Bring known bugs in 1_2_X up-to-date from HEAD

<H2>Apache 1.2 Bugs</H2>

<OL><a name="listenbug"></a>

    <LI>On some architectures if your configuration uses multiple

<OL>

    <LI><b>Exists in 1.2.1 only.</b>

    On Solaris 2.x the server will stop running after receiving a

    SIGHUP.  Four workarounds exist (choose one):<p>

    <ul>

	<li>Retrieve <a href="http://www.apache.org/dist/patches/apply_to_1.2.1/solaris_hup.patch">this patch</a>.

	<code>cd</code> to your <code>apache_1.2.1</code> directory, and

	type <code>patch -s -p1 < /path/to/patchfile</code>.  Then rebuild

	Apache.<p>

	<li>Use SIGUSR1 instead of SIGHUP, see <a href="../stopping.html">

	Stopping and Restarting Apache</a> for more details.<p>

	<li>Add <code>-DNO_SLACK</code> to

	<code>EXTRA_CFLAGS</code> in

	your <code>Configuration</code> file, re-run <code>Configure</code>

	and rebuild your server.  This disables the

	<a href="descriptors.html">descriptor slack workaround</a><p>

	<li>(Least preferable) Use Apache 1.2.0 instead of 1.2.1.<p>

    </ul>

    This problem will be tracked as

    <a href="http://www.apache.org/bugdb.cgi/full/832">PR#832</a>.<p>

    </li>

    <LI>(Exists in 1.2.0 and in 1.2.1 after either of the

    <code>NO_SLACK</code> or patch provided by the previous bug are applied.)

    Solaris 2.5.1 (and probably other versions of Solaris) appear to have

    a race condition completely unrelated to all the others.  It is possible

    during a SIGHUP that the server will fail to start because it will not

    be able to re-open its sockets.  To our knowledge this has only shown

    up during testing when we pummel the server with as many SIGHUP requests

    per second as we can.  This appears unrelated to the similar sounding bug

    described in <a href="http://www.apache.org/bugdb.cgi/full/832">PR#832</a>.

    <p>

    <LI><a name="listenbug">On some architectures</A>

    if your configuration uses multiple

    <a href="../mod/core.html#listen">Listen</a> directives then it is possible

    that the server will starve one of the sockets while serving hits on

    another.  The work-around is to add

    <a href="http://www.apache.org/bugdb.cgi/full/467">PR#467</a>.

    <P></LI>

    <a name="cnegbug"></a>

    <LI><b>Fixed in 1.2.1.</b>

    Apache's <A HREF="../content-negotiation.html">Content

    <a name="cnegbug">Apache's</A> <A HREF="../content-negotiation.html">Content

    Negotiation</A> should pick the smallest variant if there

    are several that are equally acceptable. A bug in 1.2 means it no

    longer does this unless all the variants have character sets.

    <LI>If compilation fails complaining about "unknown symbol __inet_ntoa()"

    then you have probably installed version 8 of bind. You will need to

    explicitly link with the bind library by adding <CODE>-lbind</CODE>

    to <CODE>EXTRA_LFLAGS</CODE> in <CODE>Configuration</CODE>. See

    to <CODE>EXTRA_LDFLAGS</CODE> in <CODE>Configuration</CODE>. See

    <A HREF="http://www.apache.org/bugdb.cgi/full/616">PR#616</A>

    and the

    <A HREF="FAQ.html#bind8.1">Apache FAQ</A>.