blog/archive/2013/11/11.rss

   1 <?xml version="1.0" encoding="ISO-8859-1"?>
   2 <rss version='2.0' xmlns:lj='http://www.livejournal.org/rss/lj/1.0/'>
   3         <channel>
   4                 <title>Petter Reinholdtsen - Entries from November 2013</title>
   5                 <description>Entries from November 2013</description>
   6                 <link>http://people.skolelinux.org/pere/blog/</link>
   7
   8
   9         <item>
  10                 <title>Lets make a wireless community network in Oslo!</title>
  11                 <link>http://people.skolelinux.org/pere/blog/Lets_make_a_wireless_community_network_in_Oslo_.html</link>
  12                 <guid isPermaLink="true">http://people.skolelinux.org/pere/blog/Lets_make_a_wireless_community_network_in_Oslo_.html</guid>
  13                 <pubDate>Wed, 13 Nov 2013 21:00:00 +0100</pubDate>
  14                 <description>&lt;p&gt;Today NUUG and Hackeriet announced
  15 &lt;a href=&quot;http://www.nuug.no/news/Bli_med___bygge_dugnadsnett_for_alle_i_Oslo.shtml&quot;&gt;our
  16 plans to join forces and create a wireless community network in
  17 Oslo&lt;/a&gt;.  The workshop to help people get started will take place
  18 Thursday 2013-11-28, but we already are collecting the geolocation of
  19 people joining forces to make this happen.  We have
  20 &lt;a href=&quot;https://github.com/petterreinholdtsen/meshfx-node/blob/master/oslo-nodes.geojson&quot;&gt;9
  21 locations plotted on the map&lt;/a&gt;, but we will need more before we have
  22 a connected mesh spread across Oslo.  If this sound interesting to
  23 you, please join us at the workshop.  If you are too impatient to wait
  24 15 days, please join us on the IRC channel
  25 &lt;a href=&quot;irc://irc.freenode.net/%23nuug&quot;&gt;#nuug on irc.freenode.net&lt;/a&gt;
  26 right away. :)&lt;/p&gt;
  27 </description>
  28         </item>
  29
  30         <item>
  31                 <title>Running TP-Link MR3040 as a batman-adv mesh node using openwrt</title>
  32                 <link>http://people.skolelinux.org/pere/blog/Running_TP_Link_MR3040_as_a_batman_adv_mesh_node_using_openwrt.html</link>
  33                 <guid isPermaLink="true">http://people.skolelinux.org/pere/blog/Running_TP_Link_MR3040_as_a_batman_adv_mesh_node_using_openwrt.html</guid>
  34                 <pubDate>Sun, 10 Nov 2013 23:00:00 +0100</pubDate>
  35                 <description>&lt;p&gt;Continuing my research into mesh networking, I was recommended to
  36 use TP-Link 3040 and 3600 access points as mesh nodes, and the pair I
  37 bought arrived on Friday.  Here are my notes on how to set up the
  38 MR3040 as a mesh node using
  39 &lt;a href=&quot;http://www.openwrt.org/&quot;&gt;OpenWrt&lt;/a&gt;.&lt;/p&gt;
  40
  41 &lt;p&gt;I started by following the instructions on the OpenWRT wiki for
  42 &lt;a href=&quot;http://wiki.openwrt.org/toh/tp-link/tl-mr3040&quot;&gt;TL-MR3040&lt;/a&gt;,
  43 and downloaded
  44 &lt;a href=&quot;http://downloads.openwrt.org/snapshots/trunk/ar71xx/openwrt-ar71xx-generic-tl-mr3040-v2-squashfs-factory.bin&quot;&gt;the
  45 recommended firmware image&lt;/a&gt;
  46 (openwrt-ar71xx-generic-tl-mr3040-v2-squashfs-factory.bin) and
  47 uploaded it into the original web interface.  The flashing went fine,
  48 and the machine was available via telnet on the ethernet port.  After
  49 logging in and setting the root password, ssh was available and I
  50 could start to set it up as a batman-adv mesh node.&lt;/p&gt;
  51
  52 &lt;p&gt;I started off by reading the instructions from
  53 &lt;a href=&quot;http://wirelessafrica.meraka.org.za/wiki/index.php?title=Antoine&#39;s_Research&quot;&gt;Wireless
  54 Africa&lt;/a&gt;, which had quite a lot of useful information, but
  55 eventually I followed the recipe from the Open Mesh wiki for
  56 &lt;a href=&quot;http://www.open-mesh.org/projects/batman-adv/wiki/Batman-adv-openwrt-config&quot;&gt;using
  57 batman-adv on OpenWrt&lt;/a&gt;.  A small snag was the fact that the
  58 &lt;tt&gt;opkg install kmod-batman-adv&lt;/tt&gt; command did not work as it
  59 should.  The batman-adv kernel module would fail to load because its
  60 dependency crc16 was not already loaded.  I
  61 &lt;a href=&quot;https://dev.openwrt.org/ticket/14452&quot;&gt;reported the bug&lt;/a&gt; to
  62 the openwrt project and hope it will be fixed soon.  But the problem
  63 only seem to affect initial testing of batman-adv, as configuration
  64 seem to work when booting from scratch.&lt;/p&gt;
  65
  66 &lt;p&gt;The setup is done using files in /etc/config/.  I did not bridge
  67 the Ethernet and mesh interfaces this time, to be able to hook up the
  68 box on my local network and log into it for configuration updates.
  69 The following files were changed and look like this after modifying
  70 them:&lt;/p&gt;
  71
  72 &lt;p&gt;&lt;tt&gt;/etc/config/network&lt;/tt&gt;&lt;/p&gt;
  73
  74 &lt;pre&gt;
  75
  76 config interface &#39;loopback&#39;
  77         option ifname &#39;lo&#39;
  78         option proto &#39;static&#39;
  79         option ipaddr &#39;127.0.0.1&#39;
  80         option netmask &#39;255.0.0.0&#39;
  81
  82 config globals &#39;globals&#39;
  83         option ula_prefix &#39;fdbf:4c12:3fed::/48&#39;
  84
  85 config interface &#39;lan&#39;
  86         option ifname &#39;eth0&#39;
  87         option type &#39;bridge&#39;
  88         option proto &#39;dhcp&#39;
  89         option ipaddr &#39;192.168.1.1&#39;
  90         option netmask &#39;255.255.255.0&#39;
  91         option hostname &#39;tl-mr3040&#39;
  92         option ip6assign &#39;60&#39;
  93
  94 config interface &#39;mesh&#39;
  95         option ifname &#39;adhoc0&#39;
  96         option mtu &#39;1528&#39;
  97         option proto &#39;batadv&#39;
  98         option mesh &#39;bat0&#39;
  99 &lt;/pre&gt;
 100
 101 &lt;p&gt;&lt;tt&gt;/etc/config/wireless&lt;/tt&gt;&lt;/p&gt;
 102 &lt;pre&gt;
 103
 104 config wifi-device &#39;radio0&#39;
 105         option type &#39;mac80211&#39;
 106         option channel &#39;11&#39;
 107         option hwmode &#39;11ng&#39;
 108         option path &#39;platform/ar933x_wmac&#39;
 109         option htmode &#39;HT20&#39;
 110         list ht_capab &#39;SHORT-GI-20&#39;
 111         list ht_capab &#39;SHORT-GI-40&#39;
 112         list ht_capab &#39;RX-STBC1&#39;
 113         list ht_capab &#39;DSSS_CCK-40&#39;
 114         option disabled &#39;0&#39;
 115
 116 config wifi-iface &#39;wmesh&#39;
 117         option device &#39;radio0&#39;
 118         option ifname &#39;adhoc0&#39;
 119         option network &#39;mesh&#39;
 120         option encryption &#39;none&#39;
 121         option mode &#39;adhoc&#39;
 122         option bssid &#39;02:BA:00:00:00:01&#39;
 123         option ssid &#39;meshfx@hackeriet&#39;
 124 &lt;/pre&gt;
 125 &lt;p&gt;&lt;tt&gt;/etc/config/batman-adv&lt;/tt&gt;&lt;/p&gt;
 126 &lt;pre&gt;
 127
 128 config &#39;mesh&#39; &#39;bat0&#39;
 129         option interfaces &#39;adhoc0&#39;
 130         option &#39;aggregated_ogms&#39;
 131         option &#39;ap_isolation&#39;
 132         option &#39;bonding&#39;
 133         option &#39;fragmentation&#39;
 134         option &#39;gw_bandwidth&#39;
 135         option &#39;gw_mode&#39;
 136         option &#39;gw_sel_class&#39;
 137         option &#39;log_level&#39;
 138         option &#39;orig_interval&#39;
 139         option &#39;vis_mode&#39;
 140         option &#39;bridge_loop_avoidance&#39;
 141         option &#39;distributed_arp_table&#39;
 142         option &#39;network_coding&#39;
 143         option &#39;hop_penalty&#39;
 144
 145 # yet another batX instance
 146 # config &#39;mesh&#39; &#39;bat5&#39;
 147 #       option &#39;interfaces&#39; &#39;second_mesh&#39;
 148 &lt;/pre&gt;
 149
 150 &lt;p&gt;The mesh node is now operational.  I have yet to test its range,
 151 but I hope it is good.  I have not yet tested the TP-Link 3600 box
 152 still wrapped up in plastic.&lt;/p&gt;
 153 </description>
 154         </item>
 155
 156         <item>
 157                 <title>Debian init.d boot script example for rsyslog</title>
 158                 <link>http://people.skolelinux.org/pere/blog/Debian_init_d_boot_script_example_for_rsyslog.html</link>
 159                 <guid isPermaLink="true">http://people.skolelinux.org/pere/blog/Debian_init_d_boot_script_example_for_rsyslog.html</guid>
 160                 <pubDate>Sat, 2 Nov 2013 22:40:00 +0100</pubDate>
 161                 <description>&lt;p&gt;If one of the points of switching to a new init system in Debian is
 162 &lt;a href=&quot;http://thomas.goirand.fr/blog/?p=147&quot;&gt;to get rid of huge
 163 init.d scripts&lt;/a&gt;, I doubt we need to switch away from sysvinit and
 164 init.d scripts at all.  Here is an example init.d script, ie a rewrite
 165 of /etc/init.d/rsyslog:&lt;/p&gt;
 166
 167 &lt;p&gt;&lt;pre&gt;
 168 #!/lib/init/init-d-script
 169 ### BEGIN INIT INFO
 170 # Provides:          rsyslog
 171 # Required-Start:    $remote_fs $time
 172 # Required-Stop:     umountnfs $time
 173 # X-Stop-After:      sendsigs
 174 # Default-Start:     2 3 4 5
 175 # Default-Stop:      0 1 6
 176 # Short-Description: enhanced syslogd
 177 # Description:       Rsyslog is an enhanced multi-threaded syslogd.
 178 #                    It is quite compatible to stock sysklogd and can be
 179 #                    used as a drop-in replacement.
 180 ### END INIT INFO
 181 DESC=&quot;enhanced syslogd&quot;
 182 DAEMON=/usr/sbin/rsyslogd
 183 &lt;/pre&gt;&lt;/p&gt;
 184
 185 &lt;p&gt;Pretty minimalistic to me... For the record, the original sysv-rc
 186 script was 137 lines, and the above is just 15 lines, most of it meta
 187 info/comments.&lt;/p&gt;
 188
 189 &lt;p&gt;How to do this, you ask?  Well, one create a new script
 190 /lib/init/init-d-script looking something like this:
 191
 192 &lt;p&gt;&lt;pre&gt;
 193 #!/bin/sh
 194
 195 # Define LSB log_* functions.
 196 # Depend on lsb-base (&gt;= 3.2-14) to ensure that this file is present
 197 # and status_of_proc is working.
 198 . /lib/lsb/init-functions
 199
 200 #
 201 # Function that starts the daemon/service
 202
 203 #
 204 do_start()
 205 {
 206         # Return
 207         #   0 if daemon has been started
 208         #   1 if daemon was already running
 209         #   2 if daemon could not be started
 210         start-stop-daemon --start --quiet --pidfile $PIDFILE --exec $DAEMON --test &gt; /dev/null \
 211                 || return 1
 212         start-stop-daemon --start --quiet --pidfile $PIDFILE --exec $DAEMON -- \
 213                 $DAEMON_ARGS \
 214                 || return 2
 215         # Add code here, if necessary, that waits for the process to be ready
 216         # to handle requests from services started subsequently which depend
 217         # on this one.  As a last resort, sleep for some time.
 218 }
 219
 220 #
 221 # Function that stops the daemon/service
 222 #
 223 do_stop()
 224 {
 225         # Return
 226         #   0 if daemon has been stopped
 227         #   1 if daemon was already stopped
 228         #   2 if daemon could not be stopped
 229         #   other if a failure occurred
 230         start-stop-daemon --stop --quiet --retry=TERM/30/KILL/5 --pidfile $PIDFILE --name $NAME
 231         RETVAL=&quot;$?&quot;
 232         [ &quot;$RETVAL&quot; = 2 ] &amp;&amp; return 2
 233         # Wait for children to finish too if this is a daemon that forks
 234         # and if the daemon is only ever run from this initscript.
 235         # If the above conditions are not satisfied then add some other code
 236         # that waits for the process to drop all resources that could be
 237         # needed by services started subsequently.  A last resort is to
 238         # sleep for some time.
 239         start-stop-daemon --stop --quiet --oknodo --retry=0/30/KILL/5 --exec $DAEMON
 240         [ &quot;$?&quot; = 2 ] &amp;&amp; return 2
 241         # Many daemons don&#39;t delete their pidfiles when they exit.
 242         rm -f $PIDFILE
 243         return &quot;$RETVAL&quot;
 244 }
 245
 246 #
 247 # Function that sends a SIGHUP to the daemon/service
 248 #
 249 do_reload() {
 250         #
 251         # If the daemon can reload its configuration without
 252         # restarting (for example, when it is sent a SIGHUP),
 253         # then implement that here.
 254         #
 255         start-stop-daemon --stop --signal 1 --quiet --pidfile $PIDFILE --name $NAME
 256         return 0
 257 }
 258
 259 SCRIPTNAME=$1
 260 scriptbasename=&quot;$(basename $1)&quot;
 261 echo &quot;SN: $scriptbasename&quot;
 262 if [ &quot;$scriptbasename&quot; != &quot;init-d-library&quot; ] ; then
 263     script=&quot;$1&quot;
 264     shift
 265     . $script
 266 else
 267     exit 0
 268 fi
 269
 270 NAME=$(basename $DAEMON)
 271 PIDFILE=/var/run/$NAME.pid
 272
 273 # Exit if the package is not installed
 274 #[ -x &quot;$DAEMON&quot; ] || exit 0
 275
 276 # Read configuration variable file if it is present
 277 [ -r /etc/default/$NAME ] &amp;&amp; . /etc/default/$NAME
 278
 279 # Load the VERBOSE setting and other rcS variables
 280 . /lib/init/vars.sh
 281
 282 case &quot;$1&quot; in
 283   start)
 284         [ &quot;$VERBOSE&quot; != no ] &amp;&amp; log_daemon_msg &quot;Starting $DESC&quot; &quot;$NAME&quot;
 285         do_start
 286         case &quot;$?&quot; in
 287                 0|1) [ &quot;$VERBOSE&quot; != no ] &amp;&amp; log_end_msg 0 ;;
 288                 2) [ &quot;$VERBOSE&quot; != no ] &amp;&amp; log_end_msg 1 ;;
 289         esac
 290         ;;
 291   stop)
 292         [ &quot;$VERBOSE&quot; != no ] &amp;&amp; log_daemon_msg &quot;Stopping $DESC&quot; &quot;$NAME&quot;
 293         do_stop
 294         case &quot;$?&quot; in
 295                 0|1) [ &quot;$VERBOSE&quot; != no ] &amp;&amp; log_end_msg 0 ;;
 296                 2) [ &quot;$VERBOSE&quot; != no ] &amp;&amp; log_end_msg 1 ;;
 297         esac
 298         ;;
 299   status)
 300         status_of_proc &quot;$DAEMON&quot; &quot;$NAME&quot; &amp;&amp; exit 0 || exit $?
 301         ;;
 302   #reload|force-reload)
 303         #
 304         # If do_reload() is not implemented then leave this commented out
 305         # and leave &#39;force-reload&#39; as an alias for &#39;restart&#39;.
 306         #
 307         #log_daemon_msg &quot;Reloading $DESC&quot; &quot;$NAME&quot;
 308         #do_reload
 309         #log_end_msg $?
 310         #;;
 311   restart|force-reload)
 312         #
 313         # If the &quot;reload&quot; option is implemented then remove the
 314         # &#39;force-reload&#39; alias
 315         #
 316         log_daemon_msg &quot;Restarting $DESC&quot; &quot;$NAME&quot;
 317         do_stop
 318         case &quot;$?&quot; in
 319           0|1)
 320                 do_start
 321                 case &quot;$?&quot; in
 322                         0) log_end_msg 0 ;;
 323                         1) log_end_msg 1 ;; # Old process is still running
 324                         *) log_end_msg 1 ;; # Failed to start
 325                 esac
 326                 ;;
 327           *)
 328                 # Failed to stop
 329                 log_end_msg 1
 330                 ;;
 331         esac
 332         ;;
 333   *)
 334         echo &quot;Usage: $SCRIPTNAME {start|stop|status|restart|force-reload}&quot; &gt;&amp;2
 335         exit 3
 336         ;;
 337 esac
 338
 339 :
 340 &lt;/pre&gt;&lt;/p&gt;
 341
 342 &lt;p&gt;It is based on /etc/init.d/skeleton, and could be improved quite a
 343 lot.  I did not really polish the approach, so it might not always
 344 work out of the box, but you get the idea.  I did not try very hard to
 345 optimize it nor make it more robust either.&lt;/p&gt;
 346
 347 &lt;p&gt;A better argument for switching init system in Debian than reducing
 348 the size of init scripts (which is a good thing to do anyway), is to
 349 get boot system that is able to handle the kernel events sensibly and
 350 robustly, and do not depend on the boot to run sequentially.  The boot
 351 and the kernel have not behaved sequentially in years.&lt;/p&gt;
 352 </description>
 353         </item>
 354
 355         <item>
 356                 <title>Browser plugin for SPICE (spice-xpi) uploaded to Debian</title>
 357                 <link>http://people.skolelinux.org/pere/blog/Browser_plugin_for_SPICE__spice_xpi__uploaded_to_Debian.html</link>
 358                 <guid isPermaLink="true">http://people.skolelinux.org/pere/blog/Browser_plugin_for_SPICE__spice_xpi__uploaded_to_Debian.html</guid>
 359                 <pubDate>Fri, 1 Nov 2013 11:00:00 +0100</pubDate>
 360                 <description>&lt;p&gt;&lt;a href=&quot;http://www.spice-space.org/&quot;&gt;The SPICE protocol&lt;/a&gt; for
 361 remote display access is the preferred solution with oVirt and RedHat
 362 Enterprise Virtualization, and I was sad to discover the other day
 363 that the browser plugin needed to use these systems seamlessly was
 364 missing in Debian.  The &lt;a href=&quot;http://bugs.debian.org/668284&quot;&gt;request
 365 for a package&lt;/a&gt; was from 2012-04-10 with no progress since
 366 2013-04-01, so I decided to wrap up a package based on the great work
 367 from Cajus Pollmeier and put it in a collab-maint maintained git
 368 repository to get a package I could use.  I would very much like
 369 others to help me maintain the package (or just take over, I do not
 370 mind), but as no-one had volunteered so far, I just uploaded it to
 371 NEW.  I hope it will be available in Debian in a few days.&lt;/p&gt;
 372
 373 &lt;p&gt;The source is now available from
 374 &lt;a href=&quot;http://anonscm.debian.org/gitweb/?p=collab-maint/spice-xpi.git;a=summary&quot;&gt;http://anonscm.debian.org/gitweb/?p=collab-maint/spice-xpi.git;a=summary&lt;/a&gt;.&lt;/p&gt;
 375 </description>
 376         </item>
 377
 378         </channel>
 379 </rss>