httpd-cvs mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Dean Gaudet <>
Subject cvs commit: apachen/htdocs/manual upgrading_to_1_3.html index.html
Date Mon, 06 Oct 1997 03:06:39 GMT
dgaudet     97/10/05 20:06:38

  Modified:    htdocs/manual index.html
  Added:       htdocs/manual upgrading_to_1_3.html
  Document changes from 1.2 to 1.3 that will affect existing installations.
  Revision  Changes    Path
  1.17      +1 -0      apachen/htdocs/manual/index.html
  Index: index.html
  RCS file: /export/home/cvs/apachen/htdocs/manual/index.html,v
  retrieving revision 1.16
  retrieving revision 1.17
  diff -u -r1.16 -r1.17
  --- index.html	1997/08/01 17:06:32	1.16
  +++ index.html	1997/10/06 03:06:35	1.17
  @@ -20,6 +20,7 @@
   <h3><a name="new">Release Notes</a></h3>
   <li><a href="new_features_1_3.html">New features in Apache 1.3</a>
  +<li><a href="upgrading_to_1_3.html">Upgrading to Apache 1.3</a>
   <li><a href="new_features_1_2.html">New features in Apache 1.2</a>
   <li><a href="new_features_1_1.html">New features in Apache 1.1</a>
   <li><a href="new_features_1_0.html">New features in Apache 1.0</a>
  1.1                  apachen/htdocs/manual/upgrading_to_1_3.html
  Index: upgrading_to_1_3.html
  <title>Upgrading to 1.3 from 1.2</title>
  <!-- Background white, links blue (unvisited), navy (visited), red (active) -->
  <!--#include virtual="header.html" -->
  <h1 ALIGN="CENTER">Upgrading to 1.3 from 1.2</h1>
  <p>In order to assist folks upgrading we are now going to maintain a
  document describing information critical to existing Apache users.  Note
  that it only lists differences between recent major releases, so
  for example, folks using Apache 1.1 or earlier will have to figure out
  what changed up to Apache 1.2 before this document can be considered
  relevant.  Old users could look at the <code>src/CHANGES</code> file
  which tracks code changes.
  <p>These are intended to be brief notes, and you should be able to find
  more information in either the <a href="new_features_1_3.html">New Features</a>
  document, or in the <code>src/CHANGES</code> file.
  <h3>Compile-Time Configuration Changes</h3>
    <LI>The source code has been <a href="sourcereorg.html">reorganized</a>,
      which affects anyone with custom modules or modifications.  But also,
      the <code>Module</code> directive has been changed to the
      <code>AddModule</code> directive.
    <LI>The <code>Configuration</code> variable <code>EXTRA_LFLAGS</code>
      been renamed <code>EXTRA_LDFLAGS</code>.
    <LI>The <code>-DMAXIMUM_DNS</code> definition has been obsoleted by
      changes to <code>mod_access</code> enforcing double-reverse DNS lookups
      all the time.
    <LI><code>mod_dir</code> has been split into two pieces
      <code><a href="mod/mod_autoindex.html">mod_autoindex</a></code>,
      <code><a href="mod/mod_dir.html">mod_dir</a></code>.
    <LI><a href="mod/mod_browser.html"><code>mod_browser</code></a>
has been
      replaced by <a href="mod/mod_setenvif.html"><code>mod_setenvif</code></a>.
    <LI>IRIX systems with untrusted users who can write CGIs which execute
      as the same uid as httpd should consider using <code>suexec</code>,
      or adding <code>-DUSE_FCNTL_SERIALIZED_ACCEPT</code> to
      <code>EXTRA_CFLAGS</code>.  This is slower, more information is available
      on the <a href="misc/perf-tuning.html#serialize">performance tuning
      page</a>.  There is a mild denial of service attack possible with the
      default config, but the default config is an order of magnitude faster.
    <li><code>mod_auth_msql</code> has been removed from the distribution.
  <h3>Run-Time Configuration Changes</h3>
    <li><code>HostnameLookups</code> defaults to Off.
    <li>The undocumented
      <a href="mod/mod_access.html"><code>mod_access</code></a>
      syntax "allow user-agents" was removed.  The replacement is the
      more general "allow from env".
    <li>When using wildcards in pathnames (such as * and ?) they no longer
      match / (slash).  That is, they more closely behave how the shell
      behaves.  This affects <code>&lt;Directory&gt;</code> directives,
      for example.
    <li>If no <code>TransferLog</code> directive is given then nothing will
      be logged.
      (Previously it would default to <code>logs/access_log</code>.)
    <li>Apache now has <a href="mod/core.html#loglevel">configurable error
      logging levels</a>, and the default eliminates some messages that
      earlier versions always generated.
    <li>When booting, Apache will now detach itself from stdin, stdout,
       and stderr.  stderr will not be detached until after the config
       files have been read so you will be able to see initial error
       messages.  After that all errors are logged in the error_log.
       This makes it more convenient to start Apache via rsh, ssh,
       or crontabs.
  <h3>Misc Changes</h3>
    <li><code>ServerType inetd</code> has been deprecated.  It still exists,
      but bugs are unlikely to be fixed.
    <li><code>httpd_monitor</code> has been deprecated.  The replacement
      to use <code>mod_status</code> and make a request to a URL such as
      Apache now provides an effectively unbuffered connection for
      CGI scripts.  This means that data will be sent to the client
      as soon as the CGI pauses or stops output; previously, Apache would
      buffer the output up to a fixed buffer size before sending, which
      could result in the user viewing an empty page until the CGI finished
      or output a complete buffer.  It is no longer necessary to use an
      "nph-" CGI to get unbuffered output.  Given that most CGIs are written
      in a language that by default does buffering (e.g. perl) this
      shouldn't have a detrimental effect on performance.
      <p>"nph-" CGIs, which formerly provided a direct socket to the client
      without any server post-processing, were not fully compatible with
      HTTP/1.1 or SSL support.  As such they would have had to implement
      the transport details, such as encryption or chunking, in order
      to work properly in certain situations.  Now, the only difference
      between nph and non-nph scripts is "non-parsed headers".
      <code>dbmmanage</code> has been overhauled.
  <!--#include virtual="footer.html" -->

View raw message