[RFR] templates://hadoop/{hadoop-namenoded.templates}
Thomas Koch
thomas at koch.ro
Tue Mar 30 07:27:37 UTC 2010
Thank you Christian for the thorough review. It feels very good to get the
help of "older" DDs.
I'm very comfortable with your improvements and would like to apply them
without further changes.
Best regards, Thomas
Christian PERRIER:
> Please find, for review, the debconf templates and packages descriptions
> for the hadoop source package.
>
> This review will last from Tuesday, March 30, 2010 to Friday, April 09,
> 2010.
>
> Please send reviews as unified diffs (diff -u) against the original
> files. Comments about your proposed changes will be appreciated.
>
> Your review should be sent as an answer to this mail.
>
> When appropriate, I will send intermediate requests for review, with
> "[RFRn]" (n>=2) as a subject tag.
>
> When we will reach a consensus, I send a "Last Chance For
> Comments" mail with "[LCFC]" as a subject tag.
>
> Finally, the reviewed templates will be sent to the package maintainer
> as a bug report, and a mail will be sent to this list with "[BTS]" as
> a subject tag.
>
> Rationale:
> --- hadoop.old/debian/hadoop-namenoded.templates 2010-03-22
> 09:56:11.717948376 +0100 +++
> hadoop/debian/hadoop-namenoded.templates 2010-03-30 07:22:12.123757400
> +0200 @@ -1,17 +1,17 @@
> Template: hadoop-namenoded/format
> Type: boolean
> Default: false
> -_Description: Should the namenode's filesystem be formatted now?
> +_Description: Should the namenode's file system be formatted?
>
> in other packages, we standardized on "file system". Applying this all
> along this review
>
> The namenode manages the Hadoop Distributed FileSystem (HDFS). Like a
> - normal filesystem, it needs to be formatted prior to first use. If the
> - HDFS filesystem is not formatted, the namenode daemon will fail to
> + normal file system, it needs to be formatted prior to first use. If the
> + HDFS file system is not formatted, the namenode daemon will fail to
> start.
> .
> - This operation does not affect the "normal" filesystem on this
> - computer. If you're using HDFS for the first time and don't have data
> - from previous installations on this computer, it should be save to
> - proceed with yes.
> + This operation does not affect other file systems on this
> + computer. You can safely choose to format the file system if you're
> + using HDFS for the first time and don't have data from previous
> + installations on this computer.
>
> I guess that the main point is to warn users that all "other" FS are
> not at risk here. So, let's mention this slightly differently (they're
> not more "normal" than anything else...and there might be more than
> one file system on the system, of course).
>
> "procees with yes" is highly discouraged as it makes reference to the
> way the question is shown in *some* debconf interfaces (a yes/no
> question) and, anyway, it's always tricky for translators to know
> whether they should translate the "yes" or not (the answer being "it
> depends"..:-)).
>
> .
> - You can later on format the filesystem yourself with
> - .
> - su -c"hadoop namenode -format" hadoop
> + If you choose not to format the file system right now, you can do it
> + later by executing "hadoop namenode -format" with the hadoop user
> + privileges.
>
> Don't waste space by splitting in two paragraphs. That will anyway
> loko ugly oin some interfaces. I rephrased the paragraph so that it
> doesn't depend on using "su" or not (which is not the point as the
> point is executing the command as "hadoop").
>
>
> --- hadoop.old/debian/control 2010-03-22 09:56:11.717948376 +0100
> +++ hadoop/debian/control 2010-03-26 18:30:25.615052315 +0100
> @@ -44,14 +44,54 @@
> libslf4j-java,
> libxmlenc-java
> Suggests: libhsqldb-java
> -Description: software platform for processing vast amounts of data
> - This package contains the core java libraries.
> +Description: platform for processing vast amounts of data - Java libraries
>
> I standardized all binary packages description as
> "general desc - specific desc"
>
> "general desc" drops "software". After all, this is all about software
> anyway? :-)
>
> Proper(?) capitalization fo Java
>
> + Hadoop is a software platform that lets one easily write and
> + run applications that process vast amounts of data.
> + .
> + Here's what makes Hadoop especially useful:
> + * Scalable: Hadoop can reliably store and process petabytes.
> + * Economical: It distributes the data and processing across clusters
> + of commonly available computers. These clusters can number
> + into the thousands of nodes.
> + * Efficient: By distributing the data, Hadoop can process it in parallel
> + on the nodes where the data is located. This makes it
> + extremely rapid.
> + * Reliable: Hadoop automatically maintains multiple copies of data and
> + automatically redeploys computing tasks based on failures.
> + .
> + Hadoop implements MapReduce, using the Hadoop Distributed File System
> (HDFS). + MapReduce divides applications into many small blocks of work.
> HDFS creates + multiple replicas of data blocks for reliability, placing
> them on compute + nodes around the cluster. MapReduce can then process the
> data where it is + located.
> + .
> + This package contains the core Java libraries.
>
> This package...and all others will carry the same boilerplate (first 3
> paragraphs). I used the part you had in one of the packages.
>
> Then 1 or 2 paragraphs will carry the part that's specific to the said
> package.
>
>
> Package: libhadoop-index-java
> Architecture: all
> Depends: ${misc:Depends}, libhadoop-java (= ${binary:Version}),
> liblucene2-java
> -Description: Hadoop contrib to create lucene indexes
> +Description: platform for processing vast amounts of data - create Lucene
> indexes
>
> The original synopsis was quite odd (verb sentence). Keep the "create
> <foo>" style, but I'd actually maybe prefer "Lucene index creation".
>
> Other changes below are mostly applying the same principles....
>
>
> + Hadoop is a software platform that lets one easily write and
> + run applications that process vast amounts of data.
> + .
> + Here's what makes Hadoop especially useful:
> + * Scalable: Hadoop can reliably store and process petabytes.
> + * Economical: It distributes the data and processing across clusters
> + of commonly available computers. These clusters can number
> + into the thousands of nodes.
> + * Efficient: By distributing the data, Hadoop can process it in parallel
> + on the nodes where the data is located. This makes it
> + extremely rapid.
> + * Reliable: Hadoop automatically maintains multiple copies of data and
> + automatically redeploys computing tasks based on failures.
> + .
> + Hadoop implements MapReduce, using the Hadoop Distributed File System
> (HDFS). + MapReduce divides applications into many small blocks of work.
> HDFS creates + multiple replicas of data blocks for reliability, placing
> them on compute + nodes around the cluster. MapReduce can then process the
> data where it is + located.
> + .
> This contrib package provides a utility to build or update an index
> using Map/Reduce.
> .
> @@ -65,7 +105,7 @@
> Architecture: all
> Depends: ${misc:Depends}, libhadoop-java (= ${binary:Version}),
> default-jre-headless | java6-runtime-headless
> -Description: software platform for processing vast amounts of data
> +Description: platform for processing vast amounts of data - binaries
> Hadoop is a software platform that lets one easily write and
> run applications that process vast amounts of data.
> .
> @@ -94,8 +134,22 @@
> Architecture: all
> Depends: ${misc:Depends}, hadoop-bin (= ${binary:Version}), daemon,
> adduser, lsb-base (>= 3.2-14)
> -Description: Creates user and directories for hadoop daemons
> - Prepares some common things for all hadoop daemon packages:
> +Description: platform for processing vast amounts of data - common files
> + Hadoop is a software platform that lets one easily write and
> + run applications that process vast amounts of data.
> + .
> + Here's what makes Hadoop especially useful:
> + * Scalable: Hadoop can reliably store and process petabytes.
> + * Economical: It distributes the data and processing across clusters
> + of commonly available computers. These clusters can number
> + into the thousands of nodes.
> + * Efficient: By distributing the data, Hadoop can process it in parallel
> + on the nodes where the data is located. This makes it
> + extremely rapid.
> + * Reliable: Hadoop automatically maintains multiple copies of data and
> + automatically redeploys computing tasks based on failures.
> + .
> + This package prepares some common things for all hadoop daemon packages:
> * creates the user hadoop
> * creates data and log directories owned by the hadoop user
> * manages the update-alternatives mechanism for hadoop configuration
> @@ -105,14 +159,42 @@
> Section: doc
> Architecture: all
> Depends: ${misc:Depends}, libhadoop-java (= ${binary:Version})
> -Description: Contains the javadoc for hadoop
> - contains the api documentation of hadoop
> +Description: platform for processing vast amounts of data - Java
> documentation + Hadoop is a software platform that lets one easily write
> and
> + run applications that process vast amounts of data.
> + .
> + Here's what makes Hadoop especially useful:
> + * Scalable: Hadoop can reliably store and process petabytes.
> + * Economical: It distributes the data and processing across clusters
> + of commonly available computers. These clusters can number
> + into the thousands of nodes.
> + * Efficient: By distributing the data, Hadoop can process it in parallel
> + on the nodes where the data is located. This makes it
> + extremely rapid.
> + * Reliable: Hadoop automatically maintains multiple copies of data and
> + automatically redeploys computing tasks based on failures.
> + .
> + This package provides the API documentation of Hadoop.
>
> Package: hadoop-tasktrackerd
> Section: misc
> Architecture: all
> Depends: ${misc:Depends}, hadoop-daemons-common (= ${binary:Version})
> -Description: Task Tracker for Hadoop
> +Description: platform for processing vast amounts of data - Task Tracker
> + Hadoop is a software platform that lets one easily write and
> + run applications that process vast amounts of data.
> + .
> + Here's what makes Hadoop especially useful:
> + * Scalable: Hadoop can reliably store and process petabytes.
> + * Economical: It distributes the data and processing across clusters
> + of commonly available computers. These clusters can number
> + into the thousands of nodes.
> + * Efficient: By distributing the data, Hadoop can process it in parallel
> + on the nodes where the data is located. This makes it
> + extremely rapid.
> + * Reliable: Hadoop automatically maintains multiple copies of data and
> + automatically redeploys computing tasks based on failures.
> + .
> The Task Tracker is the Hadoop service that accepts MapReduce tasks and
> computes results. Each node in a Hadoop cluster that should be doing
> computation should run a Task Tracker.
> @@ -121,34 +203,90 @@
> Section: misc
> Architecture: all
> Depends: ${misc:Depends}, hadoop-daemons-common (= ${binary:Version})
> -Description: Job Tracker for Hadoop
> - The jobtracker is a central service which is responsible for managing
> - the tasktracker services running on all nodes in a Hadoop Cluster.
> - The jobtracker allocates work to the tasktracker nearest to the data
> +Description: platform for processing vast amounts of data - Job Tracker
> + Hadoop is a software platform that lets one easily write and
> + run applications that process vast amounts of data.
> + .
> + Here's what makes Hadoop especially useful:
> + * Scalable: Hadoop can reliably store and process petabytes.
> + * Economical: It distributes the data and processing across clusters
> + of commonly available computers. These clusters can number
> + into the thousands of nodes.
> + * Efficient: By distributing the data, Hadoop can process it in parallel
> + on the nodes where the data is located. This makes it
> + extremely rapid.
> + * Reliable: Hadoop automatically maintains multiple copies of data and
> + automatically redeploys computing tasks based on failures.
> + .
> + The Job Tracker is a central service which is responsible for managing
> + the Task Tracker services running on all nodes in an Hadoop Cluster.
> + The Job Tracker allocates work to the tasktracker nearest to the data
> with an available work slot.
>
> Package: hadoop-namenoded
> Section: misc
> Architecture: all
> Depends: ${misc:Depends}, hadoop-daemons-common (= ${binary:Version})
> -Description: Name Node for Hadoop
> +Description: platform for processing vast amounts of data - name node
> + Hadoop is a software platform that lets one easily write and
> + run applications that process vast amounts of data.
> + .
> + Here's what makes Hadoop especially useful:
> + * Scalable: Hadoop can reliably store and process petabytes.
> + * Economical: It distributes the data and processing across clusters
> + of commonly available computers. These clusters can number
> + into the thousands of nodes.
> + * Efficient: By distributing the data, Hadoop can process it in parallel
> + on the nodes where the data is located. This makes it
> + extremely rapid.
> + * Reliable: Hadoop automatically maintains multiple copies of data and
> + automatically redeploys computing tasks based on failures.
> + .
> The Hadoop Distributed Filesystem (HDFS) requires one unique server, the
> - namenode, which manages the block locations of files on the filesystem.
> + name node, which manages the block locations of files on the file system.
>
> Package: hadoop-secondarynamenoded
> Section: misc
> Architecture: all
> Depends: ${misc:Depends}, hadoop-daemons-common (= ${binary:Version})
> -Description: Secondary Name Node for Hadoop
> - The Secondary Name Node is responsible for checkpointing file system
> images. - It is _not_ a failover pair for the namenode, and may safely be
> run on the +Description: platform for processing vast amounts of data -
> secondary name node + Hadoop is a software platform that lets one easily
> write and
> + run applications that process vast amounts of data.
> + .
> + Here's what makes Hadoop especially useful:
> + * Scalable: Hadoop can reliably store and process petabytes.
> + * Economical: It distributes the data and processing across clusters
> + of commonly available computers. These clusters can number
> + into the thousands of nodes.
> + * Efficient: By distributing the data, Hadoop can process it in parallel
> + on the nodes where the data is located. This makes it
> + extremely rapid.
> + * Reliable: Hadoop automatically maintains multiple copies of data and
> + automatically redeploys computing tasks based on failures.
> + .
> + The secondary name node is responsible for checkpointing file system
> images. + It is _not_ a failover pair for the name node, and may safely be
> run on the same machine.
>
> Package: hadoop-datanoded
> Section: misc
> Architecture: all
> Depends: ${misc:Depends}, hadoop-daemons-common (= ${binary:Version})
> -Description: Data Node for Hadoop
> - The Data Nodes in the Hadoop Cluster are responsible for serving up
> +Description: platform for processing vast amounts of data - data node
> + Hadoop is a software platform that lets one easily write and
> + run applications that process vast amounts of data.
> + .
> + Here's what makes Hadoop especially useful:
> + * Scalable: Hadoop can reliably store and process petabytes.
> + * Economical: It distributes the data and processing across clusters
> + of commonly available computers. These clusters can number
> + into the thousands of nodes.
> + * Efficient: By distributing the data, Hadoop can process it in parallel
> + on the nodes where the data is located. This makes it
> + extremely rapid.
> + * Reliable: Hadoop automatically maintains multiple copies of data and
> + automatically redeploys computing tasks based on failures.
> + .
> + The data nodes in the Hadoop Cluster are responsible for serving up
> blocks of data over the network to Hadoop Distributed Filesystem
> (HDFS) clients.
>
Thomas Koch, http://www.koch.ro
More information about the pkg-java-maintainers
mailing list