blog/tags/bootsystem/index.html

   1 <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
   2           "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
   3 <html xmlns="http://www.w3.org/1999/xhtml" dir="ltr">
   4   <head>
   5     <meta http-equiv="Content-Type" content="text/html;charset=utf-8" />
   6     <title>Petter Reinholdtsen: Entries Tagged bootsystem</title>
   7     <link rel="stylesheet" type="text/css" media="screen" href="http://people.skolelinux.org/pere/blog/style.css" />
   8     <link rel="stylesheet" type="text/css" media="screen" href="http://people.skolelinux.org/pere/blog/vim.css" />
   9     <link rel="alternate" title="RSS Feed" href="bootsystem.rss" type="application/rss+xml" />
  10   </head>
  11   <body>
  12     <div class="title">
  13  <h1>
  14      <a href="http://people.skolelinux.org/pere/blog/">Petter Reinholdtsen</a>
  15
  16  </h1>
  17
  18 </div>
  19
  20
  21     <h3>Entries tagged "bootsystem".</h3>
  22
  23     <div class="entry">
  24       <div class="title">
  25         <a href="http://people.skolelinux.org/pere/blog/How_is_booting_into_runlevel_1_different_from_single_user_boots_.html">How is booting into runlevel 1 different from single user boots?</a>
  26       </div>
  27       <div class="date">
  28          4th August 2011
  29       </div>
  30       <div class="body">
  31         <p>Wouter Verhelst have some
  32 <a href="http://grep.be/blog/en/retorts/pere_kubuntu_boot">interesting
  33 comments and opinions</a> on my blog post on
  34 <a href="http://people.skolelinux.org/pere/blog/What_should_start_from__etc_rcS_d__in_Debian____almost_nothing.html">the
  35 need to clean up /etc/rcS.d/ in Debian</a> and my blog post about
  36 <a href="http://people.skolelinux.org/pere/blog/What_is_missing_in_the_Debian_desktop__or_why_my_parents_use_Kubuntu.html">the
  37 default KDE desktop in Debian</a>.  I only have time to address one
  38 small piece of his comment now, and though it best to address the
  39 misunderstanding he bring forward:</p>
  40
  41 <p><blockquote>
  42 Currently, a system admin has four options: [...] boot to a
  43 single-user system (by adding 'single' to the kernel command line;
  44 this runs rcS and rc1 scripts)
  45 </blockquote></p>
  46
  47 <p>This make me believe Wouter believe booting into single user mode
  48 and booting into runlevel 1 is the same.  I am not surprised he
  49 believe this, because it would make sense and is a quite sensible
  50 thing to believe.  But because the boot in Debian is slightly broken,
  51 runlevel 1 do not work properly and it isn't the same as single user
  52 mode.  I'll try to explain what is actually happing, but it is a bit
  53 hard to explain.</p>
  54
  55 <p>Single user mode is defined like this in /etc/inittab:
  56 "<tt>~~:S:wait:/sbin/sulogin</tt>".  This means the only thing that is
  57 executed in single user mode is sulogin.  Single user mode is a boot
  58 state "between" the runlevels, and when booting into single user mode,
  59 only the scripts in /etc/rcS.d/ are executed before the init process
  60 enters the single user state.  When switching to runlevel 1, the state
  61 is in fact not ending in runlevel 1, but it passes through runlevel 1
  62 and end up in the single user mode (see /etc/rc1.d/S03single, which
  63 runs "init -t1 S" to switch to single user mode at the end of runlevel
  64 1.  It is confusing that the 'S' (single user) init mode is not the
  65 mode enabled by /etc/rcS.d/ (which is more like the initial boot
  66 mode).</p>
  67
  68 <p>This summary might make it clearer.  When booting for the first
  69 time into single user mode, the following commands are executed:
  70 "<tt>/etc/init.d/rc S; /sbin/sulogin</tt>".  When booting into
  71 runlevel 1, the following commands are executed: "<tt>/etc/init.d/rc
  72 S; /etc/init.d/rc 1; /sbin/sulogin</tt>".  A problem show up when
  73 trying to continue after visiting single user mode.  Not all services
  74 are started again as they should, causing the machine to end up in an
  75 unpredicatble state.  This is why Debian admins recommend rebooting
  76 after visiting single user mode.</p>
  77
  78 <p>A similar problem with runlevel 1 is caused by the amount of
  79 scripts executed from /etc/rcS.d/.  When switching from say runlevel 2
  80 to runlevel 1, the services started from /etc/rcS.d/ are not properly
  81 stopped when passing through the scripts in /etc/rc1.d/, and not
  82 started again when switching away from runlevel 1 to the runlevels
  83 2-5.  I believe the problem is best fixed by moving all the scripts
  84 out of /etc/rcS.d/ that are not <strong>required</strong> to get a
  85 functioning single user mode during boot.</p>
  86
  87 <p>I have spent several years investigating the Debian boot system,
  88 and discovered this problem a few years ago.  I suspect it originates
  89 from when sysvinit was introduced into Debian, a long time ago.</p>
  90
  91       </div>
  92       <div class="tags">
  93
  94
  95         Tags: <a href="http://people.skolelinux.org/pere/blog/tags/bootsystem">bootsystem</a>, <a href="http://people.skolelinux.org/pere/blog/tags/debian">debian</a>, <a href="http://people.skolelinux.org/pere/blog/tags/english">english</a>.
  96
  97
  98       </div>
  99     </div>
 100     <div class="padding"></div>
 101
 102     <div class="entry">
 103       <div class="title">
 104         <a href="http://people.skolelinux.org/pere/blog/What_should_start_from__etc_rcS_d__in_Debian____almost_nothing.html">What should start from /etc/rcS.d/ in Debian? - almost nothing</a>
 105       </div>
 106       <div class="date">
 107         30th July 2011
 108       </div>
 109       <div class="body">
 110         <p>In the Debian boot system, several packages include scripts that
 111 are started from /etc/rcS.d/.  In fact, there is a bite more of them
 112 than make sense, and this causes a few problems.  What kind of
 113 problems, you might ask.  There are at least two problems.  The first
 114 is that it is not possible to recover a machine after switching to
 115 runlevel 1.  One need to actually reboot to get the machine back to
 116 the expected state.  The other is that single user boot will sometimes
 117 run into problems because some of the subsystems are activated before
 118 the root login is presented, causing problems when trying to recover a
 119 machine from a problem in that subsystem.  A minor additional point is
 120 that moving more scripts out of rcS.d/ and into the other rc#.d/
 121 directories will increase the amount of scripts that can run in
 122 parallel during boot, and thus decrease the boot time.</p>
 123
 124 <p>So, which scripts should start from rcS.d/.  In short, only the
 125 scripts that _have_ to execute before the root login prompt is
 126 presented during a single user boot should go there.  Everything else
 127 should go into the numeric runlevels.  This means things like
 128 lm-sensors, fuse and x11-common should not run from rcS.d, but from
 129 the numeric runlevels.  Today in Debian, there are around 115 init.d
 130 scripts that are started from rcS.d/, and most of them should be moved
 131 out.  Do your package have one of them?  Please help us make single
 132 user and runlevel 1 better by moving it.</p>
 133
 134 <p>Scripts setting up the screen, keyboard, system partitions
 135 etc. should still be started from rcS.d/, but there is for example no
 136 need to have the network enabled before the single user login prompt
 137 is presented.</p>
 138
 139 <p>As always, things are not so easy to fix as they sound.  To keep
 140 Debian systems working while scripts migrate and during upgrades, the
 141 scripts need to be moved from rcS.d/ to rc2.d/ in reverse dependency
 142 order, ie the scripts that nothing in rcS.d/ depend on can be moved,
 143 and the next ones can only be moved when their dependencies have been
 144 moved first.  This migration must be done sequentially while we ensure
 145 that the package system upgrade packages in the right order to keep
 146 the system state correct.  This will require some coordination when it
 147 comes to network related packages, but most of the packages with
 148 scripts that should migrate do not have anything in rcS.d/ depending
 149 on them.  Some packages have already been updated, like the sudo
 150 package, while others are still left to do.  I wish I had time to work
 151 on this myself, but real live constrains make it unlikely that I will
 152 find time to push this forward.</p>
 153
 154       </div>
 155       <div class="tags">
 156
 157
 158         Tags: <a href="http://people.skolelinux.org/pere/blog/tags/bootsystem">bootsystem</a>, <a href="http://people.skolelinux.org/pere/blog/tags/debian">debian</a>, <a href="http://people.skolelinux.org/pere/blog/tags/english">english</a>.
 159
 160
 161       </div>
 162     </div>
 163     <div class="padding"></div>
 164
 165     <div class="entry">
 166       <div class="title">
 167         <a href="http://people.skolelinux.org/pere/blog/Automatic_upgrade_testing_from_Lenny_to_Squeeze.html">Automatic upgrade testing from Lenny to Squeeze</a>
 168       </div>
 169       <div class="date">
 170         11th June 2010
 171       </div>
 172       <div class="body">
 173         <p>The last few days I have done some upgrade testing in Debian, to
 174 see if the upgrade from Lenny to Squeeze will go smoothly.  A few bugs
 175 have been discovered and reported in the process
 176 (<a href="http://bugs.debian.org/585410">#585410</a> in nagios3-cgi,
 177 <a href="http://bugs.debian.org/584879">#584879</a> already fixed in
 178 enscript and <a href="http://bugs.debian.org/584861">#584861</a> in
 179 kdebase-workspace-data), and to get a more regular testing going on, I
 180 am working on a script to automate the test.</p>
 181
 182 <p>The idea is to create a Lenny chroot and use tasksel to install a
 183 Gnome or KDE desktop installation inside the chroot before upgrading
 184 it.  To ensure no services are started in the chroot, a policy-rc.d
 185 script is inserted.  To make sure tasksel believe it is to install a
 186 desktop on a laptop, the tasksel tests are replaced in the chroot
 187 (only acceptable because this is a throw-away chroot).</p>
 188
 189 <p>A naive upgrade from Lenny to Squeeze using aptitude dist-upgrade
 190 currently always fail because udev refuses to upgrade with the kernel
 191 in Lenny, so to avoid that problem the file /etc/udev/kernel-upgrade
 192 is created.  The bug report
 193 <a href="http://bugs.debian.org/566000">#566000</a> make me suspect
 194 this problem do not trigger in a chroot, but I touch the file anyway
 195 to make sure the upgrade go well.  Testing on virtual and real
 196 hardware have failed me because of udev so far, and creating this file
 197 do the trick in such settings anyway.  This is a
 198 <a href="http://www.linuxquestions.org/questions/debian-26/failed-dist-upgrade-due-to-udev-config_sysfs_deprecated-nonsense-804130/">known
 199 issue</a> and the current udev behaviour is intended by the udev
 200 maintainer because he lack the resources to rewrite udev to keep
 201 working with old kernels or something like that.  I really wish the
 202 udev upstream would keep udev backwards compatible, to avoid such
 203 upgrade problem, but given that they fail to do so, I guess
 204 documenting the way out of this mess is the best option we got for
 205 Debian Squeeze.</p>
 206
 207 <p>Anyway, back to the task at hand, testing upgrades.  This test
 208 script, which I call <tt>upgrade-test</tt> for now, is doing the
 209 trick:</p>
 210
 211 <blockquote><pre>
 212 #!/bin/sh
 213 set -ex
 214
 215 if [ "$1" ] ; then
 216     desktop=$1
 217 else
 218     desktop=gnome
 219 fi
 220
 221 from=lenny
 222 to=squeeze
 223
 224 exec &lt; /dev/null
 225 unset LANG
 226 mirror=http://ftp.skolelinux.org/debian
 227 tmpdir=chroot-$from-upgrade-$to-$desktop
 228 fuser -mv .
 229 debootstrap $from $tmpdir $mirror
 230 chroot $tmpdir aptitude update
 231 cat > $tmpdir/usr/sbin/policy-rc.d &lt;&lt;EOF
 232 #!/bin/sh
 233 exit 101
 234 EOF
 235 chmod a+rx $tmpdir/usr/sbin/policy-rc.d
 236 exit_cleanup() {
 237     umount $tmpdir/proc
 238 }
 239 mount -t proc proc $tmpdir/proc
 240 # Make sure proc is unmounted also on failure
 241 trap exit_cleanup EXIT INT
 242
 243 chroot $tmpdir aptitude -y install debconf-utils
 244
 245 # Make sure tasksel autoselection trigger.  It need the test scripts
 246 # to return the correct answers.
 247 echo tasksel tasksel/desktop multiselect $desktop | \
 248     chroot $tmpdir debconf-set-selections
 249
 250 # Include the desktop and laptop task
 251 for test in desktop laptop ; do
 252     echo > $tmpdir/usr/lib/tasksel/tests/$test &lt;&lt;EOF
 253 #!/bin/sh
 254 exit 2
 255 EOF
 256     chmod a+rx $tmpdir/usr/lib/tasksel/tests/$test
 257 done
 258
 259 DEBIAN_FRONTEND=noninteractive
 260 DEBIAN_PRIORITY=critical
 261 export DEBIAN_FRONTEND DEBIAN_PRIORITY
 262 chroot $tmpdir tasksel --new-install
 263
 264 echo deb $mirror $to main > $tmpdir/etc/apt/sources.list
 265 chroot $tmpdir aptitude update
 266 touch $tmpdir/etc/udev/kernel-upgrade
 267 chroot $tmpdir aptitude -y dist-upgrade
 268 fuser -mv
 269 </pre></blockquote>
 270
 271 <p>I suspect it would be useful to test upgrades with both apt-get and
 272 with aptitude, but I have not had time to look at how they behave
 273 differently so far.  I hope to get a cron job running to do the test
 274 regularly and post the result on the web.  The Gnome upgrade currently
 275 work, while the KDE upgrade fail because of the bug in
 276 kdebase-workspace-data</p>
 277
 278 <p>I am not quite sure what kind of extract from the huge upgrade logs
 279 (KDE 167 KiB, Gnome 516 KiB) it make sense to include in this blog
 280 post, so I will refrain from trying.  I can report that for Gnome,
 281 aptitude report 760 packages upgraded, 448 newly installed, 129 to
 282 remove and 1 not upgraded and 1024MB need to be downloaded while for
 283 KDE the same numbers are 702 packages upgraded, 507 newly installed,
 284 193 to remove and 0 not upgraded and 1117MB need to be downloaded</p>
 285
 286 <p>I am very happy to notice that the Gnome desktop + laptop upgrade
 287 is able to migrate to dependency based boot sequencing and parallel
 288 booting without a hitch.  Was unsure if there were still bugs with
 289 packages failing to clean up their obsolete init.d script during
 290 upgrades, and no such problem seem to affect the Gnome desktop+laptop
 291 packages.</p>
 292
 293       </div>
 294       <div class="tags">
 295
 296
 297         Tags: <a href="http://people.skolelinux.org/pere/blog/tags/bootsystem">bootsystem</a>, <a href="http://people.skolelinux.org/pere/blog/tags/debian">debian</a>, <a href="http://people.skolelinux.org/pere/blog/tags/debian edu">debian edu</a>, <a href="http://people.skolelinux.org/pere/blog/tags/english">english</a>.
 298
 299
 300       </div>
 301     </div>
 302     <div class="padding"></div>
 303
 304     <div class="entry">
 305       <div class="title">
 306         <a href="http://people.skolelinux.org/pere/blog/Upstart_or_sysvinit___as_init_d_scripts_see_it.html">Upstart or sysvinit - as init.d scripts see it</a>
 307       </div>
 308       <div class="date">
 309          6th June 2010
 310       </div>
 311       <div class="body">
 312         <p>If Debian is to migrate to upstart on Linux, I expect some init.d
 313 scripts to migrate (some of) their operations to upstart job while
 314 keeping the init.d for hurd and kfreebsd.  The packages with such
 315 needs will need a way to get their init.d scripts to behave
 316 differently when used with sysvinit and with upstart.  Because of
 317 this, I had a look at the environment variables set when a init.d
 318 script is running under upstart, and when it is not.</p>
 319
 320 <p>With upstart, I notice these environment variables are set when a
 321 script is started from rcS.d/ (ignoring some irrelevant ones like
 322 COLUMNS):</p>
 323
 324 <blockquote><pre>
 325 DEFAULT_RUNLEVEL=2
 326 previous=N
 327 PREVLEVEL=
 328 RUNLEVEL=
 329 runlevel=S
 330 UPSTART_EVENTS=startup
 331 UPSTART_INSTANCE=
 332 UPSTART_JOB=rc-sysinit
 333 </pre></blockquote>
 334
 335 <p>With sysvinit, these environment variables are set for the same
 336 script.</p>
 337
 338 <blockquote><pre>
 339 INIT_VERSION=sysvinit-2.88
 340 previous=N
 341 PREVLEVEL=N
 342 RUNLEVEL=S
 343 runlevel=S
 344 </pre></blockquote>
 345
 346 <p>The RUNLEVEL and PREVLEVEL environment variables passed on from
 347 sysvinit are not set by upstart.  Not sure if it is intentional or not
 348 to not be compatible with sysvinit in this regard.</p>
 349
 350 <p>For scripts needing to behave differently when upstart is used,
 351 looking for the UPSTART_JOB environment variable seem to be a good
 352 choice.</p>
 353
 354       </div>
 355       <div class="tags">
 356
 357
 358         Tags: <a href="http://people.skolelinux.org/pere/blog/tags/bootsystem">bootsystem</a>, <a href="http://people.skolelinux.org/pere/blog/tags/debian">debian</a>, <a href="http://people.skolelinux.org/pere/blog/tags/english">english</a>.
 359
 360
 361       </div>
 362     </div>
 363     <div class="padding"></div>
 364
 365     <div class="entry">
 366       <div class="title">
 367         <a href="http://people.skolelinux.org/pere/blog/KDM_fail_at_boot_with_NVidia_cards___and_no_one_try_to_fix_it_.html">KDM fail at boot with NVidia cards - and no one try to fix it?</a>
 368       </div>
 369       <div class="date">
 370          1st June 2010
 371       </div>
 372       <div class="body">
 373         <p>It is strange to watch how a bug in Debian causing KDM to fail to
 374 start at boot when an NVidia video card is used is handled.  The
 375 problem seem to be that the nvidia X.org driver uses a long time to
 376 initialize, and this duration is longer than kdm is configured to
 377 wait.</p>
 378
 379 <p>I came across two bugs related to this issue,
 380 <a href="http://bugs.debian.org/583312">#583312</a> initially filed
 381 against initscripts and passed on to nvidia-glx when it became obvious
 382 that the nvidia drivers were involved, and
 383 <a href="http://bugs.debian.org/524751">#524751</a> initially filed against
 384 kdm and passed on to src:nvidia-graphics-drivers for unknown reasons.</p>
 385
 386 <p>To me, it seem that no-one is interested in actually solving the
 387 problem nvidia video card owners experience and make sure the Debian
 388 distribution work out of the box for these users.  The nvidia driver
 389 maintainers expect kdm to be set up to wait longer, while kdm expect
 390 the nvidia driver maintainers to fix the driver to start faster, and
 391 while they wait for each other I guess the users end up switching to a
 392 distribution that work for them.  I have no idea what the solution is,
 393 but I am pretty sure that waiting for each other is not it.</p>
 394
 395 <p>I wonder why we end up handling bugs this way.</p>
 396
 397       </div>
 398       <div class="tags">
 399
 400
 401         Tags: <a href="http://people.skolelinux.org/pere/blog/tags/bootsystem">bootsystem</a>, <a href="http://people.skolelinux.org/pere/blog/tags/debian">debian</a>, <a href="http://people.skolelinux.org/pere/blog/tags/debian edu">debian edu</a>, <a href="http://people.skolelinux.org/pere/blog/tags/english">english</a>.
 402
 403
 404       </div>
 405     </div>
 406     <div class="padding"></div>
 407
 408     <div class="entry">
 409       <div class="title">
 410         <a href="http://people.skolelinux.org/pere/blog/Parallellized_boot_seem_to_hold_up_well_in_Debian_testing.html">Parallellized boot seem to hold up well in Debian/testing</a>
 411       </div>
 412       <div class="date">
 413         27th May 2010
 414       </div>
 415       <div class="body">
 416         <p>A few days ago, parallel booting was enabled in Debian/testing.
 417 The feature seem to hold up pretty well, but three fairly serious
 418 issues are known and should be solved:
 419
 420 <p><ul>
 421
 422 <li>The wicd package seen to
 423 <a href="http://bugs.debian.org/508289">break NFS mounting</a> and
 424 <a href="http://bugs.debian.org/581586">network setup</a> when
 425 parallel booting is enabled.  No idea why, but the wicd maintainer
 426 seem to be on the case.</li>
 427
 428 <li>The nvidia X driver seem to
 429 <a href="http://bugs.debian.org/583312">have a race condition</a>
 430 triggered more easily when parallel booting is in effect.  The
 431 maintainer is on the case.</li>
 432
 433 <li>The sysv-rc package fail to properly enable dependency based boot
 434 sequencing (the shutdown is broken) when old file-rc users
 435 <a href="http://bugs.debian.org/575080">try to switch back</a> to
 436 sysv-rc.  One way to solve it would be for file-rc to create
 437 /etc/init.d/.legacy-bootordering, and another is to try to make
 438 sysv-rc more robust.  Will investigate some more and probably upload a
 439 workaround in sysv-rc to help those trying to move from file-rc to
 440 sysv-rc get a working shutdown.</li>
 441
 442 </ul></p>
 443
 444 <p>All in all not many surprising issues, and all of them seem
 445 solvable before Squeeze is released.  In addition to these there are
 446 some packages with bugs in their dependencies and run level settings,
 447 which I expect will be fixed in a reasonable time span.</p>
 448
 449 <p>If you report any problems with dependencies in init.d scripts to
 450 the BTS, please usertag the report to get it to show up at
 451 <a href="http://bugs.debian.org/cgi-bin/pkgreport.cgi?users=initscripts-ng-devel@lists.alioth.debian.org">the
 452 list of usertagged bugs related to this</a>.</p>
 453
 454 <p>Update: Correct bug number to file-rc issue.</p>
 455
 456       </div>
 457       <div class="tags">
 458
 459
 460         Tags: <a href="http://people.skolelinux.org/pere/blog/tags/bootsystem">bootsystem</a>, <a href="http://people.skolelinux.org/pere/blog/tags/debian">debian</a>, <a href="http://people.skolelinux.org/pere/blog/tags/debian edu">debian edu</a>, <a href="http://people.skolelinux.org/pere/blog/tags/english">english</a>.
 461
 462
 463       </div>
 464     </div>
 465     <div class="padding"></div>
 466
 467     <div class="entry">
 468       <div class="title">
 469         <a href="http://people.skolelinux.org/pere/blog/Parallellized_boot_is_now_the_default_in_Debian_unstable.html">Parallellized boot is now the default in Debian/unstable</a>
 470       </div>
 471       <div class="date">
 472         14th May 2010
 473       </div>
 474       <div class="body">
 475         <p>Since this evening, parallel booting is the default in
 476 Debian/unstable for machines using dependency based boot sequencing.
 477 Apparently the testing of concurrent booting has been wider than
 478 expected, if I am to believe the
 479 <a href="http://lists.debian.org/debian-devel/2010/05/msg00122.html">input
 480 on debian-devel@</a>, and I concluded a few days ago to move forward
 481 with the feature this weekend, to give us some time to detect any
 482 remaining problems before Squeeze is frozen.  If serious problems are
 483 detected, it is simple to change the default back to sequential boot.
 484 The upload of the new sysvinit package also activate a new upstream
 485 version.</p>
 486
 487 More information about
 488 <a href="http://wiki.debian.org/LSBInitScripts/DependencyBasedBoot">dependency
 489 based boot sequencing</a> is available from the Debian wiki.  It is
 490 currently possible to disable parallel booting when one run into
 491 problems caused by it, by adding this line to /etc/default/rcS:</p>
 492
 493 <blockquote><pre>
 494 CONCURRENCY=none
 495 </pre></blockquote>
 496
 497 <p>If you report any problems with dependencies in init.d scripts to
 498 the BTS, please usertag the report to get it to show up at
 499 <a href="http://bugs.debian.org/cgi-bin/pkgreport.cgi?users=initscripts-ng-devel@lists.alioth.debian.org">the
 500 list of usertagged bugs related to this</a>.</p>
 501
 502       </div>
 503       <div class="tags">
 504
 505
 506         Tags: <a href="http://people.skolelinux.org/pere/blog/tags/bootsystem">bootsystem</a>, <a href="http://people.skolelinux.org/pere/blog/tags/debian">debian</a>, <a href="http://people.skolelinux.org/pere/blog/tags/debian edu">debian edu</a>, <a href="http://people.skolelinux.org/pere/blog/tags/english">english</a>.
 507
 508
 509       </div>
 510     </div>
 511     <div class="padding"></div>
 512
 513     <div class="entry">
 514       <div class="title">
 515         <a href="http://people.skolelinux.org/pere/blog/systemd__an_interesting_alternative_to_upstart.html">systemd, an interesting alternative to upstart</a>
 516       </div>
 517       <div class="date">
 518         13th May 2010
 519       </div>
 520       <div class="body">
 521         <p>The last few days a new boot system called
 522 <a href="http://www.freedesktop.org/wiki/Software/systemd">systemd</a>
 523 has been
 524 <a href="http://0pointer.de/blog/projects/systemd.html">introduced</a>
 525
 526 to the free software world.  I have not yet had time to play around
 527 with it, but it seem to be a very interesting alternative to
 528 <a href="http://upstart.ubuntu.com/">upstart</a>, and might prove to be
 529 a good alternative for Debian when we are able to switch to an event
 530 based boot system.  Tollef is
 531 <a href="http://bugs.debian.org/580814">in the process</a> of getting
 532 systemd into Debian, and I look forward to seeing how well it work.  I
 533 like the fact that systemd handles init.d scripts with dependency
 534 information natively, allowing them to run in parallel where upstart
 535 at the moment do not.</p>
 536
 537 <p>Unfortunately do systemd have the same problem as upstart regarding
 538 platform support.  It only work on recent Linux kernels, and also need
 539 some new kernel features enabled to function properly.  This means
 540 kFreeBSD and Hurd ports of Debian will need a port or a different boot
 541 system.  Not sure how that will be handled if systemd proves to be the
 542 way forward.</p>
 543
 544 <p>In the mean time, based on the
 545 <a href="http://lists.debian.org/debian-devel/2010/05/msg00122.html">input
 546 on debian-devel@</a> regarding parallel booting in Debian, I have
 547 decided to enable full parallel booting as the default in Debian as
 548 soon as possible (probably this weekend or early next week), to see if
 549 there are any remaining serious bugs in the init.d dependencies.  A
 550 new version of the sysvinit package implementing this change is
 551 already in experimental.  If all go well, Squeeze will be released
 552 with parallel booting enabled by default.</p>
 553
 554       </div>
 555       <div class="tags">
 556
 557
 558         Tags: <a href="http://people.skolelinux.org/pere/blog/tags/bootsystem">bootsystem</a>, <a href="http://people.skolelinux.org/pere/blog/tags/debian">debian</a>, <a href="http://people.skolelinux.org/pere/blog/tags/english">english</a>, <a href="http://people.skolelinux.org/pere/blog/tags/nuug">nuug</a>.
 559
 560
 561       </div>
 562     </div>
 563     <div class="padding"></div>
 564
 565     <div class="entry">
 566       <div class="title">
 567         <a href="http://people.skolelinux.org/pere/blog/Parallellizing_the_boot_in_Debian_Squeeze___ready_for_wider_testing.html">Parallellizing the boot in Debian Squeeze - ready for wider testing</a>
 568       </div>
 569       <div class="date">
 570          6th May 2010
 571       </div>
 572       <div class="body">
 573         <p>These days, the init.d script dependencies in Squeeze are quite
 574 complete, so complete that it is actually possible to run all the
 575 init.d scripts in parallell based on these dependencies.  If you want
 576 to test your Squeeze system, make sure
 577 <a href="http://wiki.debian.org/LSBInitScripts/DependencyBasedBoot">dependency
 578 based boot sequencing</a> is enabled, and add this line to
 579 /etc/default/rcS:</p>
 580
 581 <blockquote><pre>
 582 CONCURRENCY=makefile
 583 </pre></blockquote>
 584
 585 <p>That is it.  It will cause sysv-rc to use the startpar tool to run
 586 scripts in parallel using the dependency information stored in
 587 /etc/init.d/.depend.boot, /etc/init.d/.depend.start and
 588 /etc/init.d/.depend.stop to order the scripts.  Startpar is configured
 589 to try to start the kdm and gdm scripts as early as possible, and will
 590 start the facilities required by kdm or gdm as early as possible to
 591 make this happen.</p>
 592
 593 <p>Give it a try, and see if you like the result.  If some services
 594 fail to start properly, it is most likely because they have incomplete
 595 init.d script dependencies in their startup script (or some of their
 596 dependent scripts have incomplete dependencies).  Report bugs and get
 597 the package maintainers to fix it. :)</p>
 598
 599 <p>Running scripts in parallel could be the default in Debian when we
 600 manage to get the init.d script dependencies complete and correct.  I
 601 expect we will get there in Squeeze+1, if we get manage to test and
 602 fix the remaining issues.</p>
 603
 604 <p>If you report any problems with dependencies in init.d scripts to
 605 the BTS, please usertag the report to get it to show up at
 606 <a href="http://bugs.debian.org/cgi-bin/pkgreport.cgi?users=initscripts-ng-devel@lists.alioth.debian.org">the
 607 list of usertagged bugs related to this</a>.</p>
 608
 609       </div>
 610       <div class="tags">
 611
 612
 613         Tags: <a href="http://people.skolelinux.org/pere/blog/tags/bootsystem">bootsystem</a>, <a href="http://people.skolelinux.org/pere/blog/tags/debian">debian</a>, <a href="http://people.skolelinux.org/pere/blog/tags/english">english</a>.
 614
 615
 616       </div>
 617     </div>
 618     <div class="padding"></div>
 619
 620     <div class="entry">
 621       <div class="title">
 622         <a href="http://people.skolelinux.org/pere/blog/Debian_has_switched_to_dependency_based_boot_sequencing.html">Debian has switched to dependency based boot sequencing</a>
 623       </div>
 624       <div class="date">
 625         27th July 2009
 626       </div>
 627       <div class="body">
 628         <p>Since this evening, with the upload of sysvinit version 2.87dsf-2,
 629 and the upload of insserv version 1.12.0-10 yesterday, Debian unstable
 630 have been migrated to using dependency based boot sequencing.  This
 631 conclude work me and others have been doing for the last three days.
 632 It feels great to see this finally part of the default Debian
 633 installation.  Now we just need to weed out the last few problems that
 634 are bound to show up, to get everything ready for Squeeze.</p>
 635
 636 <p>The next step is migrating /sbin/init from sysvinit to upstart, and
 637 fixing the more fundamental problem of handing the event based
 638 non-predictable kernel in the early boot.</p>
 639
 640       </div>
 641       <div class="tags">
 642
 643
 644         Tags: <a href="http://people.skolelinux.org/pere/blog/tags/bootsystem">bootsystem</a>, <a href="http://people.skolelinux.org/pere/blog/tags/debian">debian</a>, <a href="http://people.skolelinux.org/pere/blog/tags/english">english</a>, <a href="http://people.skolelinux.org/pere/blog/tags/nuug">nuug</a>.
 645
 646
 647       </div>
 648     </div>
 649     <div class="padding"></div>
 650
 651     <div class="entry">
 652       <div class="title">
 653         <a href="http://people.skolelinux.org/pere/blog/Taking_over_sysvinit_development.html">Taking over sysvinit development</a>
 654       </div>
 655       <div class="date">
 656         22nd July 2009
 657       </div>
 658       <div class="body">
 659         <p>After several years of frustration with the lack of activity from
 660 the existing sysvinit upstream developer, I decided a few weeks ago to
 661 take over the package and become the new upstream.  The number of
 662 patches to track for the Debian package was becoming a burden, and the
 663 lack of synchronization between the distribution made it hard to keep
 664 the package up to date.</p>
 665
 666 <p>On the new sysvinit team is the SuSe maintainer Dr. Werner Fink,
 667 and my Debian co-maintainer Kel Modderman.  About 10 days ago, I made
 668 a new upstream tarball with version number 2.87dsf (for Debian, SuSe
 669 and Fedora), based on the patches currently in use in these
 670 distributions.  We Debian maintainers plan to move to this tarball as
 671 the new upstream as soon as we find time to do the merge.  Since the
 672 new tarball was created, we agreed with Werner at SuSe to make a new
 673 upstream project at <a href="http://savannah.nongnu.org/">Savannah</a>, and continue
 674 development there.  The project is registered and currently waiting
 675 for approval by the Savannah administrators, and as soon as it is
 676 approved, we will import the old versions from svn and continue
 677 working on the future release.</p>
 678
 679 <p>It is a bit ironic that this is done now, when some of the involved
 680 distributions are moving to upstart as a syvinit replacement.</p>
 681
 682       </div>
 683       <div class="tags">
 684
 685
 686         Tags: <a href="http://people.skolelinux.org/pere/blog/tags/bootsystem">bootsystem</a>, <a href="http://people.skolelinux.org/pere/blog/tags/debian">debian</a>, <a href="http://people.skolelinux.org/pere/blog/tags/english">english</a>, <a href="http://people.skolelinux.org/pere/blog/tags/nuug">nuug</a>.
 687
 688
 689       </div>
 690     </div>
 691     <div class="padding"></div>
 692
 693     <div class="entry">
 694       <div class="title">
 695         <a href="http://people.skolelinux.org/pere/blog/Debian_boots_quicker_and_quicker.html">Debian boots quicker and quicker</a>
 696       </div>
 697       <div class="date">
 698         24th June 2009
 699       </div>
 700       <div class="body">
 701         <p>I spent Monday and tuesday this week in London with a lot of the
 702 people involved in the boot system on Debian and Ubuntu, to see if we
 703 could find more ways to speed up the boot system.  This was an Ubuntu
 704 funded
 705 <a href="https://wiki.ubuntu.com/FoundationsTeam/BootPerformance/DebianUbuntuSprint">developer
 706 gathering</a>. It was quite productive.  We also discussed the future
 707 of boot systems, and ways to handle the increasing number of boot
 708 issues introduced by the Linux kernel becoming more and more
 709 asynchronous and event base.  The Ubuntu approach using udev and
 710 upstart might be a good way forward.  Time will show.</p>
 711
 712 <p>Anyway, there are a few ways at the moment to speed up the boot
 713 process in Debian.  All of these should be applied to get a quick
 714 boot:</p>
 715
 716 <ul>
 717
 718 <li>Use dash as /bin/sh.</li>
 719
 720 <li>Disable the init.d/hwclock*.sh scripts and make sure the hardware
 721    clock is in UTC.</li>
 722
 723 <li>Install and activate the insserv package to enable
 724    <a href="http://wiki.debian.org/LSBInitScripts/DependencyBasedBoot">dependency
 725    based boot sequencing</a>, and enable concurrent booting.</li>
 726
 727 </ul>
 728
 729 These points are based on the Google summer of code work done by
 730 <a href="http://initscripts-ng.alioth.debian.org/soc2006-bootsystem/">Carlos
 731 Villegas</a>.
 732
 733 <p>Support for makefile-style concurrency during boot was uploaded to
 734 unstable yesterday.  When we tested it, we were able to cut 6 seconds
 735 from the boot sequence.  It depend on very correct dependency
 736 declaration in all init.d scripts, so I expect us to find edge cases
 737 where the dependences in some scripts are slightly wrong when we start
 738 using this.</p>
 739
 740 <p>On our IRC channel for this effort, #pkg-sysvinit, a new idea was
 741 introduced by Raphael Geissert today, one that could affect the
 742 startup speed as well.  Instead of starting some scripts concurrently
 743 from rcS.d/ and another set of scripts from rc2.d/, it would be
 744 possible to run a of them in the same process.  A quick way to test
 745 this would be to enable insserv and run 'mv /etc/rc2.d/S* /etc/rcS.d/;
 746 insserv'.  Will need to test if that work. :)</p>
 747
 748       </div>
 749       <div class="tags">
 750
 751
 752         Tags: <a href="http://people.skolelinux.org/pere/blog/tags/bootsystem">bootsystem</a>, <a href="http://people.skolelinux.org/pere/blog/tags/debian">debian</a>, <a href="http://people.skolelinux.org/pere/blog/tags/english">english</a>.
 753
 754
 755       </div>
 756     </div>
 757     <div class="padding"></div>
 758
 759     <p style="text-align: right;"><a href="bootsystem.rss"><img src="http://people.skolelinux.org/pere/blog/xml.gif" alt="RSS Feed" width="36" height="14" /></a></p>
 760     <div id="sidebar">
 761
 762
 763
 764 <h2>Archive</h2>
 765 <ul>
 766
 767 <li>2012
 768 <ul>
 769
 770 <li><a href="http://people.skolelinux.org/pere/blog/archive/2012/01/">January (7)</a></li>
 771
 772 <li><a href="http://people.skolelinux.org/pere/blog/archive/2012/02/">February (10)</a></li>
 773
 774 <li><a href="http://people.skolelinux.org/pere/blog/archive/2012/03/">March (17)</a></li>
 775
 776 <li><a href="http://people.skolelinux.org/pere/blog/archive/2012/04/">April (12)</a></li>
 777
 778 <li><a href="http://people.skolelinux.org/pere/blog/archive/2012/05/">May (12)</a></li>
 779
 780 <li><a href="http://people.skolelinux.org/pere/blog/archive/2012/06/">June (20)</a></li>
 781
 782 <li><a href="http://people.skolelinux.org/pere/blog/archive/2012/07/">July (17)</a></li>
 783
 784 <li><a href="http://people.skolelinux.org/pere/blog/archive/2012/08/">August (6)</a></li>
 785
 786 <li><a href="http://people.skolelinux.org/pere/blog/archive/2012/09/">September (9)</a></li>
 787
 788 <li><a href="http://people.skolelinux.org/pere/blog/archive/2012/10/">October (17)</a></li>
 789
 790 <li><a href="http://people.skolelinux.org/pere/blog/archive/2012/11/">November (10)</a></li>
 791
 792 <li><a href="http://people.skolelinux.org/pere/blog/archive/2012/12/">December (2)</a></li>
 793
 794 </ul></li>
 795
 796 <li>2011
 797 <ul>
 798
 799 <li><a href="http://people.skolelinux.org/pere/blog/archive/2011/01/">January (16)</a></li>
 800
 801 <li><a href="http://people.skolelinux.org/pere/blog/archive/2011/02/">February (6)</a></li>
 802
 803 <li><a href="http://people.skolelinux.org/pere/blog/archive/2011/03/">March (6)</a></li>
 804
 805 <li><a href="http://people.skolelinux.org/pere/blog/archive/2011/04/">April (7)</a></li>
 806
 807 <li><a href="http://people.skolelinux.org/pere/blog/archive/2011/05/">May (3)</a></li>
 808
 809 <li><a href="http://people.skolelinux.org/pere/blog/archive/2011/06/">June (2)</a></li>
 810
 811 <li><a href="http://people.skolelinux.org/pere/blog/archive/2011/07/">July (7)</a></li>
 812
 813 <li><a href="http://people.skolelinux.org/pere/blog/archive/2011/08/">August (6)</a></li>
 814
 815 <li><a href="http://people.skolelinux.org/pere/blog/archive/2011/09/">September (4)</a></li>
 816
 817 <li><a href="http://people.skolelinux.org/pere/blog/archive/2011/10/">October (2)</a></li>
 818
 819 <li><a href="http://people.skolelinux.org/pere/blog/archive/2011/11/">November (3)</a></li>
 820
 821 <li><a href="http://people.skolelinux.org/pere/blog/archive/2011/12/">December (1)</a></li>
 822
 823 </ul></li>
 824
 825 <li>2010
 826 <ul>
 827
 828 <li><a href="http://people.skolelinux.org/pere/blog/archive/2010/01/">January (2)</a></li>
 829
 830 <li><a href="http://people.skolelinux.org/pere/blog/archive/2010/02/">February (1)</a></li>
 831
 832 <li><a href="http://people.skolelinux.org/pere/blog/archive/2010/03/">March (3)</a></li>
 833
 834 <li><a href="http://people.skolelinux.org/pere/blog/archive/2010/04/">April (3)</a></li>
 835
 836 <li><a href="http://people.skolelinux.org/pere/blog/archive/2010/05/">May (9)</a></li>
 837
 838 <li><a href="http://people.skolelinux.org/pere/blog/archive/2010/06/">June (14)</a></li>
 839
 840 <li><a href="http://people.skolelinux.org/pere/blog/archive/2010/07/">July (12)</a></li>
 841
 842 <li><a href="http://people.skolelinux.org/pere/blog/archive/2010/08/">August (13)</a></li>
 843
 844 <li><a href="http://people.skolelinux.org/pere/blog/archive/2010/09/">September (7)</a></li>
 845
 846 <li><a href="http://people.skolelinux.org/pere/blog/archive/2010/10/">October (9)</a></li>
 847
 848 <li><a href="http://people.skolelinux.org/pere/blog/archive/2010/11/">November (13)</a></li>
 849
 850 <li><a href="http://people.skolelinux.org/pere/blog/archive/2010/12/">December (12)</a></li>
 851
 852 </ul></li>
 853
 854 <li>2009
 855 <ul>
 856
 857 <li><a href="http://people.skolelinux.org/pere/blog/archive/2009/01/">January (8)</a></li>
 858
 859 <li><a href="http://people.skolelinux.org/pere/blog/archive/2009/02/">February (8)</a></li>
 860
 861 <li><a href="http://people.skolelinux.org/pere/blog/archive/2009/03/">March (12)</a></li>
 862
 863 <li><a href="http://people.skolelinux.org/pere/blog/archive/2009/04/">April (10)</a></li>
 864
 865 <li><a href="http://people.skolelinux.org/pere/blog/archive/2009/05/">May (9)</a></li>
 866
 867 <li><a href="http://people.skolelinux.org/pere/blog/archive/2009/06/">June (3)</a></li>
 868
 869 <li><a href="http://people.skolelinux.org/pere/blog/archive/2009/07/">July (4)</a></li>
 870
 871 <li><a href="http://people.skolelinux.org/pere/blog/archive/2009/08/">August (3)</a></li>
 872
 873 <li><a href="http://people.skolelinux.org/pere/blog/archive/2009/09/">September (1)</a></li>
 874
 875 <li><a href="http://people.skolelinux.org/pere/blog/archive/2009/10/">October (2)</a></li>
 876
 877 <li><a href="http://people.skolelinux.org/pere/blog/archive/2009/11/">November (3)</a></li>
 878
 879 <li><a href="http://people.skolelinux.org/pere/blog/archive/2009/12/">December (3)</a></li>
 880
 881 </ul></li>
 882
 883 <li>2008
 884 <ul>
 885
 886 <li><a href="http://people.skolelinux.org/pere/blog/archive/2008/11/">November (5)</a></li>
 887
 888 <li><a href="http://people.skolelinux.org/pere/blog/archive/2008/12/">December (7)</a></li>
 889
 890 </ul></li>
 891
 892 </ul>
 893
 894
 895
 896 <h2>Tags</h2>
 897 <ul>
 898
 899  <li><a href="http://people.skolelinux.org/pere/blog/tags/3d-printer">3d-printer (13)</a></li>
 900
 901  <li><a href="http://people.skolelinux.org/pere/blog/tags/amiga">amiga (1)</a></li>
 902
 903  <li><a href="http://people.skolelinux.org/pere/blog/tags/aros">aros (1)</a></li>
 904
 905  <li><a href="http://people.skolelinux.org/pere/blog/tags/bankid">bankid (3)</a></li>
 906
 907  <li><a href="http://people.skolelinux.org/pere/blog/tags/bitcoin">bitcoin (3)</a></li>
 908
 909  <li><a href="http://people.skolelinux.org/pere/blog/tags/bootsystem">bootsystem (12)</a></li>
 910
 911  <li><a href="http://people.skolelinux.org/pere/blog/tags/bsa">bsa (2)</a></li>
 912
 913  <li><a href="http://people.skolelinux.org/pere/blog/tags/debian">debian (57)</a></li>
 914
 915  <li><a href="http://people.skolelinux.org/pere/blog/tags/debian edu">debian edu (116)</a></li>
 916
 917  <li><a href="http://people.skolelinux.org/pere/blog/tags/digistan">digistan (9)</a></li>
 918
 919  <li><a href="http://people.skolelinux.org/pere/blog/tags/docbook">docbook (7)</a></li>
 920
 921  <li><a href="http://people.skolelinux.org/pere/blog/tags/drivstoffpriser">drivstoffpriser (4)</a></li>
 922
 923  <li><a href="http://people.skolelinux.org/pere/blog/tags/english">english (161)</a></li>
 924
 925  <li><a href="http://people.skolelinux.org/pere/blog/tags/fiksgatami">fiksgatami (21)</a></li>
 926
 927  <li><a href="http://people.skolelinux.org/pere/blog/tags/fildeling">fildeling (12)</a></li>
 928
 929  <li><a href="http://people.skolelinux.org/pere/blog/tags/freeculture">freeculture (9)</a></li>
 930
 931  <li><a href="http://people.skolelinux.org/pere/blog/tags/frikanalen">frikanalen (9)</a></li>
 932
 933  <li><a href="http://people.skolelinux.org/pere/blog/tags/intervju">intervju (32)</a></li>
 934
 935  <li><a href="http://people.skolelinux.org/pere/blog/tags/kart">kart (17)</a></li>
 936
 937  <li><a href="http://people.skolelinux.org/pere/blog/tags/ldap">ldap (8)</a></li>
 938
 939  <li><a href="http://people.skolelinux.org/pere/blog/tags/lenker">lenker (5)</a></li>
 940
 941  <li><a href="http://people.skolelinux.org/pere/blog/tags/ltsp">ltsp (1)</a></li>
 942
 943  <li><a href="http://people.skolelinux.org/pere/blog/tags/multimedia">multimedia (25)</a></li>
 944
 945  <li><a href="http://people.skolelinux.org/pere/blog/tags/norsk">norsk (217)</a></li>
 946
 947  <li><a href="http://people.skolelinux.org/pere/blog/tags/nuug">nuug (147)</a></li>
 948
 949  <li><a href="http://people.skolelinux.org/pere/blog/tags/offentlig innsyn">offentlig innsyn (6)</a></li>
 950
 951  <li><a href="http://people.skolelinux.org/pere/blog/tags/open311">open311 (2)</a></li>
 952
 953  <li><a href="http://people.skolelinux.org/pere/blog/tags/opphavsrett">opphavsrett (39)</a></li>
 954
 955  <li><a href="http://people.skolelinux.org/pere/blog/tags/personvern">personvern (60)</a></li>
 956
 957  <li><a href="http://people.skolelinux.org/pere/blog/tags/raid">raid (1)</a></li>
 958
 959  <li><a href="http://people.skolelinux.org/pere/blog/tags/reprap">reprap (11)</a></li>
 960
 961  <li><a href="http://people.skolelinux.org/pere/blog/tags/rfid">rfid (2)</a></li>
 962
 963  <li><a href="http://people.skolelinux.org/pere/blog/tags/robot">robot (4)</a></li>
 964
 965  <li><a href="http://people.skolelinux.org/pere/blog/tags/rss">rss (1)</a></li>
 966
 967  <li><a href="http://people.skolelinux.org/pere/blog/tags/ruter">ruter (4)</a></li>
 968
 969  <li><a href="http://people.skolelinux.org/pere/blog/tags/scraperwiki">scraperwiki (2)</a></li>
 970
 971  <li><a href="http://people.skolelinux.org/pere/blog/tags/sikkerhet">sikkerhet (28)</a></li>
 972
 973  <li><a href="http://people.skolelinux.org/pere/blog/tags/sitesummary">sitesummary (4)</a></li>
 974
 975  <li><a href="http://people.skolelinux.org/pere/blog/tags/skepsis">skepsis (4)</a></li>
 976
 977  <li><a href="http://people.skolelinux.org/pere/blog/tags/standard">standard (39)</a></li>
 978
 979  <li><a href="http://people.skolelinux.org/pere/blog/tags/stavekontroll">stavekontroll (3)</a></li>
 980
 981  <li><a href="http://people.skolelinux.org/pere/blog/tags/stortinget">stortinget (5)</a></li>
 982
 983  <li><a href="http://people.skolelinux.org/pere/blog/tags/surveillance">surveillance (12)</a></li>
 984
 985  <li><a href="http://people.skolelinux.org/pere/blog/tags/sysadmin">sysadmin (1)</a></li>
 986
 987  <li><a href="http://people.skolelinux.org/pere/blog/tags/valg">valg (7)</a></li>
 988
 989  <li><a href="http://people.skolelinux.org/pere/blog/tags/video">video (35)</a></li>
 990
 991  <li><a href="http://people.skolelinux.org/pere/blog/tags/vitenskap">vitenskap (4)</a></li>
 992
 993  <li><a href="http://people.skolelinux.org/pere/blog/tags/web">web (26)</a></li>
 994
 995 </ul>
 996
 997
 998     </div>
 999     <p style="text-align: right">
1000  Created by <a href="http://steve.org.uk/Software/chronicle">Chronicle v4.4</a>
1001 </p>
1002
1003   </body>
1004 </html>