Repository Maintenance

<?xml version="1.0" encoding="utf-8" standalone="no"?> <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"> <html xmlns="http://www.w3.org/1999/xhtml"> <head> <meta http-equiv="Content-Type" content="text/html; charset=utf-8" /> <title>Repository Maintenance</title> <link rel="stylesheet" type="text/css" href="styles.css" /> <meta name="generator" content="DocBook XSL Stylesheets Vsnapshot" /> <link rel="home" href="index.html" title="Version Control with Subversion" /> <link rel="up" href="svn.reposadmin.html" title="Chapter 5. Repository Administration" /> <link rel="prev" href="svn.reposadmin.create.html" title="Creating and Configuring Your Repository" /> <link rel="next" href="svn.reposadmin.maint.moving-and-removing.html" title="Moving and Removing Repositories" /> </head> <body> <div xmlns="" id="vcws-version-notice"> <p>This documentation was written to describe the 1.7.x series of Apache™ Subversion®. If you are running a different version of Subversion, you are strongly encouraged to visit <a href="http://www.svnbook.com/">http://www.svnbook.com/</a> and instead consult the version of this documentation appropriate for your version of Subversion.</p> </div> <div class="navheader"> <table width="100%" summary="Navigation header"> <tr> <th colspan="3" align="center">Repository Maintenance</th> </tr> <tr> <td width="20%" align="left"><a accesskey="p" href="svn.reposadmin.create.html">Prev</a> </td> <th width="60%" align="center">Chapter 5. Repository Administration</th> <td width="20%" align="right"> <a accesskey="n" href="svn.reposadmin.maint.moving-and-removing.html">Next</a></td> </tr> </table> <hr /> </div> <div class="sect1"> <div class="titlepage"> <div> <div> <h2 class="title" style="clear: both"><a id="svn.reposadmin.maint"></a>Repository Maintenance</h2> </div> </div> </div> <p>Maintaining a Subversion repository can be daunting, mostly due to the complexities inherent in systems that have a database backend. Doing the task well is all about knowing the tools—what they are, when to use them, and how. This section will introduce you to the repository administration tools provided by Subversion and discuss how to wield them to accomplish tasks such as repository data migration, upgrades, backups, and cleanups.</p> <div class="sect2"> <div class="titlepage"> <div> <div> <h3 class="title"><a id="svn.reposadmin.maint.tk"></a>An Administrator's Toolkit</h3> </div> </div> </div> <p>Subversion provides a handful of utilities useful for creating, inspecting, modifying, and repairing your repository. Let's look more closely at each of those tools. Afterward, we'll briefly examine some of the utilities included in the Berkeley DB distribution that provide functionality specific to your repository's database backend not otherwise provided by Subversion's own tools.</p> <div class="sect3"> <div class="titlepage"> <div> <div> <h4 class="title"><a id="svn.reposadmin.maint.tk.svnadmin"></a>svnadmin</h4> </div> </div> </div> <p>The <span class="command"><strong>svnadmin</strong></span> program is the repository administrator's best friend. Besides providing the ability to create Subversion repositories, this program allows you to perform several maintenance operations on those repositories. The syntax of <span class="command"><strong>svnadmin</strong></span> is similar to that of other Subversion command-line programs:</p> <div class="informalexample"> <pre class="screen"> $ svnadmin help general usage: svnadmin SUBCOMMAND REPOS_PATH [ARGS & OPTIONS ...] Type 'svnadmin help <subcommand>' for help on a specific subcommand. Type 'svnadmin --version' to see the program version and FS modules. Available subcommands: crashtest create deltify … </pre> </div> <p>Previously in this chapter (in <a class="xref" href="svn.reposadmin.create.html#svn.reposadmin.basics.creating" title="Creating the Repository">the section called “Creating the Repository”</a>), we were introduced to the <span class="command"><strong>svnadmin create</strong></span> subcommand. Most of the other <span class="command"><strong>svnadmin</strong></span> subcommands we will cover later in this chapter. And you can consult <a class="xref" href="svn.ref.svnadmin.html" title="svnadmin—Subversion Repository Administration">the section called “svnadmin—Subversion Repository Administration”</a> in <a class="xref" href="svn.ref.html" title="Chapter 9. Subversion Complete Reference">Chapter 9, <em>Subversion Complete Reference</em></a> for a full rundown of subcommands and what each of them offers.</p> </div> <div class="sect3"> <div class="titlepage"> <div> <div> <h4 class="title"><a id="svn.reposadmin.maint.tk.svnlook"></a>svnlook</h4> </div> </div> </div> <p><span class="command"><strong>svnlook</strong></span> is a tool provided by Subversion for examining the various revisions and <em class="firstterm">transactions</em> (which are revisions in the making) in a repository. No part of this program attempts to change the repository. <span class="command"><strong>svnlook</strong></span> is typically used by the repository hooks for reporting the changes that are about to be committed (in the case of the pre-commit hook) or that were just committed (in the case of the post-commit hook) to the repository. A repository administrator may use this tool for diagnostic purposes.</p> <p><span class="command"><strong>svnlook</strong></span> has a straightforward syntax:</p> <div class="informalexample"> <pre class="screen"> $ svnlook help general usage: svnlook SUBCOMMAND REPOS_PATH [ARGS & OPTIONS ...] Note: any subcommand which takes the '--revision' and '--transaction' options will, if invoked without one of those options, act on the repository's youngest revision. Type 'svnlook help <subcommand>' for help on a specific subcommand. Type 'svnlook --version' to see the program version and FS modules. … </pre> </div> <p>Most of <span class="command"><strong>svnlook</strong></span>'s subcommands can operate on either a revision or a transaction tree, printing information about the tree itself, or how it differs from the previous revision of the repository. You use the <code class="option">--revision</code> (<code class="option">-r</code>) and <code class="option">--transaction</code> (<code class="option">-t</code>) options to specify which revision or transaction, respectively, to examine. In the absence of both the <code class="option">--revision</code> (<code class="option">-r</code>) and <code class="option">--transaction</code> (<code class="option">-t</code>) options, <span class="command"><strong>svnlook</strong></span> will examine the youngest (or <code class="literal">HEAD</code>) revision in the repository. So the following two commands do exactly the same thing when 19 is the youngest revision in the repository located at <code class="filename">/var/svn/repos</code>:</p> <div class="informalexample"> <pre class="screen"> $ svnlook info /var/svn/repos $ svnlook info /var/svn/repos -r 19 </pre> </div> <p>One exception to these rules about subcommands is the <span class="command"><strong>svnlook youngest</strong></span> subcommand, which takes no options and simply prints out the repository's youngest revision number:</p> <div class="informalexample"> <pre class="screen"> $ svnlook youngest /var/svn/repos 19 $ </pre> </div> <div class="note" style="margin-left: 0.5in; margin-right: 0.5in;"> <table border="0" summary="Note"> <tr> <td rowspan="2" align="center" valign="top" width="25"> <img alt="[Note]" src="images/note.png" /> </td> <th align="left">Note</th> </tr> <tr> <td align="left" valign="top"> <p>Keep in mind that the only transactions you can browse are uncommitted ones. Most repositories will have no such transactions because transactions are usually either committed (in which case, you should access them as revision with the <code class="option">--revision</code> (<code class="option">-r</code>) option) or aborted and removed.</p> </td> </tr> </table> </div> <p>Output from <span class="command"><strong>svnlook</strong></span> is designed to be both human- and machine-parsable. Take, as an example, the output of the <span class="command"><strong>svnlook info</strong></span> subcommand:</p> <div class="informalexample"> <pre class="screen"> $ svnlook info /var/svn/repos sally 2002-11-04 09:29:13 -0600 (Mon, 04 Nov 2002) 27 Added the usual Greek tree. $ </pre> </div> <p>The output of <span class="command"><strong>svnlook info</strong></span> consists of the following, in the order given:</p> <div class="orderedlist"> <ol class="orderedlist" type="1"> <li class="listitem"> <p>The author, followed by a newline</p> </li> <li class="listitem"> <p>The date, followed by a newline</p> </li> <li class="listitem"> <p>The number of characters in the log message, followed by a newline</p> </li> <li class="listitem"> <p>The log message itself, followed by a newline</p> </li> </ol> </div> <p>This output is human-readable, meaning items such as the datestamp are displayed using a textual representation instead of something more obscure (such as the number of nanoseconds since the Tastee Freez guy drove by). But the output is also machine-parsable—because the log message can contain multiple lines and be unbounded in length, <span class="command"><strong>svnlook</strong></span> provides the length of that message before the message itself. This allows scripts and other wrappers around this command to make intelligent decisions about the log message, such as how much memory to allocate for the message, or at least how many bytes to skip in the event that this output is not the last bit of data in the stream.</p> <p><span class="command"><strong>svnlook</strong></span> can perform a variety of other queries: displaying subsets of bits of information we've mentioned previously, recursively listing versioned directory trees, reporting which paths were modified in a given revision or transaction, showing textual and property differences made to files and directories, and so on. See <a class="xref" href="svn.ref.svnlook.html" title="svnlook—Subversion Repository Examination">the section called “svnlook—Subversion Repository Examination”</a> in <a class="xref" href="svn.ref.html" title="Chapter 9. Subversion Complete Reference">Chapter 9, <em>Subversion Complete Reference</em></a> for a full reference of <span class="command"><strong>svnlook</strong></span>'s features.</p> </div> <div class="sect3"> <div class="titlepage"> <div> <div> <h4 class="title"><a id="svn.reposadmin.maint.tk.svndumpfilter"></a>svndumpfilter</h4> </div> </div> </div> <p>While it won't be the most commonly used tool at the administrator's disposal, <span class="command"><strong>svndumpfilter</strong></span> provides a very particular brand of useful functionality—the ability to quickly and easily modify streams of Subversion repository history data by acting as a path-based filter.</p> <p>The syntax of <span class="command"><strong>svndumpfilter</strong></span> is as follows:</p> <div class="informalexample"> <pre class="screen"> $ svndumpfilter help general usage: svndumpfilter SUBCOMMAND [ARGS & OPTIONS ...] Type 'svndumpfilter help <subcommand>' for help on a specific subcommand. Type 'svndumpfilter --version' to see the program version. Available subcommands: exclude include help (?, h) </pre> </div> <p>There are only two interesting subcommands: <span class="command"><strong>svndumpfilter exclude</strong></span> and <span class="command"><strong>svndumpfilter include</strong></span>. They allow you to make the choice between implicit or explicit inclusion of paths in the stream. You can learn more about these subcommands and <span class="command"><strong>svndumpfilter</strong></span>'s unique purpose later in this chapter, in <a class="xref" href="svn.reposadmin.maint.html#svn.reposadmin.maint.filtering" title="Filtering Repository History">the section called “Filtering Repository History”</a>.</p> </div> <div class="sect3"> <div class="titlepage"> <div> <div> <h4 class="title"><a id="svn.reposadmin.maint.tk.svnrdump"></a>svnrdump</h4> </div> </div> </div> <p>The <span class="command"><strong>svnrdump</strong></span> program is, to put it simply, essentially just network-aware flavors of the <span class="command"><strong>svnadmin dump</strong></span> and <span class="command"><strong>svnadmin load</strong></span> subcommands, rolled up into a separate program.</p> <div class="informalexample"> <pre class="screen"> $ svnrdump help general usage: svnrdump SUBCOMMAND URL [-r LOWER[:UPPER]] Type 'svnrdump help <subcommand>' for help on a specific subcommand. Type 'svnrdump --version' to see the program version and RA modules. Available subcommands: dump load help (?, h) $ </pre> </div> <p>We discuss the use of <span class="command"><strong>svnrdump</strong></span> and the aforementioned <span class="command"><strong>svnadmin</strong></span> commands later in this chapter (see <a class="xref" href="svn.reposadmin.maint.html#svn.reposadmin.maint.migrate" title="Migrating Repository Data Elsewhere">the section called “Migrating Repository Data Elsewhere”</a>).</p> </div> <div class="sect3"> <div class="titlepage"> <div> <div> <h4 class="title"><a id="svn.reposadmin.maint.tk.svnsync"></a>svnsync</h4> </div> </div> </div> <p>The <span class="command"><strong>svnsync</strong></span> program provides all the functionality required for maintaining a read-only mirror of a Subversion repository. The program really has one job—to transfer one repository's versioned history into another repository. And while there are few ways to do that, its primary strength is that it can operate remotely—the <span class="quote">“<span class="quote">source</span>”</span> and <span class="quote">“<span class="quote">sink</span>”</span><a href="#ftn.idm5421" class="footnote" id="idm5421"><sup class="footnote">[43]</sup></a> repositories may be on different computers from each other and from <span class="command"><strong>svnsync</strong></span> itself.</p> <p>As you might expect, <span class="command"><strong>svnsync</strong></span> has a syntax that looks very much like every other program we've mentioned in this chapter:</p> <div class="informalexample"> <pre class="screen"> $ svnsync help general usage: svnsync SUBCOMMAND DEST_URL [ARGS & OPTIONS ...] Type 'svnsync help <subcommand>' for help on a specific subcommand. Type 'svnsync --version' to see the program version and RA modules. Available subcommands: initialize (init) synchronize (sync) copy-revprops info help (?, h) $ </pre> </div> <p>We talk more about replicating repositories with <span class="command"><strong>svnsync</strong></span> later in this chapter (see <a class="xref" href="svn.reposadmin.maint.html#svn.reposadmin.maint.replication" title="Repository Replication">the section called “Repository Replication”</a>).</p> </div> <div class="sect3"> <div class="titlepage"> <div> <div> <h4 class="title"><a id="svn.reposadmin.maint.tk.fsfsreshard"></a>fsfs-reshard.py</h4> </div> </div> </div> <p>While not an official member of the Subversion toolchain, the <span class="command"><strong>fsfs-reshard.py</strong></span> script (found in the <code class="filename">tools/server-side</code> directory of the Subversion source distribution) is a useful performance tuning tool for administrators of FSFS-backed Subversion repositories. As described in the sidebar <a class="xref" href="svn.reposadmin.planning.html#svn.reposadmin.basics.backends.fsfs.revfiles" title="Revision files and shards">Revision files and shards</a>, FSFS repositories use individual files to house information about each revision. Sometimes these files all live in a single directory; sometimes they are sharded across many directories. But the neat thing is that the number of directories used to house these files is configurable. That's where <span class="command"><strong>fsfs-reshard.py</strong></span> comes in.</p> <p><span class="command"><strong>fsfs-reshard.py</strong></span> reshuffles the repository's file structure into a new arrangement that reflects the requested number of sharding subdirectories and updates the repository configuration to preserve this change. When used in conjunction with the <span class="command"><strong>svnadmin upgrade</strong></span> command, this is especially useful for upgrading a pre-1.5 Subversion (unsharded) repository to the latest filesystem format and sharding its data files (which Subversion will not automatically do for you). This script can also be used for fine-tuning an already sharded repository.</p> </div> <div class="sect3"> <div class="titlepage"> <div> <div> <h4 class="title"><a id="svn.reposadmin.maint.tk.bdbutil"></a>Berkeley DB utilities</h4> </div> </div> </div> <p>If you're using a Berkeley DB repository, all of your versioned filesystem's structure and data live in a set of database tables within the <code class="filename">db/</code> subdirectory of your repository. This subdirectory is a regular Berkeley DB environment directory and can therefore be used in conjunction with any of the Berkeley database tools, typically provided as part of the Berkeley DB distribution.</p> <p>For day-to-day Subversion use, these tools are unnecessary. Most of the functionality typically needed for Subversion repositories has been duplicated in the <span class="command"><strong>svnadmin</strong></span> tool. For example, <span class="command"><strong>svnadmin list-unused-dblogs</strong></span> and <span class="command"><strong>svnadmin list-dblogs</strong></span> perform a subset of what is provided by the Berkeley <span class="command"><strong>db_archive</strong></span> utility, and <span class="command"><strong>svnadmin recover</strong></span> reflects the common use cases of the <span class="command"><strong>db_recover</strong></span> utility.</p> <p>However, there are still a few Berkeley DB utilities that you might find useful. The <span class="command"><strong>db_dump</strong></span> and <span class="command"><strong>db_load</strong></span> programs write and read, respectively, a custom file format that describes the keys and values in a Berkeley DB database. Since Berkeley databases are not portable across machine architectures, this format is a useful way to transfer those databases from machine to machine, irrespective of architecture or operating system. As we describe later in this chapter, you can also use <span class="command"><strong>svnadmin dump</strong></span> and <span class="command"><strong>svnadmin load</strong></span> for similar purposes, but <span class="command"><strong>db_dump</strong></span> and <span class="command"><strong>db_load</strong></span> can do certain jobs just as well and much faster. They can also be useful if the experienced Berkeley DB hacker needs to do in-place tweaking of the data in a BDB-backed repository for some reason, which is something Subversion's utilities won't allow. Also, the <span class="command"><strong>db_stat</strong></span> utility can provide useful information about the status of your Berkeley DB environment, including detailed statistics about the locking and storage subsystems.</p> <p>For more information on the Berkeley DB tool chain, visit the documentation section of the Berkeley DB section of Oracle's web site, located at <a class="ulink" href="http://www.oracle.com/technology/documentation/berkeley-db/db/" target="_top">http://www.oracle.com/technology/documentation/berkeley-db/db/</a>.</p> </div> </div> <div class="sect2"> <div class="titlepage"> <div> <div> <h3 class="title"><a id="svn.reposadmin.maint.setlog"></a>Commit Log Message Correction</h3> </div> </div> </div> <p>Sometimes a user will have an error in her log message (a misspelling or some misinformation, perhaps). If the repository is configured (using the pre-revprop-change hook; see <a class="xref" href="svn.reposadmin.create.html#svn.reposadmin.create.hooks" title="Implementing Repository Hooks">the section called “Implementing Repository Hooks”</a>) to accept changes to this log message after the commit is finished, the user can <span class="quote">“<span class="quote">fix</span>”</span> her log message remotely using <span class="command"><strong>svn propset</strong></span> (see <a class="xref" href="svn.ref.svn.c.propset.html" title="svn propset (pset, ps)">svn propset (pset, ps)</a> in <a class="xref" href="svn.ref.html" title="Chapter 9. Subversion Complete Reference">Chapter 9, <em>Subversion Complete Reference</em></a>). However, because of the potential to lose information forever, Subversion repositories are not, by default, configured to allow changes to unversioned properties—except by an administrator.</p> <p>If a log message needs to be changed by an administrator, this can be done using <span class="command"><strong>svnadmin setlog</strong></span>. This command changes the log message (the <code class="literal">svn:log</code> property) on a given revision of a repository, reading the new value from a provided file.</p> <div class="informalexample"> <pre class="screen"> $ echo "Here is the new, correct log message" > newlog.txt $ svnadmin setlog myrepos newlog.txt -r 388 </pre> </div> <p>The <span class="command"><strong>svnadmin setlog</strong></span> command, by default, is still bound by the same protections against modifying unversioned properties as a remote client is—the pre-revprop-change and post-revprop-change hooks are still triggered, and therefore must be set up to accept changes of this nature. But an administrator can get around these protections by passing the <code class="option">--bypass-hooks</code> option to the <span class="command"><strong>svnadmin setlog</strong></span> command.</p> <div class="warning" style="margin-left: 0.5in; margin-right: 0.5in;"> <table border="0" summary="Warning"> <tr> <td rowspan="2" align="center" valign="top" width="25"> <img alt="[Warning]" src="images/warning.png" /> </td> <th align="left">Warning</th> </tr> <tr> <td align="left" valign="top"> <p>Remember, though, that by bypassing the hooks, you are likely avoiding such things as email notifications of property changes, backup systems that track unversioned property changes, and so on. In other words, be very careful about what you are changing, and how you change it.</p> </td> </tr> </table> </div> </div> <div class="sect2"> <div class="titlepage"> <div> <div> <h3 class="title"><a id="svn.reposadmin.maint.diskspace"></a>Managing Disk Space</h3> </div> </div> </div> <p>While the cost of storage has dropped incredibly in the past few years, disk usage is still a valid concern for administrators seeking to version large amounts of data. Every bit of version history information stored in the live repository needs to be backed up elsewhere, perhaps multiple times as part of rotating backup schedules. It is useful to know what pieces of Subversion's repository data need to remain on the live site, which need to be backed up, and which can be safely removed.</p> <div class="sect3"> <div class="titlepage"> <div> <div> <h4 class="title"><a id="svn.reposadmin.maint.diskspace.deltas"></a>How Subversion saves disk space</h4> </div> </div> </div> <p>To keep the repository small, Subversion uses <em class="firstterm">deltification</em> (or delta-based storage) within the repository itself. Deltification involves encoding the representation of a chunk of data as a collection of differences against some other chunk of data. If the two pieces of data are very similar, this deltification results in storage savings for the deltified chunk—rather than taking up space equal to the size of the original data, it takes up only enough space to say, <span class="quote">“<span class="quote">I look just like this other piece of data over here, except for the following couple of changes.</span>”</span> The result is that most of the repository data that tends to be bulky—namely, the contents of versioned files—is stored at a much smaller size than the original full-text representation of that data.</p> <p>While deltified storage has been a part of Subversion's design since the very beginning, there have been additional improvements made over the years. Subversion repositories created with Subversion 1.4 or later benefit from compression of the full-text representations of file contents. Repositories created with Subversion 1.6 or later further enjoy the disk space savings afforded by <em class="firstterm">representation sharing</em>, a feature which allows multiple files or file revisions with identical file content to refer to a single shared instance of that data rather than each having their own distinct copy thereof.</p> <div class="note" style="margin-left: 0.5in; margin-right: 0.5in;"> <table border="0" summary="Note"> <tr> <td rowspan="2" align="center" valign="top" width="25"> <img alt="[Note]" src="images/note.png" /> </td> <th align="left">Note</th> </tr> <tr> <td align="left" valign="top"> <p>Because all of the data that is subject to deltification in a BDB-backed repository is stored in a single Berkeley DB database file, reducing the size of the stored values will not immediately reduce the size of the database file itself. Berkeley DB will, however, keep internal records of unused areas of the database file and consume those areas first before growing the size of the database file. So while deltification doesn't produce immediate space savings, it can drastically slow future growth of the database.</p> </td> </tr> </table> </div> </div> <div class="sect3"> <div class="titlepage"> <div> <div> <h4 class="title"><a id="svn.reposadmin.maint.diskspace.deadtxns"></a>Removing dead transactions</h4> </div> </div> </div> <p>Though they are uncommon, there are circumstances in which a Subversion commit process might fail, leaving behind in the repository the remnants of the revision-to-be that wasn't—an uncommitted transaction and all the file and directory changes associated with it. This could happen for several reasons: perhaps the client operation was inelegantly terminated by the user, or a network failure occurred in the middle of an operation. Regardless of the reason, dead transactions can happen. They don't do any real harm, other than consuming disk space. A fastidious administrator may nonetheless wish to remove them.</p> <p>You can use the <span class="command"><strong>svnadmin lstxns</strong></span> command to list the names of the currently outstanding transactions:</p> <div class="informalexample"> <pre class="screen"> $ svnadmin lstxns myrepos 19 3a1 a45 $ </pre> </div> <p>Each item in the resultant output can then be used with <span class="command"><strong>svnlook</strong></span> (and its <code class="option">--transaction</code> (<code class="option">-t</code>) option) to determine who created the transaction, when it was created, what types of changes were made in the transaction—information that is helpful in determining whether the transaction is a safe candidate for removal! If you do indeed want to remove a transaction, its name can be passed to <span class="command"><strong>svnadmin rmtxns</strong></span>, which will perform the cleanup of the transaction. In fact, <span class="command"><strong>svnadmin rmtxns</strong></span> can take its input directly from the output of <span class="command"><strong>svnadmin lstxns</strong></span>!</p> <div class="informalexample"> <pre class="screen"> $ svnadmin rmtxns myrepos `svnadmin lstxns myrepos` $ </pre> </div> <p>If you use these two subcommands like this, you should consider making your repository temporarily inaccessible to clients. That way, no one can begin a legitimate transaction before you start your cleanup. <a class="xref" href="svn.reposadmin.maint.html#svn.reposadmin.maint.diskspace.deadtxns.ex-1" title="Example 5.1. txn-info.sh (reporting outstanding transactions)">Example 5.1, “txn-info.sh (reporting outstanding transactions)”</a> contains a bit of shell-scripting that can quickly generate information about each outstanding transaction in your repository.</p> <div class="example"> <a id="svn.reposadmin.maint.diskspace.deadtxns.ex-1"></a> <p class="title"> <strong>Example 5.1. txn-info.sh (reporting outstanding transactions)</strong> </p> <div class="example-contents"> <pre class="programlisting"> #!/bin/sh ### Generate informational output for all outstanding transactions in ### a Subversion repository. REPOS="${1}" if [ "x$REPOS" = x ] ; then echo "usage: $0 REPOS_PATH" exit fi for TXN in `svnadmin lstxns ${REPOS}`; do echo "---[ Transaction ${TXN} ]-------------------------------------------" svnlook info "${REPOS}" -t "${TXN}" done </pre> </div> </div> <br class="example-break" /> <p>The output of the script is basically a concatenation of several chunks of <span class="command"><strong>svnlook info</strong></span> output (see <a class="xref" href="svn.reposadmin.maint.html#svn.reposadmin.maint.tk.svnlook" title="svnlook">the section called “svnlook”</a>) and will look something like this:</p> <div class="informalexample"> <pre class="screen"> $ txn-info.sh myrepos ---[ Transaction 19 ]------------------------------------------- sally 2001-09-04 11:57:19 -0500 (Tue, 04 Sep 2001) 0 ---[ Transaction 3a1 ]------------------------------------------- harry 2001-09-10 16:50:30 -0500 (Mon, 10 Sep 2001) 39 Trying to commit over a faulty network. ---[ Transaction a45 ]------------------------------------------- sally 2001-09-12 11:09:28 -0500 (Wed, 12 Sep 2001) 0 $ </pre> </div> <p>A long-abandoned transaction usually represents some sort of failed or interrupted commit. A transaction's datestamp can provide interesting information—for example, how likely is it that an operation begun nine months ago is still active?</p> <p>In short, transaction cleanup decisions need not be made unwisely. Various sources of information—including Apache's error and access logs, Subversion's operational logs, Subversion revision history, and so on—can be employed in the decision-making process. And of course, an administrator can often simply communicate with a seemingly dead transaction's owner (via email, e.g.) to verify that the transaction is, in fact, in a zombie state.</p> </div> <div class="sect3"> <div class="titlepage"> <div> <div> <h4 class="title"><a id="svn.reposadmin.maint.diskspace.bdblogs"></a>Purging unused Berkeley DB logfiles</h4> </div> </div> </div> <p>Until recently, the largest offender of disk space usage with respect to BDB-backed Subversion repositories were the logfiles in which Berkeley DB performs its prewrites before modifying the actual database files. These files capture all the actions taken along the route of changing the database from one state to another—while the database files, at any given time, reflect a particular state, the logfiles contain all of the many changes along the way <span class="emphasis"><em>between</em></span> states. Thus, they can grow and accumulate quite rapidly.</p> <p>Fortunately, beginning with the 4.2 release of Berkeley DB, the database environment has the ability to remove its own unused logfiles automatically. Any repositories created using <span class="command"><strong>svnadmin</strong></span> when compiled against Berkeley DB version 4.2 or later will be configured for this automatic logfile removal. If you don't want this feature enabled, simply pass the <code class="option">--bdb-log-keep</code> option to the <span class="command"><strong>svnadmin create</strong></span> command. If you forget to do this or change your mind at a later time, simply edit the <code class="filename">DB_CONFIG</code> file found in your repository's <code class="filename">db</code> directory, comment out the line that contains the <code class="literal">set_flags DB_LOG_AUTOREMOVE</code> directive, and then run <span class="command"><strong>svnadmin recover</strong></span> on your repository to force the configuration changes to take effect. See <a class="xref" href="svn.reposadmin.create.html#svn.reposadmin.create.bdb" title="Berkeley DB Configuration">the section called “Berkeley DB Configuration”</a> for more information about database configuration.</p> <p>Without some sort of automatic logfile removal in place, logfiles will accumulate as you use your repository. This is actually somewhat of a feature of the database system—you should be able to recreate your entire database using nothing but the logfiles, so these files can be useful for catastrophic database recovery. But typically, you'll want to archive the logfiles that are no longer in use by Berkeley DB, and then remove them from disk to conserve space. Use the <span class="command"><strong>svnadmin list-unused-dblogs</strong></span> command to list the unused logfiles:</p> <div class="informalexample"> <pre class="screen"> $ svnadmin list-unused-dblogs /var/svn/repos /var/svn/repos/log.0000000031 /var/svn/repos/log.0000000032 /var/svn/repos/log.0000000033 … $ rm `svnadmin list-unused-dblogs /var/svn/repos` ## disk space reclaimed! </pre> </div> <div class="warning" style="margin-left: 0.5in; margin-right: 0.5in;"> <table border="0" summary="Warning"> <tr> <td rowspan="2" align="center" valign="top" width="25"> <img alt="[Warning]" src="images/warning.png" /> </td> <th align="left">Warning</th> </tr> <tr> <td align="left" valign="top"> <p>BDB-backed repositories whose logfiles are used as part of a backup or disaster recovery plan should <span class="emphasis"><em>not</em></span> make use of the logfile autoremoval feature. Reconstruction of a repository's data from logfiles can only be accomplished only when <span class="emphasis"><em>all</em></span> the logfiles are available. If some of the logfiles are removed from disk before the backup system has a chance to copy them elsewhere, the incomplete set of backed-up logfiles is essentially useless.</p> </td> </tr> </table> </div> </div> <div class="sect3"> <div class="titlepage"> <div> <div> <h4 class="title"><a id="svn.reposadmin.maint.diskspace.fsfspacking"></a>Packing FSFS filesystems</h4> </div> </div> </div> <p>As described in the sidebar <a class="xref" href="svn.reposadmin.planning.html#svn.reposadmin.basics.backends.fsfs.revfiles" title="Revision files and shards">Revision files and shards</a>, FSFS-backed Subversion repositories create, by default, a new on-disk file for each revision added to the repository. Having thousands of these files present on your Subversion server—even when housed in separate shard directories—can lead to inefficiencies.</p> <p>The first problem is that the operating system has to reference many different files over a short period of time. This leads to inefficient use of disk caches and, as a result, more time spent seeking across large disks. Because of this, Subversion pays a performance penalty when accessing your versioned data.</p> <p>The second problem is a bit more subtle. Because of the ways that most filesystems allocate disk space, each file claims more space on the disk than it actually uses. The amount of extra space required to house a single file can average anywhere from 2 to 16 kilobytes <span class="emphasis"><em>per file</em></span>, depending on the underlying filesystem in use. This translates directly into a per-revision disk usage penalty for FSFS-backed repositories. The effect is most pronounced in repositories which have many small revisions, since the overhead involved in storing the revision file quickly outgrows the size of the actual data being stored.</p> <p>To solve these problems, Subversion 1.6 introduced the <span class="command"><strong>svnadmin pack</strong></span> command. By concatenating all the files of a completed shard into a single <span class="quote">“<span class="quote">pack</span>”</span> file and then removing the original per-revision files, <span class="command"><strong>svnadmin pack</strong></span> reduces the file count within a given shard down to just a single file. In doing so, it aids filesystem caches and reduces (to one) the number of times a file storage overhead penalty is paid.</p> <p>Subversion can pack existing sharded repositories which have been upgraded to the 1.6 filesystem format or later (see <a class="xref" href="svn.ref.svnadmin.c.upgrade.html" title="svnadmin upgrade">svnadmin upgrade</a>) in <a class="xref" href="svn.ref.html" title="Chapter 9. Subversion Complete Reference">Chapter 9, <em>Subversion Complete Reference</em></a>. To do so, just run <span class="command"><strong>svnadmin pack</strong></span> on the repository:</p> <div class="informalexample"> <pre class="screen"> $ svnadmin pack /var/svn/repos Packing shard 0...done. Packing shard 1...done. Packing shard 2...done. … Packing shard 34...done. Packing shard 35...done. Packing shard 36...done. $ </pre> </div> <p>Because the packing process obtains the required locks before doing its work, you can run it on live repositories, or even as part of a post-commit hook. Repacking packed shards is legal, but will have no effect on the disk usage of the repository.</p> <p><span class="command"><strong>svnadmin pack</strong></span> has no effect on BDB-backed Subversion repositories.</p> </div> </div> <div class="sect2"> <div class="titlepage"> <div> <div> <h3 class="title"><a id="svn.reposadmin.maint.recovery"></a>Berkeley DB Recovery</h3> </div> </div> </div> <p>As mentioned in <a class="xref" href="svn.reposadmin.planning.html#svn.reposadmin.basics.backends.bdb" title="Berkeley DB">the section called “Berkeley DB”</a>, a Berkeley DB repository can sometimes be left in a frozen state if not closed properly. When this happens, an administrator needs to rewind the database back into a consistent state. This is unique to BDB-backed repositories, though—if you are using FSFS-backed ones instead, this won't apply to you. And for those of you using Subversion 1.4 with Berkeley DB 4.4 or later, you should find that Subversion has become much more resilient in these types of situations. Still, wedged Berkeley DB repositories do occur, and an administrator needs to know how to safely deal with this circumstance.</p> <p>To protect the data in your repository, Berkeley DB uses a locking mechanism. This mechanism ensures that portions of the database are not simultaneously modified by multiple database accessors, and that each process sees the data in the correct state when that data is being read from the database. When a process needs to change something in the database, it first checks for the existence of a lock on the target data. If the data is not locked, the process locks the data, makes the change it wants to make, and then unlocks the data. Other processes are forced to wait until that lock is removed before they are permitted to continue accessing that section of the database. (This has nothing to do with the locks that you, as a user, can apply to versioned files within the repository; we try to clear up the confusion caused by this terminology collision in the sidebar <a class="xref" href="svn.advanced.locking.html#svn.advanced.locking.meanings" title="The Three Meanings of “Lock”">The Three Meanings of <span class="quote">“<span class="quote">Lock</span>”</span></a>.)</p> <p>In the course of using your Subversion repository, fatal errors or interruptions can prevent a process from having the chance to remove the locks it has placed in the database. The result is that the backend database system gets <span class="quote">“<span class="quote">wedged.</span>”</span> When this happens, any attempts to access the repository hang indefinitely (since each new accessor is waiting for a lock to go away—which isn't going to happen).</p> <p>If this happens to your repository, don't panic. The Berkeley DB filesystem takes advantage of database transactions, checkpoints, and prewrite journaling to ensure that only the most catastrophic of events<a href="#ftn.idm5572" class="footnote" id="idm5572"><sup class="footnote">[44]</sup></a> can permanently destroy a database environment. A sufficiently paranoid repository administrator will have made off-site backups of the repository data in some fashion, but don't head off to the tape backup storage closet just yet.</p> <p>Instead, use the following recipe to attempt to <span class="quote">“<span class="quote">unwedge</span>”</span> your repository:</p> <div class="orderedlist"> <ol class="orderedlist" type="1"> <li class="listitem"> <p>Make sure no processes are accessing (or attempting to access) the repository. For networked repositories, this also means shutting down the Apache HTTP Server or svnserve daemon.</p> </li> <li class="listitem"> <p>Become the user who owns and manages the repository. This is important, as recovering a repository while running as the wrong user can tweak the permissions of the repository's files in such a way that your repository will still be inaccessible even after it is <span class="quote">“<span class="quote">unwedged.</span>”</span></p> </li> <li class="listitem"> <p>Run the command <strong class="userinput"><code>svnadmin recover /var/svn/repos</code></strong>. You should see output such as this:</p> <div class="informalexample"> <pre class="screen"> Repository lock acquired. Please wait; recovering the repository may take some time... Recovery completed. The latest repos revision is 19. </pre> </div> <p>This command may take many minutes to complete.</p> </li> <li class="listitem"> <p>Restart the server process.</p> </li> </ol> </div> <p>This procedure fixes almost every case of repository wedging. Make sure that you run this command as the user that owns and manages the database, not just as <code class="literal">root</code>. Part of the recovery process might involve re-creating from scratch various database files (shared memory regions, e.g.). Recovering as <code class="literal">root</code> will create those files such that they are owned by <code class="literal">root</code>, which means that even after you restore connectivity to your repository, regular users will be unable to access it.</p> <p>If the previous procedure, for some reason, does not successfully unwedge your repository, you should do two things. First, move your broken repository directory aside (perhaps by renaming it to something like <code class="filename">repos.BROKEN</code>) and then restore your latest backup of it. Then, send an email to the Subversion users mailing list (at <code class="email"><<a class="email" href="mailto:users@subversion.apache.org">users@subversion.apache.org</a>></code>) describing your problem in detail. Data integrity is an extremely high priority to the Subversion developers.</p> </div> <div class="sect2"> <div class="titlepage"> <div> <div> <h3 class="title"><a id="svn.reposadmin.maint.migrate"></a>Migrating Repository Data Elsewhere</h3> </div> </div> </div> <p>A Subversion filesystem has its data spread throughout files in the repository, in a fashion generally understood by (and of interest to) only the Subversion developers themselves. However, circumstances may arise that call for all, or some subset, of that data to be copied or moved into another repository.</p> <p>Subversion provides such functionality by way of <em class="firstterm">repository dump streams</em>. A repository dump stream (often referred to as a <span class="quote">“<span class="quote">dump file</span>”</span> when stored as a file on disk) is a portable, flat file format that describes the various revisions in your repository—what was changed, by whom, when, and so on. This dump stream is the primary mechanism used to marshal versioned history—in whole or in part, with or without modification—between repositories. And Subversion provides the tools necessary for creating and loading these dump streams: the <span class="command"><strong>svnadmin dump</strong></span> and <span class="command"><strong>svnadmin load</strong></span> subcommands, respectively, and the <span class="command"><strong>svnrdump</strong></span> program.</p> <div class="warning" style="margin-left: 0.5in; margin-right: 0.5in;"> <table border="0" summary="Warning"> <tr> <td rowspan="2" align="center" valign="top" width="25"> <img alt="[Warning]" src="images/warning.png" /> </td> <th align="left">Warning</th> </tr> <tr> <td align="left" valign="top"> <p>While the Subversion repository dump format contains human-readable portions and a familiar structure (it resembles an RFC 822 format, the same type of format used for most email), it is <span class="emphasis"><em>not</em></span> a plain-text file format. It is a binary file format, highly sensitive to meddling. For example, many text editors will corrupt the file by automatically converting line endings.</p> </td> </tr> </table> </div> <p>There are many reasons for dumping and loading Subversion repository data. Early in Subversion's life, the most common reason was due to the evolution of Subversion itself. As Subversion matured, there were times when changes made to the backend database schema caused compatibility issues with previous versions of the repository, so users had to dump their repository data using the previous version of Subversion and load it into a freshly created repository with the new version of Subversion. Now, these types of schema changes haven't occurred since Subversion's 1.0 release, and the Subversion developers promise not to force users to dump and load their repositories when upgrading between minor versions (such as from 1.3 to 1.4) of Subversion. But there are still other reasons for dumping and loading, including re-deploying a Berkeley DB repository on a new OS or CPU architecture, switching between the Berkeley DB and FSFS backends, or (as we'll cover later in this chapter in <a class="xref" href="svn.reposadmin.maint.html#svn.reposadmin.maint.filtering" title="Filtering Repository History">the section called “Filtering Repository History”</a>) purging versioned data from repository history.</p> <div class="note" style="margin-left: 0.5in; margin-right: 0.5in;"> <table border="0" summary="Note"> <tr> <td rowspan="2" align="center" valign="top" width="25"> <img alt="[Note]" src="images/note.png" /> </td> <th align="left">Note</th> </tr> <tr> <td align="left" valign="top"> <p>The Subversion repository dump format describes versioned repository changes only. It will not carry any information about uncommitted transactions, user locks on filesystem paths, repository or server configuration customizations (including hook scripts), and so on.</p> </td> </tr> </table> </div> <p>The Subversion repository dump format also enables conversion from a different storage mechanism or version control system altogether. Because the dump file format is, for the most part, human-readable, it should be relatively easy to describe generic sets of changes—each of which should be treated as a new revision—using this file format. In fact, the <span class="command"><strong>cvs2svn</strong></span> utility (see <a class="xref" href="svn.forcvs.convert.html" title="Converting a Repository from CVS to Subversion">the section called “Converting a Repository from CVS to Subversion”</a>) uses the dump format to represent the contents of a CVS repository so that those contents can be copied into a Subversion repository.</p> <p>For now, we'll concern ourselves only with migration of repository data between Subversion repositories, which we'll describe in detail in the sections which follow.</p> <div class="sect3"> <div class="titlepage"> <div> <div> <h4 class="title"><a id="svn.reposadmin.maint.migrate.svnadmin"></a>Repository data migration using svnadmin</h4> </div> </div> </div> <p>Whatever your reason for migrating repository history, using the <span class="command"><strong>svnadmin dump</strong></span> and <span class="command"><strong>svnadmin load</strong></span> subcommands is straightforward. <span class="command"><strong>svnadmin dump</strong></span> will output a range of repository revisions that are formatted using Subversion's custom filesystem dump format. The dump format is printed to the standard output stream, while informative messages are printed to the standard error stream. This allows you to redirect the output stream to a file while watching the status output in your terminal window. For example:</p> <div class="informalexample"> <pre class="screen"> $ svnlook youngest myrepos 26 $ svnadmin dump myrepos > dumpfile * Dumped revision 0. * Dumped revision 1. * Dumped revision 2. … * Dumped revision 25. * Dumped revision 26. </pre> </div> <p>At the end of the process, you will have a single file (<code class="filename">dumpfile</code> in the previous example) that contains all the data stored in your repository in the requested range of revisions. Note that <span class="command"><strong>svnadmin dump</strong></span> is reading revision trees from the repository just like any other <span class="quote">“<span class="quote">reader</span>”</span> process would (e.g., <span class="command"><strong>svn checkout</strong></span>), so it's safe to run this command at any time.</p> <p>The other subcommand in the pair, <span class="command"><strong>svnadmin load</strong></span>, parses the standard input stream as a Subversion repository dump file and effectively replays those dumped revisions into the target repository for that operation. It also gives informative feedback, this time using the standard output stream:</p> <div class="informalexample"> <pre class="screen"> $ svnadmin load newrepos < dumpfile <<< Started new txn, based on original revision 1 * adding path : A ... done. * adding path : A/B ... done. … ------- Committed new rev 1 (loaded from original rev 1) >>> <<< Started new txn, based on original revision 2 * editing path : A/mu ... done. * editing path : A/D/G/rho ... done. ------- Committed new rev 2 (loaded from original rev 2) >>> … <<< Started new txn, based on original revision 25 * editing path : A/D/gamma ... done. ------- Committed new rev 25 (loaded from original rev 25) >>> <<< Started new txn, based on original revision 26 * adding path : A/Z/zeta ... done. * editing path : A/mu ... done. ------- Committed new rev 26 (loaded from original rev 26) >>> </pre> </div> <p>The result of a load is new revisions added to a repository—the same thing you get by making commits against that repository from a regular Subversion client. Just as in a commit, you can use hook programs to perform actions before and after each of the commits made during a load process. By passing the <code class="option">--use-pre-commit-hook</code> and <code class="option">--use-post-commit-hook</code> options to <span class="command"><strong>svnadmin load</strong></span>, you can instruct Subversion to execute the pre-commit and post-commit hook programs, respectively, for each loaded revision. You might use these, for example, to ensure that loaded revisions pass through the same validation steps that regular commits pass through. Of course, you should use these options with care—if your post-commit hook sends emails to a mailing list for each new commit, you might not want to spew hundreds or thousands of commit emails in rapid succession at that list! You can read more about the use of hook scripts in <a class="xref" href="svn.reposadmin.create.html#svn.reposadmin.create.hooks" title="Implementing Repository Hooks">the section called “Implementing Repository Hooks”</a>.</p> <p>Note that because <span class="command"><strong>svnadmin</strong></span> uses standard input and output streams for the repository dump and load processes, people who are feeling especially saucy can try things such as this (perhaps even using different versions of <span class="command"><strong>svnadmin</strong></span> on each side of the pipe):</p> <div class="informalexample"> <pre class="screen"> $ svnadmin create newrepos $ svnadmin dump oldrepos | svnadmin load newrepos </pre> </div> <p>By default, the dump file will be quite large—much larger than the repository itself. That's because by default every version of every file is expressed as a full text in the dump file. This is the fastest and simplest behavior, and it's nice if you're piping the dump data directly into some other process (such as a compression program, filtering program, or loading process). But if you're creating a dump file for longer-term storage, you'll likely want to save disk space by using the <code class="option">--deltas</code> option. With this option, successive revisions of files will be output as compressed, binary differences—just as file revisions are stored in a repository. This option is slower, but it results in a dump file much closer in size to the original repository.</p> <p>We mentioned previously that <span class="command"><strong>svnadmin dump</strong></span> outputs a range of revisions. Use the <code class="option">--revision</code> (<code class="option">-r</code>) option to specify a single revision, or a range of revisions, to dump. If you omit this option, all the existing repository revisions will be dumped.</p> <div class="informalexample"> <pre class="screen"> $ svnadmin dump myrepos -r 23 > rev-23.dumpfile $ svnadmin dump myrepos -r 100:200 > revs-100-200.dumpfile </pre> </div> <p>As Subversion dumps each new revision, it outputs only enough information to allow a future loader to re-create that revision based on the previous one. In other words, for any given revision in the dump file, only the items that were changed in that revision will appear in the dump. The only exception to this rule is the first revision that is dumped with the current <span class="command"><strong>svnadmin dump</strong></span> command.</p> <p>By default, Subversion will not express the first dumped revision as merely differences to be applied to the previous revision. For one thing, there is no previous revision in the dump file! And second, Subversion cannot know the state of the repository into which the dump data will be loaded (if it ever is). To ensure that the output of each execution of <span class="command"><strong>svnadmin dump</strong></span> is self-sufficient, the first dumped revision is, by default, a full representation of every directory, file, and property in that revision of the repository.</p> <p>However, you can change this default behavior. If you add the <code class="option">--incremental</code> option when you dump your repository, <span class="command"><strong>svnadmin</strong></span> will compare the first dumped revision against the previous revision in the repository—the same way it treats every other revision that gets dumped. It will then output the first revision exactly as it does the rest of the revisions in the dump range—mentioning only the changes that occurred in that revision. The benefit of this is that you can create several small dump files that can be loaded in succession, instead of one large one, like so:</p> <div class="informalexample"> <pre class="screen"> $ svnadmin dump myrepos -r 0:1000 > dumpfile1 $ svnadmin dump myrepos -r 1001:2000 --incremental > dumpfile2 $ svnadmin dump myrepos -r 2001:3000 --incremental > dumpfile3 </pre> </div> <p>These dump files could be loaded into a new repository with the following command sequence:</p> <div class="informalexample"> <pre class="screen"> $ svnadmin load newrepos < dumpfile1 $ svnadmin load newrepos < dumpfile2 $ svnadmin load newrepos < dumpfile3 </pre> </div> <p>Another neat trick you can perform with this <code class="option">--incremental</code> option involves appending to an existing dump file a new range of dumped revisions. For example, you might have a post-commit hook that simply appends the repository dump of the single revision that triggered the hook. Or you might have a script that runs nightly to append dump file data for all the revisions that were added to the repository since the last time the script ran. Used like this, <span class="command"><strong>svnadmin dump</strong></span> can be one way to back up changes to your repository over time in case of a system crash or some other catastrophic event.</p> <p>The dump format can also be used to merge the contents of several different repositories into a single repository. By using the <code class="option">--parent-dir</code> option of <span class="command"><strong>svnadmin load</strong></span>, you can specify a new virtual root directory for the load process. That means if you have dump files for three repositories—say <code class="filename">calc-dumpfile</code>, <code class="filename">cal-dumpfile</code>, and <code class="filename">ss-dumpfile</code>—you can first create a new repository to hold them all:</p> <div class="informalexample"> <pre class="screen"> $ svnadmin create /var/svn/projects $ </pre> </div> <p>Then, make new directories in the repository that will encapsulate the contents of each of the three previous repositories:</p> <div class="informalexample"> <pre class="screen"> $ svn mkdir -m "Initial project roots" \ file:///var/svn/projects/calc \ file:///var/svn/projects/calendar \ file:///var/svn/projects/spreadsheet Committed revision 1. $ </pre> </div> <p>Lastly, load the individual dump files into their respective locations in the new repository:</p> <div class="informalexample"> <pre class="screen"> $ svnadmin load /var/svn/projects --parent-dir calc < calc-dumpfile … $ svnadmin load /var/svn/projects --parent-dir calendar < cal-dumpfile … $ svnadmin load /var/svn/projects --parent-dir spreadsheet < ss-dumpfile … $ </pre> </div> </div> <div class="sect3"> <div class="titlepage"> <div> <div> <h4 class="title"><a id="svn.reposadmin.maint.migrate.svnrdump"></a>Repository data migration using svnrdump</h4> </div> </div> </div> <p>In Subversion 1.7, <span class="command"><strong>svnrdump</strong></span> joined the set of stock Subversion tools. It offers fairly specialized functionality, essentially as a network-aware version of the <span class="command"><strong>svnadmin dump</strong></span> and <span class="command"><strong>svnadmin load</strong></span> commands which we discuss in depth in <a class="xref" href="svn.reposadmin.maint.html#svn.reposadmin.maint.migrate.svnadmin" title="Repository data migration using svnadmin">the section called “Repository data migration using svnadmin”</a>. <span class="command"><strong>svnrdump dump</strong></span> will generate a dump stream from a remote repository, spewing it to standard output; <span class="command"><strong>svnrdump load</strong></span> will read a dump stream from standard input and load it into a remote repository. Using <span class="command"><strong>svnrdump</strong></span>, you can generate incremental dumps just as you might with <span class="command"><strong>svnadmin dump</strong></span>. You can even dump a subtree of the repository—something that <span class="command"><strong>svnadmin dump</strong></span> cannot do.</p> <p>The primary difference is that instead of requiring direct access to the repository, <span class="command"><strong>svnrdump</strong></span> operates remotely, using the very same Repository Access (RA) protocols that the Subversion client does. As such, you might need to provide authentication credentials. Also, your remote interations are subject to any authorization limitations configured on the Subversion server.</p> <div class="note" style="margin-left: 0.5in; margin-right: 0.5in;"> <table border="0" summary="Note"> <tr> <td rowspan="2" align="center" valign="top" width="25"> <img alt="[Note]" src="images/note.png" /> </td> <th align="left">Note</th> </tr> <tr> <td align="left" valign="top"> <p><span class="command"><strong>svnrdump dump</strong></span> requires that the remote server be running Subversion 1.4 or newer. It currently generates dump streams only of the sort which are created when you pass the <code class="option">--deltas</code> option to <span class="command"><strong>svnadmin dump</strong></span>. This isn't interesting in the typical use-cases, but might impact specific types of custom transformations you might wish to apply to the resulting dump stream.</p> </td> </tr> </table> </div> <div class="note" style="margin-left: 0.5in; margin-right: 0.5in;"> <table border="0" summary="Note"> <tr> <td rowspan="2" align="center" valign="top" width="25"> <img alt="[Note]" src="images/note.png" /> </td> <th align="left">Note</th> </tr> <tr> <td align="left" valign="top"> <p>Because it modifies revision properties after committing new revisions, <span class="command"><strong>svnrdump load</strong></span> requires that the target repository have revision property changes enabled via the pre-revprop-change hook. See <a class="xref" href="svn.ref.reposhooks.pre-revprop-change.html" title="pre-revprop-change">pre-revprop-change</a> in <a class="xref" href="svn.ref.html" title="Chapter 9. Subversion Complete Reference">Chapter 9, <em>Subversion Complete Reference</em></a> for details.</p> </td> </tr> </table> </div> <p>As you might expect, you can use <span class="command"><strong>svnadmin</strong></span> and <span class="command"><strong>svnrdump</strong></span> in concert. You can, for example, use <span class="command"><strong>svnrdump dump</strong></span> to generate a dump stream from a remote repository, and pipe the results thereof through <span class="command"><strong>svnadmin load</strong></span> to copy all that repository history into a local repository. Or you can do the reverse, copying history from a local repository into a remote one.</p> <div class="tip" style="margin-left: 0.5in; margin-right: 0.5in;"> <table border="0" summary="Tip"> <tr> <td rowspan="2" align="center" valign="top" width="25"> <img alt="[Tip]" src="images/tip.png" /> </td> <th align="left">Tip</th> </tr> <tr> <td align="left" valign="top"> <p>By using <code class="literal">file://</code> URLs, <span class="command"><strong>svnrdump</strong></span> can also access local repositories, but it will be doing so via Subversion's Repository Access (RA) abstraction layer—you'll get better performance out of <span class="command"><strong>svnadmin</strong></span> in such situations.</p> </td> </tr> </table> </div> </div> </div> <div class="sect2"> <div class="titlepage"> <div> <div> <h3 class="title"><a id="svn.reposadmin.maint.filtering"></a>Filtering Repository History</h3> </div> </div> </div> <p>Since Subversion stores your versioned history using, at the very least, binary differencing algorithms and data compression (optionally in a completely opaque database system), attempting manual tweaks is unwise if not quite difficult, and at any rate strongly discouraged. And once data has been stored in your repository, Subversion generally doesn't provide an easy way to remove that data.<a href="#ftn.idm5718" class="footnote" id="idm5718"><sup class="footnote">[45]</sup></a> But inevitably, there will be times when you would like to manipulate the history of your repository. You might need to strip out all instances of a file that was accidentally added to the repository (and shouldn't be there for whatever reason).<a href="#ftn.idm5720" class="footnote" id="idm5720"><sup class="footnote">[46]</sup></a> Or, perhaps you have multiple projects sharing a single repository, and you decide to split them up into their own repositories. To accomplish tasks such as these, administrators need a more manageable and malleable representation of the data in their repositories—the Subversion repository dump format.</p> <p>As we described earlier in <a class="xref" href="svn.reposadmin.maint.html#svn.reposadmin.maint.migrate" title="Migrating Repository Data Elsewhere">the section called “Migrating Repository Data Elsewhere”</a>, the Subversion repository dump format is a human-readable representation of the changes that you've made to your versioned data over time. Use the <span class="command"><strong>svnadmin dump</strong></span> or <span class="command"><strong>svnrdump dump</strong></span> command to generate the dump data, and <span class="command"><strong>svnadmin load</strong></span> or <span class="command"><strong>svnrdump load</strong></span> to populate a new repository with it. The great thing about the human-readability aspect of the dump format is that, if you aren't careless about it, you can manually inspect and modify it. Of course, the downside is that if you have three years' worth of repository activity encapsulated in what is likely to be a very large dump file, it could take you a long, long time to manually inspect and modify it.</p> <p>That's where <span class="command"><strong>svndumpfilter</strong></span> becomes useful. This program acts as a path-based filter for repository dump streams. Simply give it either a list of paths you wish to keep or a list of paths you wish to not keep, and then pipe your repository dump data through this filter. The result will be a modified stream of dump data that contains only the versioned paths you (explicitly or implicitly) requested.</p> <p>Let's look at a realistic example of how you might use this program. Earlier in this chapter (see <a class="xref" href="svn.reposadmin.planning.html#svn.reposadmin.projects.chooselayout" title="Planning Your Repository Organization">the section called “Planning Your Repository Organization”</a>), we discussed the process of deciding how to choose a layout for the data in your repositories—using one repository per project or combining them, arranging stuff within your repository, and so on. But sometimes after new revisions start flying in, you rethink your layout and would like to make some changes. A common change is the decision to move multiple projects that are sharing a single repository into separate repositories for each project.</p> <p>Our imaginary repository contains three projects: <code class="literal">calc</code>, <code class="literal">calendar</code>, and <code class="literal">spreadsheet</code>. They have been living side-by-side in a layout like this:</p> <div class="informalexample"> <div class="literallayout"> <p><br /> /<br /> calc/<br /> trunk/<br /> branches/<br /> tags/<br /> calendar/<br /> trunk/<br /> branches/<br /> tags/<br /> spreadsheet/<br /> trunk/<br /> branches/<br /> tags/<br /> </p> </div> </div> <p>To get these three projects into their own repositories, we first dump the whole repository:</p> <div class="informalexample"> <pre class="screen"> $ svnadmin dump /var/svn/repos > repos-dumpfile * Dumped revision 0. * Dumped revision 1. * Dumped revision 2. * Dumped revision 3. … $ </pre> </div> <p>Next, run that dump file through the filter, each time including only one of our top-level directories. This results in three new dump files:</p> <div class="informalexample"> <pre class="screen"> $ svndumpfilter include calc < repos-dumpfile > calc-dumpfile … $ svndumpfilter include calendar < repos-dumpfile > cal-dumpfile … $ svndumpfilter include spreadsheet < repos-dumpfile > ss-dumpfile … $ </pre> </div> <p>At this point, you have to make a decision. Each of your dump files will create a valid repository, but will preserve the paths exactly as they were in the original repository. This means that even though you would have a repository solely for your <code class="literal">calc</code> project, that repository would still have a top-level directory named <code class="filename">calc</code>. If you want your <code class="filename">trunk</code>, <code class="filename">tags</code>, and <code class="filename">branches</code> directories to live in the root of your repository, you might wish to edit your dump files, tweaking the <code class="literal">Node-path</code> and <code class="literal">Node-copyfrom-path</code> headers so that they no longer have that first <code class="filename">calc/</code> path component. Also, you'll want to remove the section of dump data that creates the <code class="filename">calc</code> directory. It will look something like the following:</p> <div class="informalexample"> <pre class="programlisting"> Node-path: calc Node-action: add Node-kind: dir Content-length: 0 </pre> </div> <div class="warning" style="margin-left: 0.5in; margin-right: 0.5in;"> <table border="0" summary="Warning"> <tr> <td rowspan="2" align="center" valign="top" width="25"> <img alt="[Warning]" src="images/warning.png" /> </td> <th align="left">Warning</th> </tr> <tr> <td align="left" valign="top"> <p>If you do plan on manually editing the dump file to remove a top-level directory, make sure your editor is not set to automatically convert end-of-line characters to the native format (e.g., <code class="literal">\r\n</code> to <code class="literal">\n</code>), as the content will then not agree with the metadata. This will render the dump file useless.</p> </td> </tr> </table> </div> <p>All that remains now is to create your three new repositories, and load each dump file into the right repository, ignoring the UUID found in the dump stream:</p> <div class="informalexample"> <pre class="screen"> $ svnadmin create calc $ svnadmin load --ignore-uuid calc < calc-dumpfile <<< Started new transaction, based on original revision 1 * adding path : Makefile ... done. * adding path : button.c ... done. … $ svnadmin create calendar $ svnadmin load --ignore-uuid calendar < cal-dumpfile <<< Started new transaction, based on original revision 1 * adding path : Makefile ... done. * adding path : cal.c ... done. … $ svnadmin create spreadsheet $ svnadmin load --ignore-uuid spreadsheet < ss-dumpfile <<< Started new transaction, based on original revision 1 * adding path : Makefile ... done. * adding path : ss.c ... done. … $ </pre> </div> <p>Both of <span class="command"><strong>svndumpfilter</strong></span>'s subcommands accept options for deciding how to deal with <span class="quote">“<span class="quote">empty</span>”</span> revisions. If a given revision contains only changes to paths that were filtered out, that now-empty revision could be considered uninteresting or even unwanted. So to give the user control over what to do with those revisions, <span class="command"><strong>svndumpfilter</strong></span> provides the following command-line options:</p> <div class="variablelist"> <dl class="variablelist"> <dt> <span class="term"> <code class="option">--drop-empty-revs</code> </span> </dt> <dd> <p>Do not generate empty revisions at all—just omit them.</p> </dd> <dt> <span class="term"> <code class="option">--renumber-revs</code> </span> </dt> <dd> <p>If empty revisions are dropped (using the <code class="option">--drop-empty-revs</code> option), change the revision numbers of the remaining revisions so that there are no gaps in the numeric sequence.</p> </dd> <dt> <span class="term"> <code class="option">--preserve-revprops</code> </span> </dt> <dd> <p>If empty revisions are not dropped, preserve the revision properties (log message, author, date, custom properties, etc.) for those empty revisions. Otherwise, empty revisions will contain only the original datestamp, and a generated log message that indicates that this revision was emptied by <span class="command"><strong>svndumpfilter</strong></span>.</p> </dd> </dl> </div> <p>While <span class="command"><strong>svndumpfilter</strong></span> can be very useful and a huge timesaver, there are unfortunately a couple of gotchas. First, this utility is overly sensitive to path semantics. Pay attention to whether paths in your dump file are specified with or without leading slashes. You'll want to look at the <code class="literal">Node-path</code> and <code class="literal">Node-copyfrom-path</code> headers.</p> <div class="informalexample"> <pre class="programlisting"> … Node-path: spreadsheet/Makefile … </pre> </div> <p>If the paths have leading slashes, you should include leading slashes in the paths you pass to <span class="command"><strong>svndumpfilter include</strong></span> and <span class="command"><strong>svndumpfilter exclude</strong></span> (and if they don't, you shouldn't). Further, if your dump file has an inconsistent usage of leading slashes for some reason,<a href="#ftn.idm5795" class="footnote" id="idm5795"><sup class="footnote">[47]</sup></a> you should probably normalize those paths so that they all have, or all lack, leading slashes.</p> <p>Also, copied paths can give you some trouble. Subversion supports copy operations in the repository, where a new path is created by copying some already existing path. It is possible that at some point in the lifetime of your repository, you might have copied a file or directory from some location that <span class="command"><strong>svndumpfilter</strong></span> is excluding, to a location that it is including. To make the dump data self-sufficient, <span class="command"><strong>svndumpfilter</strong></span> needs to still show the addition of the new path—including the contents of any files created by the copy—and not represent that addition as a copy from a source that won't exist in your filtered dump data stream. But because the Subversion repository dump format shows only what was changed in each revision, the contents of the copy source might not be readily available. If you suspect that you have any copies of this sort in your repository, you might want to rethink your set of included/excluded paths, perhaps including the paths that served as sources of your troublesome copy operations, too.</p> <p>Finally, <span class="command"><strong>svndumpfilter</strong></span> takes path filtering quite literally. If you are trying to copy the history of a project rooted at <code class="filename">trunk/my-project</code> and move it into a repository of its own, you would, of course, use the <span class="command"><strong>svndumpfilter include</strong></span> command to keep all the changes in and under <code class="filename">trunk/my-project</code>. But the resultant dump file makes no assumptions about the repository into which you plan to load this data. Specifically, the dump data might begin with the revision that added the <code class="filename">trunk/my-project</code> directory, but it will <span class="emphasis"><em>not</em></span> contain directives that would create the <code class="filename">trunk</code> directory itself (because <code class="filename">trunk</code> doesn't match the include filter). You'll need to make sure that any directories that the new dump stream expects to exist actually do exist in the target repository before trying to load the stream into that repository.</p> </div> <div class="sect2"> <div class="titlepage"> <div> <div> <h3 class="title"><a id="svn.reposadmin.maint.replication"></a>Repository Replication</h3> </div> </div> </div> <p>There are several scenarios in which it is quite handy to have a Subversion repository whose version history is exactly the same as some other repository's. Perhaps the most obvious one is the maintenance of a simple backup repository, used when the primary repository has become inaccessible due to a hardware failure, network outage, or other such annoyance. Other scenarios include deploying mirror repositories to distribute heavy Subversion load across multiple servers, use as a soft-upgrade mechanism, and so on.</p> <p>Subversion provides a program for managing scenarios such as these. <span class="command"><strong>svnsync</strong></span> works by essentially asking the Subversion server to <span class="quote">“<span class="quote">replay</span>”</span> revisions, one at a time. It then uses that revision information to mimic a commit of the same to another repository. Neither repository needs to be locally accessible to the machine on which <span class="command"><strong>svnsync</strong></span> is running—its parameters are repository URLs, and it does all its work through Subversion's Repository Access (RA) interfaces. All it requires is read access to the source repository and read/write access to the destination repository.</p> <div class="note" style="margin-left: 0.5in; margin-right: 0.5in;"> <table border="0" summary="Note"> <tr> <td rowspan="2" align="center" valign="top" width="25"> <img alt="[Note]" src="images/note.png" /> </td> <th align="left">Note</th> </tr> <tr> <td align="left" valign="top"> <p>When using <span class="command"><strong>svnsync</strong></span> against a remote source repository, the Subversion server for that repository must be running Subversion version 1.4 or later.</p> </td> </tr> </table> </div> <div class="sect3"> <div class="titlepage"> <div> <div> <h4 class="title"><a id="svn.reposadmin.maint.replication.svnsync"></a>Replication with svnsync</h4> </div> </div> </div> <p>Assuming you already have a source repository that you'd like to mirror, the next thing you need is a target repository that will actually serve as that mirror. This target repository can use either of the available filesystem data-store backends (see <a class="xref" href="svn.reposadmin.planning.html#svn.reposadmin.basics.backends" title="Choosing a Data Store">the section called “Choosing a Data Store”</a>)—Subversion's abstraction layers ensure that such details don't matter. But by default, it must not yet have any version history in it. (We'll discuss an exception to this later in this section.)</p> <p>The protocol that <span class="command"><strong>svnsync</strong></span> uses to communicate revision information is highly sensitive to mismatches between the versioned histories contained in the source and target repositories. For this reason, while <span class="command"><strong>svnsync</strong></span> cannot <span class="emphasis"><em>demand</em></span> that the target repository be read-only,<a href="#ftn.idm5828" class="footnote" id="idm5828"><sup class="footnote">[48]</sup></a> allowing the revision history in the target repository to change by any mechanism other than the mirroring process is a recipe for disaster.</p> <div class="warning" style="margin-left: 0.5in; margin-right: 0.5in;"> <table border="0" summary="Warning"> <tr> <td rowspan="2" align="center" valign="top" width="25"> <img alt="[Warning]" src="images/warning.png" /> </td> <th align="left">Warning</th> </tr> <tr> <td align="left" valign="top"> <p>Do <span class="emphasis"><em>not</em></span> modify a mirror repository in such a way as to cause its version history to deviate from that of the repository it mirrors. The only commits and revision property modifications that ever occur on that mirror repository should be those performed by the <span class="command"><strong>svnsync</strong></span> tool.</p> </td> </tr> </table> </div> <p>Another requirement of the target repository is that the <span class="command"><strong>svnsync</strong></span> process be allowed to modify revision properties. Because <span class="command"><strong>svnsync</strong></span> works within the framework of that repository's hook system, the default state of the repository (which is to disallow revision property changes; see <a class="xref" href="svn.ref.reposhooks.pre-revprop-change.html" title="pre-revprop-change">pre-revprop-change</a> in <a class="xref" href="svn.ref.html" title="Chapter 9. Subversion Complete Reference">Chapter 9, <em>Subversion Complete Reference</em></a>) is insufficient. You'll need to explicitly implement the pre-revprop-change hook, and your script must allow <span class="command"><strong>svnsync</strong></span> to set and change revision properties. With those provisions in place, you are ready to start mirroring repository revisions.</p> <div class="tip" style="margin-left: 0.5in; margin-right: 0.5in;"> <table border="0" summary="Tip"> <tr> <td rowspan="2" align="center" valign="top" width="25"> <img alt="[Tip]" src="images/tip.png" /> </td> <th align="left">Tip</th> </tr> <tr> <td align="left" valign="top"> <p>It's a good idea to implement authorization measures that allow your repository replication process to perform its tasks while preventing other users from modifying the contents of your mirror repository at all.</p> </td> </tr> </table> </div> <p>Let's walk through the use of <span class="command"><strong>svnsync</strong></span> in a somewhat typical mirroring scenario. We'll pepper this discourse with practical recommendations, which you are free to disregard if they aren't required by or suitable for your environment.</p> <p>We will be mirroring the public Subversion repository which houses the source code for this very book and exposing that mirror publicly on the Internet, hosted on a different machine than the one on which the original Subversion source code repository lives. This remote host has a global configuration that permits anonymous users to read the contents of repositories on the host, but requires users to authenticate to modify those repositories. (Please forgive us for glossing over the details of Subversion server configuration for the moment—those are covered thoroughly in <a class="xref" href="svn.serverconfig.html" title="Chapter 6. Server Configuration">Chapter 6, <em>Server Configuration</em></a>.) And for no other reason than that it makes for a more interesting example, we'll be driving the replication process from a third machine—the one that we currently find ourselves using.</p> <p>First, we'll create the repository which will be our mirror. This and the next couple of steps do require shell access to the machine on which the mirror repository will live. Once the repository is all configured, though, we shouldn't need to touch it directly again.</p> <div class="informalexample"> <pre class="screen"> $ ssh admin@svn.example.com "svnadmin create /var/svn/svn-mirror" admin@svn.example.com's password: ******** $ </pre> </div> <p>At this point, we have our repository, and due to our server's configuration, that repository is now <span class="quote">“<span class="quote">live</span>”</span> on the Internet. Now, because we don't want anything modifying the repository except our replication process, we need a way to distinguish that process from other would-be committers. To do so, we use a dedicated username for our process. Only commits and revision property modifications performed by the special username <code class="literal">syncuser</code> will be allowed.</p> <p>We'll use the repository's hook system both to allow the replication process to do what it needs to do and to enforce that only it is doing those things. We accomplish this by implementing two of the repository event hooks—pre-revprop-change and start-commit. Our pre-revprop-change hook script is found in <a class="xref" href="svn.reposadmin.maint.html#svn.reposadmin.maint.replication.pre-revprop-change" title="Example 5.2. Mirror repository's pre-revprop-change hook script">Example 5.2, “Mirror repository's pre-revprop-change hook script”</a>, and basically verifies that the user attempting the property changes is our <code class="literal">syncuser</code> user. If so, the change is allowed; otherwise, it is denied.</p> <div class="example"> <a id="svn.reposadmin.maint.replication.pre-revprop-change"></a> <p class="title"> <strong>Example 5.2. Mirror repository's pre-revprop-change hook script</strong> </p> <div class="example-contents"> <pre class="programlisting"> #!/bin/sh USER="$3" if [ "$USER" = "syncuser" ]; then exit 0; fi echo "Only the syncuser user may change revision properties" >&2 exit 1 </pre> </div> </div> <br class="example-break" /> <p>That covers revision property changes. Now we need to ensure that only the <code class="literal">syncuser</code> user is permitted to commit new revisions to the repository. We do this using a start-commit hook script such as the one in <a class="xref" href="svn.reposadmin.maint.html#svn.reposadmin.maint.replication.start-commit" title="Example 5.3. Mirror repository's start-commit hook script">Example 5.3, “Mirror repository's start-commit hook script”</a>.</p> <div class="example"> <a id="svn.reposadmin.maint.replication.start-commit"></a> <p class="title"> <strong>Example 5.3. Mirror repository's start-commit hook script</strong> </p> <div class="example-contents"> <pre class="programlisting"> #!/bin/sh USER="$2" if [ "$USER" = "syncuser" ]; then exit 0; fi echo "Only the syncuser user may commit new revisions" >&2 exit 1 </pre> </div> </div> <br class="example-break" /> <p>After installing our hook scripts and ensuring that they are executable by the Subversion server, we're finished with the setup of the mirror repository. Now, we get to actually do the mirroring.</p> <p>The first thing we need to do with <span class="command"><strong>svnsync</strong></span> is to register in our target repository the fact that it will be a mirror of the source repository. We do this using the <span class="command"><strong>svnsync initialize</strong></span> subcommand. The URLs we provide point to the root directories of the target and source repositories, respectively. In Subversion 1.4, this is required—only full mirroring of repositories is permitted. Beginning with Subversion 1.5, though, you can use <span class="command"><strong>svnsync</strong></span> to mirror only some subtree of the repository, too.</p> <div class="informalexample"> <pre class="screen"> $ svnsync help init initialize (init): usage: svnsync initialize DEST_URL SOURCE_URL Initialize a destination repository for synchronization from another repository. … $ svnsync initialize http://svn.example.com/svn-mirror \ https://svn.code.sf.net/p/svnbook/source \ --sync-username syncuser --sync-password syncpass Copied properties for revision 0 (svn:sync-* properties skipped). NOTE: Normalized svn:* properties to LF line endings (1 rev-props, 0 node-props). $ </pre> </div> <p>Our target repository will now remember that it is a mirror of the public Subversion source code repository. Notice that we provided a username and password as arguments to <span class="command"><strong>svnsync</strong></span>—that was required by the pre-revprop-change hook on our mirror repository.</p> <div class="note" style="margin-left: 0.5in; margin-right: 0.5in;"> <table border="0" summary="Note"> <tr> <td rowspan="2" align="center" valign="top" width="25"> <img alt="[Note]" src="images/note.png" /> </td> <th align="left">Note</th> </tr> <tr> <td align="left" valign="top"> <p>In Subversion 1.4, the values given to <span class="command"><strong>svnsync</strong></span>'s <code class="option">--username</code> and <code class="option">--password</code> command-line options were used for authentication against both the source and destination repositories. This caused problems when a user's credentials weren't exactly the same for both repositories, especially when running in noninteractive mode (with the <code class="option">--non-interactive</code> option). This was fixed in Subversion 1.5 with the introduction of two new pairs of options. Use <code class="option">--source-username</code> and <code class="option">--source-password</code> to provide authentication credentials for the source repository; use <code class="option">--sync-username</code> and <code class="option">--sync-password</code> to provide credentials for the destination repository. (The old <code class="option">--username</code> and <code class="option">--password</code> options still exist for compatibility, but we advise against using them.)</p> </td> </tr> </table> </div> <p>And now comes the fun part. With a single subcommand, we can tell <span class="command"><strong>svnsync</strong></span> to copy all the as-yet-unmirrored revisions from the source repository to the target.<a href="#ftn.idm5888" class="footnote" id="idm5888"><sup class="footnote">[49]</sup></a> The <span class="command"><strong>svnsync synchronize</strong></span> subcommand will peek into the special revision properties previously stored on the target repository and determine how much of the source repository has been previously mirrored—in this case, the most recently mirrored revision is r0. Then it will query the source repository and determine what the latest revision in that repository is. Finally, it asks the source repository's server to start replaying all the revisions between 0 and that latest revision. As <span class="command"><strong>svnsync</strong></span> gets the resultant response from the source repository's server, it begins forwarding those revisions to the target repository's server as new commits.</p> <div class="informalexample"> <pre class="screen"> $ svnsync help synchronize synchronize (sync): usage: svnsync synchronize DEST_URL [SOURCE_URL] Transfer all pending revisions to the destination from the source with which it was initialized. … $ svnsync synchronize http://svn.example.com/svn-mirror \ https://svn.code.sf.net/p/svnbook/source Committed revision 1. Copied properties for revision 1. Committed revision 2. Copied properties for revision 2. Transmitting file data . Committed revision 3. Copied properties for revision 3. … Transmitting file data . Committed revision 4063. Copied properties for revision 4063. Transmitting file data . Committed revision 4064. Copied properties for revision 4064. Transmitting file data .... Committed revision 4065. Copied properties for revision 4065. $ </pre> </div> <p>Of particular interest here is that for each mirrored revision, there is first a commit of that revision to the target repository, and then property changes follow. This two-phase replication is required because the initial commit is performed by (and attributed to) the user <code class="literal">syncuser</code> and is datestamped with the time as of that revision's creation. <span class="command"><strong>svnsync</strong></span> has to follow up with an immediate series of property modifications that copy into the target repository all the original revision properties found for that revision in the source repository, which also has the effect of fixing the author and datestamp of the revision to match that of the source repository.</p> <p>Also noteworthy is that <span class="command"><strong>svnsync</strong></span> performs careful bookkeeping that allows it to be safely interrupted and restarted without ruining the integrity of the mirrored data. If a network glitch occurs while mirroring a repository, simply repeat the <span class="command"><strong>svnsync synchronize</strong></span> command, and it will happily pick up right where it left off. In fact, as new revisions appear in the source repository, this is exactly what you do to keep your mirror up to date.</p> <div class="warning" style="margin-left: 0.5in; margin-right: 0.5in;"> <table border="0" summary="Warning"> <tr> <td rowspan="2" align="center" valign="top" width="25"> <img alt="[Warning]" src="images/warning.png" /> </td> <th align="left">Warning</th> </tr> <tr> <td align="left" valign="top"> <p>As part of its bookkeeping, <span class="command"><strong>svnsync</strong></span> records in the mirror repository the URL with which the mirror was initialized. Because of this, invocations of <span class="command"><strong>svnsync</strong></span> which follow the initialization step do not <span class="emphasis"><em>require</em></span> that you provide the source URL on the command line again. However, for security purposes, we recommend that you continue to do so. Depending on how it is deployed, it may not be safe for <span class="command"><strong>svnsync</strong></span> to trust the source URL which it retrieves from the mirror repository, and from which it pulls versioned data.</p> </td> </tr> </table> </div> <div class="sidebar"> <div class="titlepage"> <div> <div> <p class="title"> <strong>svnsync Bookkeeping</strong> </p> </div> </div> </div> <p><span class="command"><strong>svnsync</strong></span> needs to be able to set and modify revision properties on the mirror repository because those properties are part of the data it is tasked with mirroring. As those properties change in the source repository, those changes need to be reflected in the mirror repository, too. But <span class="command"><strong>svnsync</strong></span> also uses a set of custom revision properties—stored in revision 0 of the mirror repository—for its own internal bookkeeping. These properties contain information such as the URL and UUID of the source repository, plus some additional state-tracking information.</p> <p>One of those pieces of state-tracking information is a flag that essentially just means <span class="quote">“<span class="quote">there's a synchronization in progress right now.</span>”</span> This is used to prevent multiple <span class="command"><strong>svnsync</strong></span> processes from colliding with each other while trying to mirror data to the same destination repository. Now, generally you won't need to pay any attention whatsoever to <span class="emphasis"><em>any</em></span> of these special properties (all of which begin with the prefix <code class="literal">svn:sync-</code>). Occasionally, though, if a synchronization fails unexpectedly, Subversion never has a chance to remove this particular state flag. This causes all future synchronization attempts to fail because it appears that a synchronization is still in progress when, in fact, none is. Fortunately, recovering from this situation is easy to do. In Subversion 1.7, you can use the newly introduced <code class="option">--steal-lock</code> option with <span class="command"><strong>svnsync</strong></span>'s commands. In previous Subversion versions, you need only to remove the <code class="literal">svn:sync-lock</code> property which serves as this flag from revision 0 of the mirror repository:</p> <div class="informalexample"> <pre class="screen"> $ svn propdel --revprop -r0 svn:sync-lock http://svn.example.com/svn-mirror property 'svn:sync-lock' deleted from repository revision 0 $ </pre> </div> <p>Also, <span class="command"><strong>svnsync</strong></span> stores the source repository URL provided at mirror initialization time in a bookkeeping property on the mirror repository. Future synchronization operations against that mirror which omit the source URL at the command line will consult the special <code class="literal">svn:sync-from-url</code> property stored on the mirror itself to know where to synchronize from. This value is used literally by the synchronization process, though. Be wary of using non-fully-qualified domain names (such as referring to <code class="literal">svnbook.red-bean.com</code> as simply <code class="literal">svnbook</code> because that happens to work when you are connected directly to the <code class="literal">red-bean.com</code> network), domain names which don't resolve or resolve differently depending on where you happen to be operating from, or IP addresses (which can change over time). But here again, if you need an existing mirror to start referring to a different URL for the same source repository, you can change the bookkeeping property which houses that information. Users of Subversion 1.7 or better can use <span class="command"><strong>svnsync init --allow-non-empty</strong></span> to reinitialize their mirrors with new source URL:</p> <div class="informalexample"> <pre class="screen"> $ svnsync initialize --allow-non-empty http://svn.example.com/svn-mirror \ <em class="replaceable"><code>NEW-SOURCE-URL</code></em> Copied properties for revision 4065. $ </pre> </div> <p>If you are running an older version of Subversion, you'll need to manually tweak the <code class="literal">svn:sync-from-url</code> bookkeeping property:</p> <div class="informalexample"> <pre class="screen"> $ svn propset --revprop -r0 svn:sync-from-url <em class="replaceable"><code>NEW-SOURCE-URL</code></em> \ http://svn.example.com/svn-mirror property 'svn:sync-from-url' set on repository revision 0 $ </pre> </div> <p>Another interesting thing about these special bookkeeping properties is that <span class="command"><strong>svnsync</strong></span> will not attempt to mirror any of those properties when they are found in the source repository. The reason is probably obvious, but basically boils down to <span class="command"><strong>svnsync</strong></span> not being able to distinguish the special properties it has merely copied from the source repository from those it needs to consult and maintain for its own bookkeeping needs. This situation could occur if, for example, you were maintaining a mirror of a mirror of a third repository. When <span class="command"><strong>svnsync</strong></span> sees its own special properties in revision 0 of the source repository, it simply ignores them.</p> <p>An <span class="command"><strong>svnsync info</strong></span> subcommand was added in Subversion 1.6 to easily display the special bookkeeping properties in the destination repository.</p> <div class="informalexample"> <pre class="screen"> $ svnsync help info info: usage: svnsync info DEST_URL Print information about the synchronization destination repository located at DEST_URL. … $ svnsync info http://svn.example.com/svn-mirror Source URL: https://svn.code.sf.net/p/svnbook/source Source Repository UUID: 931749d0-5854-0410-9456-f14be4d6b398 Last Merged Revision: 4065 $ </pre> </div> </div> <p>There is, however, one bit of inelegance in the process. Because Subversion revision properties can be changed at any time throughout the lifetime of the repository, and because they don't leave an audit trail that indicates when they were changed, replication processes have to pay special attention to them. If you've already mirrored the first 15 revisions of a repository and someone then changes a revision property on revision 12, <span class="command"><strong>svnsync</strong></span> won't know to go back and patch up its copy of revision 12. You'll need to tell it to do so manually by using (or with some additional tooling around) the <span class="command"><strong>svnsync copy-revprops</strong></span> subcommand, which simply rereplicates all the revision properties for a particular revision or range thereof.</p> <div class="informalexample"> <pre class="screen"> $ svnsync help copy-revprops copy-revprops: usage: 1. svnsync copy-revprops DEST_URL [SOURCE_URL] 2. svnsync copy-revprops DEST_URL REV[:REV2] … $ svnsync copy-revprops http://svn.example.com/svn-mirror 12 Copied properties for revision 12. $ </pre> </div> <p>That's repository replication via <span class="command"><strong>svnsync</strong></span> in a nutshell. You'll likely want some automation around such a process. For example, while our example was a pull-and-push setup, you might wish to have your primary repository push changes to one or more blessed mirrors as part of its post-commit and post-revprop-change hook implementations. This would enable the mirror to be up to date in as near to real time as is likely possible.</p> </div> <div class="sect3"> <div class="titlepage"> <div> <div> <h4 class="title"><a id="svn.reposadmin.maint.replication.svnsync-partial"></a>Partial replication with svnsync</h4> </div> </div> </div> <p><span class="command"><strong>svnsync</strong></span> isn't limited to full copies of everything which lives in a repository. It can handle various shades of partial replication, too. For example, while it isn't very commonplace to do so, <span class="command"><strong>svnsync</strong></span> does gracefully mirror repositories in which the user as whom it authenticates has only partial read access. It simply copies only the bits of the repository that it is permitted to see. Obviously, such a mirror is not useful as a backup solution.</p> <p>As of Subversion 1.5, <span class="command"><strong>svnsync</strong></span> also has the ability to mirror a subset of a repository rather than the whole thing. The process of setting up and maintaining such a mirror is exactly the same as when mirroring a whole repository, except that instead of specifying the source repository's root URL when running <span class="command"><strong>svnsync init</strong></span>, you specify the URL of some subdirectory within that repository. Synchronization to that mirror will now copy only the bits that changed under that source repository subdirectory. There are some limitations to this support, though. First, you can't mirror multiple disjoint subdirectories of the source repository into a single mirror repository—you'd need to instead mirror some parent directory that is common to both. Second, the filtering logic is entirely path-based, so if the subdirectory you are mirroring was renamed at some point in the past, your mirror would contain only the revisions since the directory appeared at the URL you specified. And likewise, if the source subdirectory is renamed in the future, your synchronization processes will stop mirroring data at the point that the source URL you specified is no longer valid.</p> </div> <div class="sect3"> <div class="titlepage"> <div> <div> <h4 class="title"><a id="svn.reposadmin.maint.replication.svnsync-init-nonempty"></a>A quick trick for mirror creation</h4> </div> </div> </div> <p>We mentioned previously the cost of setting up an initial mirror of an existing repository. For many folks, the sheer cost of transmitting thousands—or millions—of revisions of history to a new mirror repository via <span class="command"><strong>svnsync</strong></span> is a show-stopper. Fortunately, Subversion 1.7 provides a workaround by way of a new <code class="option">--allow-non-empty</code> option to <span class="command"><strong>svnsync initialize</strong></span>. This option allows you to initialize one repository as a mirror of another while bypassing the verification that the to-be-initialized mirror has no version history present in it. Per our previous warnings about the sensitivity of this whole replication process, you should rightly discern that this is an option to be used only with great caution. But it's wonderfully handy when you have administrative access to the source repository, where you can simply make a physical copy of the repository and then initialize that copy as a new mirror:</p> <div class="informalexample"> <pre class="screen"> $ svnadmin hotcopy /path/to/repos /path/to/mirror-repos $ ### create /path/to/mirror-repos/hooks/pre-revprop-change $ svnsync initialize file:///path/to/mirror-repos \ file:///path/to/repos svnsync: E000022: Destination repository already contains revision history; co nsider using --allow-non-empty if the repository's revisions are known to mirr or their respective revisions in the source repository $ svnsync initialize --allow-non-empty file:///path/to/mirror-repos \ file:///path/to/repos Copied properties for revision 32042. $ </pre> </div> <p>Admins who are running a version of Subversion prior to 1.7 (and thus do not have access to <span class="command"><strong>svnsync initialize</strong></span>'s <code class="option">--allow-non-empty</code> feature) can accomplish effectively the same thing that that feature does through <span class="emphasis"><em>careful</em></span> manipulation of the r0 revision properties on the copy of the repository which is slated to become a mirror of the original. Use <span class="command"><strong>svnadmin setrevprop</strong></span> to create the same bookkeeping properties that <span class="command"><strong>svnsync</strong></span> would have created there.</p> </div> <div class="sect3"> <div class="titlepage"> <div> <div> <h4 class="title"><a id="svn.reposadmin.maint.replication.wrapup"></a>Replication wrap-up</h4> </div> </div> </div> <p>We've discussed a couple of ways to replicate revision history from one repository to another. So let's look now at the user end of these operations. How does replication and the various situations which call for it affect Subversion clients?</p> <p>As far as user interaction with repositories and mirrors goes, it <span class="emphasis"><em>is</em></span> possible to have a single working copy that interacts with both, but you'll have to jump through some hoops to make it happen. First, you need to ensure that both the primary and mirror repositories have the same repository UUID (which is not the case by default). See <a class="xref" href="svn.reposadmin.maint.html#svn.reposadmin.maint.uuids" title="Managing Repository UUIDs">the section called “Managing Repository UUIDs”</a> later in this chapter for more about this.</p> <p>Once the two repositories have the same UUID, you can use <span class="command"><strong>svn relocate</strong></span> to point your working copy to whichever of the repositories you wish to operate against, a process that is described in <a class="xref" href="svn.ref.svn.c.relocate.html" title="svn relocate">svn relocate</a> in <a class="xref" href="svn.ref.html" title="Chapter 9. Subversion Complete Reference">Chapter 9, <em>Subversion Complete Reference</em></a>. There is a possible danger here, though, in that if the primary and mirror repositories aren't in close synchronization, a working copy up to date with, and pointing to, the primary repository will, if relocated to point to an out-of-date mirror, become confused about the apparent sudden loss of revisions it fully expects to be present, and it will throw errors to that effect. If this occurs, you can relocate your working copy back to the primary repository and then either wait until the mirror repository is up to date, or backdate your working copy to a revision you know is present in the sync repository, and then retry the relocation.</p> <p>Finally, be aware that the revision-based replication provided by <span class="command"><strong>svnsync</strong></span> is only that—replication of revisions. Only the kinds of information carried by the Subversion repository dump file format are available for replication. As such, tools such as <span class="command"><strong>svnsync</strong></span> (and <span class="command"><strong>svnrdump</strong></span>, which we discuss in <a class="xref" href="svn.reposadmin.maint.html#svn.reposadmin.maint.migrate.svnrdump" title="Repository data migration using svnrdump">the section called “Repository data migration using svnrdump”</a>) are limited in ways similar to that of the repository dump stream. They do not include in their replicated information such things as the hook implementations, repository or server configuration data, uncommitted transactions, or information about user locks on repository paths.</p> </div> </div> <div class="sect2"> <div class="titlepage"> <div> <div> <h3 class="title"><a id="svn.reposadmin.maint.backup"></a>Repository Backup</h3> </div> </div> </div> <p>Despite numerous advances in technology since the birth of the modern computer, one thing unfortunately rings true with crystalline clarity—sometimes things go very, very awry. Power outages, network connectivity dropouts, corrupt RAM, and crashed hard drives are but a taste of the evil that Fate is poised to unleash on even the most conscientious administrator. And so we arrive at a very important topic—how to make backup copies of your repository data.</p> <p>There are two types of backup methods available for Subversion repository administrators—full and incremental. A full backup of the repository involves squirreling away in one sweeping action all the information required to fully reconstruct that repository in the event of a catastrophe. Usually, it means, quite literally, the duplication of the entire repository directory (which includes either a Berkeley DB or FSFS environment). Incremental backups are lesser things: backups of only the portion of the repository data that has changed since the previous backup.</p> <p>As far as full backups go, the naïve approach might seem like a sane one, but unless you temporarily disable all other access to your repository, simply doing a recursive directory copy runs the risk of generating a faulty backup. In the case of Berkeley DB, the documentation describes a certain order in which database files can be copied that will guarantee a valid backup copy. A similar ordering exists for FSFS data. But you don't have to implement these algorithms yourself, because the Subversion development team has already done so. The <span class="command"><strong>svnadmin hotcopy</strong></span> command takes care of the minutia involved in making a hot backup of your repository. And its invocation is as trivial as the Unix <span class="command"><strong>cp</strong></span> or Windows <span class="command"><strong>copy</strong></span> operations:</p> <div class="informalexample"> <pre class="screen"> $ svnadmin hotcopy /var/svn/repos /var/svn/repos-backup </pre> </div> <p>The resultant backup is a fully functional Subversion repository, able to be dropped in as a replacement for your live repository should something go horribly wrong.</p> <p>When making copies of a Berkeley DB repository, you can even instruct <span class="command"><strong>svnadmin hotcopy</strong></span> to purge any unused Berkeley DB logfiles (see <a class="xref" href="svn.reposadmin.maint.html#svn.reposadmin.maint.diskspace.bdblogs" title="Purging unused Berkeley DB logfiles">the section called “Purging unused Berkeley DB logfiles”</a>) from the original repository upon completion of the copy. Simply provide the <code class="option">--clean-logs</code> option on the command line.</p> <div class="informalexample"> <pre class="screen"> $ svnadmin hotcopy --clean-logs /var/svn/bdb-repos /var/svn/bdb-repos-backup </pre> </div> <p>Additional tooling around this command is available, too. The <code class="filename">tools/backup/</code> directory of the Subversion source distribution holds the <span class="command"><strong>hot-backup.py</strong></span> script. This script adds a bit of backup management atop <span class="command"><strong>svnadmin hotcopy</strong></span>, allowing you to keep only the most recent configured number of backups of each repository. It will automatically manage the names of the backed-up repository directories to avoid collisions with previous backups and will <span class="quote">“<span class="quote">rotate off</span>”</span> older backups, deleting them so that only the most recent ones remain. Even if you also have an incremental backup, you might want to run this program on a regular basis. For example, you might consider using <span class="command"><strong>hot-backup.py</strong></span> from a program scheduler (such as <span class="command"><strong>cron</strong></span> on Unix systems), which can cause it to run nightly (or at whatever granularity of time you deem safe).</p> <p>Some administrators use a different backup mechanism built around generating and storing repository dump data. We described in <a class="xref" href="svn.reposadmin.maint.html#svn.reposadmin.maint.migrate" title="Migrating Repository Data Elsewhere">the section called “Migrating Repository Data Elsewhere”</a> how to use <span class="command"><strong>svnadmin dump</strong></span> with the <code class="option">--incremental</code> option to perform an incremental backup of a given revision or range of revisions. And of course, you can achieve a full backup variation of this by omitting the <code class="option">--incremental</code> option to that command. There is some value in these methods, in that the format of your backed-up information is flexible—it's not tied to a particular platform, versioned filesystem type, or release of Subversion or Berkeley DB. But that flexibility comes at a cost, namely that restoring that data can take a long time—longer with each new revision committed to your repository. Also, as is the case with so many of the various backup methods, revision property changes that are made to already backed-up revisions won't get picked up by a nonoverlapping, incremental dump generation. For these reasons, we recommend against relying solely on dump-based backup approaches.</p> <p>As you can see, each of the various backup types and methods has its advantages and disadvantages. The easiest is by far the full hot backup, which will always result in a perfect working replica of your repository. Should something bad happen to your live repository, you can restore from the backup with a simple recursive directory copy. Unfortunately, if you are maintaining multiple backups of your repository, these full copies will each eat up just as much disk space as your live repository. Incremental backups, by contrast, tend to be quicker to generate and smaller to store. But the restoration process can be a pain, often involving applying multiple incremental backups. And other methods have their own peculiarities. Administrators need to find the balance between the cost of making the backup and the cost of restoring it.</p> <p>The <span class="command"><strong>svnsync</strong></span> program (see <a class="xref" href="svn.reposadmin.maint.html#svn.reposadmin.maint.replication" title="Repository Replication">the section called “Repository Replication”</a>) actually provides a rather handy middle-ground approach. If you are regularly synchronizing a read-only mirror with your main repository, in a pinch your read-only mirror is probably a good candidate for replacing that main repository if it falls over. The primary disadvantage of this method is that only the versioned repository data gets synchronized—repository configuration files, user-specified repository path locks, and other items that might live in the physical repository directory but not <span class="emphasis"><em>inside</em></span> the repository's virtual versioned filesystem are not handled by <span class="command"><strong>svnsync</strong></span>.</p> <p>In any backup scenario, repository administrators need to be aware of how modifications to unversioned revision properties affect their backups. Since these changes do not themselves generate new revisions, they will not trigger post-commit hooks, and may not even trigger the pre-revprop-change and post-revprop-change hooks.<a href="#ftn.idm6024" class="footnote" id="idm6024"><sup class="footnote">[50]</sup></a> And since you can change revision properties without respect to chronological order—you can change any revision's properties at any time—an incremental backup of the latest few revisions might not catch a property modification to a revision that was included as part of a previous backup.</p> <p>Generally speaking, only the truly paranoid would need to back up their entire repository, say, every time a commit occurred. However, assuming that a given repository has some other redundancy mechanism in place with relatively fine granularity (such as per-commit emails or incremental dumps), a hot backup of the database might be something that a repository administrator would want to include as part of a system-wide nightly backup. It's your data—protect it as much as you'd like.</p> <p>Often, the best approach to repository backups is a diversified one that leverages combinations of the methods described here. The Subversion developers, for example, back up the Subversion source code repository nightly using <span class="command"><strong>hot-backup.py</strong></span> and an off-site <span class="command"><strong>rsync</strong></span> of those full backups; keep multiple archives of all the commit and property change notification emails; and have repository mirrors maintained by various volunteers using <span class="command"><strong>svnsync</strong></span>. Your solution might be similar, but should be catered to your needs and that delicate balance of convenience with paranoia. And whatever you do, validate your backups from time to time—what good is a spare tire that has a hole in it? While all of this might not save your hardware from the iron fist of Fate,<a href="#ftn.idm6032" class="footnote" id="idm6032"><sup class="footnote">[51]</sup></a> it should certainly help you recover from those trying times.</p> </div> <div class="sect2"> <div class="titlepage"> <div> <div> <h3 class="title"><a id="svn.reposadmin.maint.uuids"></a>Managing Repository UUIDs</h3> </div> </div> </div> <p>Subversion repositories have a universally unique identifier (UUID) associated with them. This is used by Subversion clients to verify the identity of a repository when other forms of verification aren't good enough (such as checking the repository URL, which can change over time). Most Subversion repository administrators rarely, if ever, need to think about repository UUIDs as anything more than a trivial implementation detail of Subversion. Sometimes, however, there is cause for attention to this detail.</p> <p>As a general rule, you want the UUIDs of your live repositories to be unique. That is, after all, the point of having UUIDs. But there are times when you want the repository UUIDs of two repositories to be exactly the same. For example, if you make a copy of a repository for backup purposes, you want the backup to be a perfect replica of the original so that, in the event that you have to restore that backup and replace the live repository, users don't suddenly see what looks like a different repository. When dumping and loading repository history (as described earlier in <a class="xref" href="svn.reposadmin.maint.html#svn.reposadmin.maint.migrate" title="Migrating Repository Data Elsewhere">the section called “Migrating Repository Data Elsewhere”</a>), you get to decide whether to apply the UUID encapsulated in the data dump stream to the repository in which you are loading the data. The particular circumstance will dictate the correct behavior.</p> <p>There are a couple of ways to set (or reset) a repository's UUID, should you need to. As of Subversion 1.5, this is as simple as using the <span class="command"><strong>svnadmin setuuid</strong></span> command. If you provide this subcommand with an explicit UUID, it will validate that the UUID is well-formed and then set the repository UUID to that value. If you omit the UUID, a brand-new UUID will be generated for your repository.</p> <div class="informalexample"> <pre class="screen"> $ svnlook uuid /var/svn/repos cf2b9d22-acb5-11dc-bc8c-05e83ce5dbec $ svnadmin setuuid /var/svn/repos # generate a new UUID $ svnlook uuid /var/svn/repos 3c3c38fe-acc0-11dc-acbc-1b37ff1c8e7c $ svnadmin setuuid /var/svn/repos \ cf2b9d22-acb5-11dc-bc8c-05e83ce5dbec # restore the old UUID $ svnlook uuid /var/svn/repos cf2b9d22-acb5-11dc-bc8c-05e83ce5dbec $ </pre> </div> <p>For folks using versions of Subversion earlier than 1.5, these tasks are a little more complicated. You can explicitly set a repository's UUID by piping a repository dump file stub that carries the new UUID specification through <strong class="userinput"><code>svnadmin load --force-uuid <em class="replaceable"><code>REPOS-PATH</code></em></code></strong>.</p> <div class="informalexample"> <pre class="screen"> $ svnadmin load --force-uuid /var/svn/repos <<EOF SVN-fs-dump-format-version: 2 UUID: cf2b9d22-acb5-11dc-bc8c-05e83ce5dbec EOF $ svnlook uuid /var/svn/repos cf2b9d22-acb5-11dc-bc8c-05e83ce5dbec $ </pre> </div> <p>Having older versions of Subversion generate a brand-new UUID is not quite as simple to do, though. Your best bet here is to find some other way to generate a UUID, and then explicitly set the repository's UUID to that value.</p> </div> <div class="footnotes"> <br /> <hr style="width:100; text-align:left;margin-left: 0" /> <div id="ftn.idm5421" class="footnote"> <p><a href="#idm5421" class="para"><sup class="para">[43] </sup></a>Or is that, the <span class="quote">“<span class="quote">sync</span>”</span>?</p> </div> <div id="ftn.idm5572" class="footnote"> <p><a href="#idm5572" class="para"><sup class="para">[44] </sup></a>For example, hard drive + huge electromagnet = disaster.</p> </div> <div id="ftn.idm5718" class="footnote"> <p><a href="#idm5718" class="para"><sup class="para">[45] </sup></a>That's rather the reason you use version control at all, right?</p> </div> <div id="ftn.idm5720" class="footnote"> <p><a href="#idm5720" class="para"><sup class="para">[46] </sup></a>Conscious, cautious removal of certain bits of versioned data is actually supported by real use cases. That's why an <span class="quote">“<span class="quote">obliterate</span>”</span> feature has been one of the most highly requested Subversion features, and one which the Subversion developers hope to soon provide.</p> </div> <div id="ftn.idm5795" class="footnote"> <p><a href="#idm5795" class="para"><sup class="para">[47] </sup></a>While <span class="command"><strong>svnadmin dump</strong></span> has a consistent leading slash policy (to not include them), other programs that generate dump data might not be so consistent.</p> </div> <div id="ftn.idm5828" class="footnote"> <p><a href="#idm5828" class="para"><sup class="para">[48] </sup></a>In fact, it can't truly be read-only, or <span class="command"><strong>svnsync</strong></span> itself would have a tough time copying revision history into it.</p> </div> <div id="ftn.idm5888" class="footnote"> <p><a href="#idm5888" class="para"><sup class="para">[49] </sup></a>Be forewarned that while it will take only a few seconds for the average reader to parse this paragraph and the sample output that follows it, the actual time required to complete such a mirroring operation is, shall we say, quite a bit longer.</p> </div> <div id="ftn.idm6024" class="footnote"> <p><a href="#idm6024" class="para"><sup class="para">[50] </sup></a><span class="command"><strong>svnadmin setlog</strong></span> can be called in a way that bypasses the hook interface altogether.</p> </div> <div id="ftn.idm6032" class="footnote"> <p><a href="#idm6032" class="para"><sup class="para">[51] </sup></a>You know—the collective term for all of her <span class="quote">“<span class="quote">fickle fingers.</span>”</span></p> </div> </div> </div> <div class="navfooter"> <hr /> <table width="100%" summary="Navigation footer"> <tr> <td width="40%" align="left"><a accesskey="p" href="svn.reposadmin.create.html">Prev</a> </td> <td width="20%" align="center"> <a accesskey="u" href="svn.reposadmin.html">Up</a> </td> <td width="40%" align="right"> <a accesskey="n" href="svn.reposadmin.maint.moving-and-removing.html">Next</a></td> </tr> <tr> <td width="40%" align="left" valign="top">Creating and Configuring Your Repository </td> <td width="20%" align="center"> <a accesskey="h" href="index.html">Home</a> </td> <td width="40%" align="right" valign="top"> Moving and Removing Repositories</td> </tr> </table> </div> <div xmlns="" id="vcws-footer"> <hr /> <p>You are reading <em>Version Control with Subversion</em> (for Subversion 1.7), by Ben Collins-Sussman, Brian W. Fitzpatrick, and C. Michael Pilato.<br />This work is licensed under the <a href="http://creativecommons.org/licenses/by/2.0/">Creative Commons Attribution License v2.0</a>.<br />To submit comments, corrections, or other contributions to the text, please visit <a href="http://www.svnbook.com/">http://www.svnbook.com/</a>.</p> </div> </body> </html>

CINXE.COM

Repository Maintenance