference in Boston, MA October 7-8, 2010. Michael reviewed the migration from Fast ESP to Lucene/Solr open source search. He discussed approaches to identifying core content areas of HTML documents such as Text-To-Tag Ratio Heuristics and Page Stereotype/Site Template Analysis, and reviewed specific use cases that we have encountered as search integration experts and discuss available tools.

TNR Global was a sponsor of Lucene Revolution. The conference gathered over 400 professionals from the enterprise search industry. We were happy to see so much interest in Lucene/Solr open source search, and get to know and learn from the folks who have done large scale implementations, including Twitter, LinkedIn, and eHarmony.  Not surprisingly, there was a lot of interest about migration from proprietory search systems to Solr, especially from FAST ESP due to Microsoft’s discontinuing FAST ESP support for Linux.  If you would like to learn more about how a migration from FAST ESP to Lucene Solr can benefit your company, contact us for a free consultation.

How to create a duplicate ESP collection without re-crawling!

In a production (or even stable) ESP environment, it is difficult to make a change to the Document Processing Pipeline and test it without wiping out the existing collection (not to mention the time it takes to perform a full re-crawl if the collection is even moderately large). In this case, the best option is to use postprocess to feed existing documents to a new (empty) collection.

Making a duplicate collection provides several benefits:

  • No re-crawling is required
  • The original collection is not affected by pipeline changes
  • You can test your new collection without touching the stable data
  • Upon determining that your changes are producing good results, you can easily migrate your front-end to the new collection while still maintaining existing stable data in the original collection (in case you want to revert your changes)

Steps to make a duplicate collection

  1. Using the ESP Admin GUI, create a new collection with the pipeline you would like to use (or test, as the case may be)
  2. Do not specify any data sources when configuring the new collection
  3. Stop the Enterprise Crawler:

    $FASTSEARCH/bin/nctrl stop crawler

  4. Run the following command where origcollection is the original collection and newcollection is the new collection (that you just created):

    $FASTSEARCH/bin/postprocess -R origcollection -k default:newcollection

    Notes about this command:

    • the default specified above is a content feeding destination, as specified in the destinations section of $FASTSEARCH/etc/CrawlerGlobalDefaults.xml. Specifying default will specify the destination as the current ESP install.
    • be sure to run the above command using either nohup or screen as it will not exit until all content has been fed to the new collection. For large collections this may take a while.
  5. Restart the Enterprise Crawler:

    $FASTSEARCH/bin/nctrl start crawler

Fast ESP Error: no doc procs registered to process a batch with priority 0

Just wanted to take this error message off of the, “Hey, we’ve seen this before… now how did we resolve this..?” pile.  This is the full text of the error:

WARNING    Could not send batch to ESP content distributor, will retry automatically.
Reason given: process() failed: exception (no_resources) no doc procs registered to 
process a batch with priority 0

At first glance, it looks pretty clear that you just need to [re]start your document processor(s).  However, this won’t necessarily solve the problem.  Turns out that the a likely reason for this to pop up is a bad Document Processing Pipeline (DPP) Stage.  The docprocs fire up, hit the bad stage (e.g. python errors etc.) and don’t recover.

To debug your DPP Stage, take a look at the logs for the document processor(s).  They’re usually located in $FASTSEARCH/var/log/procserver and, in our experience, there’s probably an uncaught python exception lurking somewhere in there.

Integrate custom services with the Fast ESP Node Controller

Add your own services to ESP's Node Controller
Add your own services to ESP's Node Controller


Integrating our own custom services with Fast ESP’s Node Controller provides us with several benefits:

  • Administrators without in-depth ESP knowledge can easily control services (e.g. start, stop, configure parameters)
  • Services can be started at boot time with the rest of ESP
  • espdeploy can be used to install our services in a multi-node cluster

The components required for this system are:

  1. The ESP Node Controller (with config file NodeConf.xml)
  2. The 3rd-party service (like a CherryPy server, log parser, etc.)
  3. A wrapper script (see below)

Steps for Integration

  1. Define the service you would like to integrate. It can be any script or binary that can be executed on the system. For example, the service might be a python script that takes command-line arguments and continues running itself (as is the case with a webserver).
  2. Create the wrapper script that sets up the proper environment and runs/stops the service properly. The wrapper script should be put in the $FASTSEARCH/bin directory (with executable permissions). Additionally, the wrapper script should pass $@ to your actual script so any/all arguments defined in $FASTSEARCH/etc/NodeConf.xml will be passed along properly from the Node Controller to your service. The following is an example of a wrapper script:
    # export the proper python path
    export PYTHONPATH=":/path/to/python"
    # run the script (backgrounded)
    python $FASTSEARCH/lib/python2.6/yourmodule/yourservice.py $@ &
    # determine the process id of the python script
    # upon receiving a SIGTERM, forward it to the process
    trap "kill -TERM $SCRIPT_PID" SIGTERM
    # wait for SIGTERM from nctrl
  3. Define the service in $FASTSEARCH/etc/NodeConf.xml
    Add the following to the end of the <startorder> tag:


    Add the following to the end of the <node> tag, customizing as appropriate:

    <!-- My Custom Service -->
    <process name="servicename" description="My Custom Service">
                    <parameters>-p 16940 -v</parameters>
                    <port base="4004"/>
  4. Reload the Node Controller configuration with the following:
    nctrl reloadcfg

And that’s it!  Now you should be able to start, stop, configure, and deploy your services using Fast ESP tools.  Enjoy!

Microsoft FAST ESP


TNR Global employs Microsoft FAST ESP (Enterprise Search Platform).  Our implementations support custom and standard formats such as text, HTML, XML, and PDF. We have configured, mirrored, scaled and maintained the ESP system in a rigorous production environment both for Linux and SharePoint.

At TNR Global, we implement and customize the Microsoft FAST Search solution to empower our customers to reach their business goals.

Our expertise with Fast ESP

  • Configuring, mirroring and scaling Microsoft FAST ESP systems using various architectural layouts
  • Custom document tagging pipeline stage development for associating database content with web content based upon document URL
  • Custom dependency-based content build and feeding systems
  • Access to low-level undocumented ESP XML-RPC APIs for better integration
  • ESP benchmarking and performance tuning
  • FAST Index XML repartitioning tools for content volume scaling
  • Proven ESP content backup & recovery techniques
  • Handling of extensive or unplanned system changes without impacting service availability
  • Ajax / Web 2.0 ESP-Suggest functionality integration that uses actual ESP query logs
  • Seamless handling of hardware failure through service mirroring and failover modes
  • Expertise with low-level search engine architecture and search/relevancy algorithms

See an example of the Microsoft FAST ESP Search implementation by TNR Global and CMG at ThomasNet.

TNR has worked with the FAST ESP product since 2004, from version FAST Data Search (FDS) 3.2 up through version FAST ESP 5.3. In 2007, FAST Search and Transfer was acquired by Microsoft. It is Microsoft’s plan to use powerful FAST style technology for their public search engine, Bing. FAST’s flexible and scalable enterprise search platform elevates the search capabilities of enterprise customers and connects people to the relevant information they seek regardless of medium. This drives revenues and reduces total cost of ownership by effectively leveraging IT infrastructure. FAST ESP is known for its scalability, relevancy, and reliability. More than 2,600 customers worldwide use FAST solutions. Contact us for a free consultation.

Microsoft FAST ESP Overview

February 10, 2009
By Michael McIntosh, Senior Search Software Engineer
TNR Global
We use Microsoft FAST ESP to power a large industrial search engine listing over 1 million companies and over 3 million indexed documents and receiving millions of visitors every month. I have been working with ESP since 2003 (then known as FDS 3.2).
Microsoft FAST ESP is extremely flexible and can deal with indexing many document types (html, pdf, word, etc). It has a very robust crawler for web documents and you can use their intermediary FastXML format to load custom document formats into the system or use their Content APIs.
One of my favorite parts of the engine is its Document Processing Pipeline which lets you make use of dozens of out-of-the-box processing plugins as well as using a Python API to write your own custom document processing stages. An example of a custom stage we wrote was one that looks at a web site URL and tries to identify which company it belongs to so additional metadata can be attached to a web document.
It has a very robust programming/integration SDK in several popular languages (C++/C#/Java) for adding content and performing queries as well as fetching system status and managing cluster services.
ESP has a query language called FAST Query Language (FQL) that is very robust and allows you to do basic Boolean searches (AND, OR, NOT) as well as phrase and term proximity searches. In addition to that, it has something called “scope search” which can be used to search document metadata (XML) that has a format that can vary from document to document.
In terms of performance, it scales fairly linearly. If you benchmark it to determine how it performs on one machine, if you add another machine it generally can double performance. You can run the system on one machine (only recommended for development), or many (for production). It is fault-tolerant (it can still serve some results if one of your load-balanced indices goes offline) and it has full fail-over support (one or more critical machines could die or be taken offline for maintenance and the system will continue to function properly)
So, its very powerful. The documentation nowadays is pretty good. So, you ask, what are the downsides?
Well, if the data you need to make searchable has a format that changes frequently, that might be a pain. FAST ESP has something called an “Index Profile” which is basically a config file it uses to determine what document fields are important and should be used for indexing. Everything fed into ESP is a “document”, even if your loading database table rows into it. Each document has several fields, typical fields being: title, body, keywords, headers, documentvectors, processingtime, etc. You can specify as many of your own custom fields as you wish.
If your content maintains mostly the same format (like web documents) its not a big issue. But if you have to make big changes to which fields should be indexed and how they should be treated, you probably need to edit the Index Profile. Some changes to the index profile are “Hot Updates”, meaning you can make the change and not interrupt service. But, some of the bigger changes are “Cold Updates” which requires a full data refeed and indexing before the change takes effect. Depending on the size of your dataset and how many machines are in your cluster, this operation could take hours or days. Cold Updates are a pain to schedule unless you have plenty of cash for extra hardware that you can bring online while your production systems are performing a cold update and reloading the data. Having to do that on production clusters more than once or twice a year requires a fair amount of planning to get right with minimum or 0% downtime.  Learn more about some of the ways we help our customers get the most from their FAST installations.