]> pere.pagekite.me Git - homepage.git/blob - blog/archive/2010/07/07.rss
Typo.
[homepage.git] / blog / archive / 2010 / 07 / 07.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 July 2010</title>
5 <description>Entries from July 2010</description>
6 <link>http://people.skolelinux.org/pere/blog/</link>
7
8
9 <item>
10 <title>Caching password, user and group on a roaming Debian laptop</title>
11 <link>http://people.skolelinux.org/pere/blog/Caching_password__user_and_group_on_a_roaming_Debian_laptop.html</link>
12 <guid isPermaLink="true">http://people.skolelinux.org/pere/blog/Caching_password__user_and_group_on_a_roaming_Debian_laptop.html</guid>
13 <pubDate>Thu, 1 Jul 2010 11:40:00 +0200</pubDate>
14 <description>&lt;p&gt;For a laptop, centralized user directories and password checking is
15 a bit troubling. Laptops are typically used also when not connected
16 to the network, and it is vital for a user to be able to log in or
17 unlock the screen saver also when a central server is unavailable.
18 This is possible by caching passwords and directory information (user
19 and group attributes) locally, and the packages to do so are available
20 in Debian. Here follow two recipes to set this up in Debian/Squeeze.
21 It is also possible to set up in Debian/Lenny, but require more manual
22 setup there because pam-auth-update is missing in Lenny.&lt;/p&gt;
23
24 &lt;h2&gt;LDAP/Kerberos + nscd + libpam-ccreds + libpam-mklocaluser/pam_mkhomedir&lt;/h2&gt;
25
26 This is the traditional method with a twist. The password caching is
27 provided by libpam-ccreds (version 10-4 or later is needed on
28 Squeeze), and the directory caching is done by nscd. The directory
29 lookup and password checking is done using LDAP. If one want to use
30 Kerberos for password checking the libpam-ldapd package can be
31 replaced with libpam-krb5 or libpam-heimdal. If one is happy having a
32 local home directory with the path listed in LDAP, one can use the
33 pam_mkhomedir module from pam-modules to make this happen instead of
34 using libpam-mklocaluser. A setup for pam-auth-update to enable
35 pam_mkhomedir will have to be written until a fix for
36 &lt;a href=&quot;http://bugs.debian.org/568577&quot;&gt;bug #568577&lt;/a&gt; is in the
37 archive. Because I believe it is a bad idea to have local home
38 directories using misleading paths like /site/server/partition/, I
39 prefer to create a local user with the home directory in /home/. This
40 is done using the libpam-mklocaluser package.&lt;/p&gt;
41
42 &lt;p&gt;These packages need to be installed and configured&lt;/p&gt;
43
44 &lt;blockquote&gt;&lt;pre&gt;
45 libnss-ldapd libpam-ldapd nscd libpam-ccreds libpam-mklocaluser
46 &lt;/pre&gt;&lt;/blockquote&gt;
47
48 &lt;p&gt;The ldapd packages will ask for LDAP connection information, and
49 one have to fill in the values that fits ones own site. Make sure the
50 PAM part uses encrypted connections, to make sure the password is not
51 sent in clear text to the LDAP server. I&#39;ve been unable to get TLS
52 certificate checking for a self signed certificate working, which make
53 LDAP authentication unsafe for Debian Edu (nslcd is not checking if it
54 is talking to the correct LDAP server), and very much welcome feedback
55 on how to get this working.&lt;/p&gt;
56
57 &lt;p&gt;Because nscd do not have a default configuration fit for offline
58 caching until &lt;a href=&quot;http://bugs.debian.org/485282&quot;&gt;bug #485282&lt;/a&gt;
59 is fixed, this configuration should be used instead of the one
60 currently in /etc/nscd.conf. The changes are in the fields
61 reload-count and positive-time-to-live, and is based on the
62 instructions I found in the
63 &lt;a href=&quot;http://www.flyn.org/laptopldap/&quot;&gt;LDAP for Mobile Laptops&lt;/a&gt;
64 instructions by Flyn Computing.&lt;/p&gt;
65
66 &lt;blockquote&gt;&lt;pre&gt;
67 debug-level 0
68 reload-count unlimited
69 paranoia no
70
71 enable-cache passwd yes
72 positive-time-to-live passwd 2592000
73 negative-time-to-live passwd 20
74 suggested-size passwd 211
75 check-files passwd yes
76 persistent passwd yes
77 shared passwd yes
78 max-db-size passwd 33554432
79 auto-propagate passwd yes
80
81 enable-cache group yes
82 positive-time-to-live group 2592000
83 negative-time-to-live group 20
84 suggested-size group 211
85 check-files group yes
86 persistent group yes
87 shared group yes
88 max-db-size group 33554432
89 auto-propagate group yes
90
91 enable-cache hosts no
92 positive-time-to-live hosts 2592000
93 negative-time-to-live hosts 20
94 suggested-size hosts 211
95 check-files hosts yes
96 persistent hosts yes
97 shared hosts yes
98 max-db-size hosts 33554432
99
100 enable-cache services yes
101 positive-time-to-live services 2592000
102 negative-time-to-live services 20
103 suggested-size services 211
104 check-files services yes
105 persistent services yes
106 shared services yes
107 max-db-size services 33554432
108 &lt;/pre&gt;&lt;/blockquote&gt;
109
110 &lt;p&gt;While we wait for a mechanism to update /etc/nsswitch.conf
111 automatically like the one provided in
112 &lt;a href=&quot;http://bugs.debian.org/496915&quot;&gt;bug #496915&lt;/a&gt;, the file
113 content need to be manually replaced to ensure LDAP is used as the
114 directory service on the machine. /etc/nsswitch.conf should normally
115 look like this:&lt;/p&gt;
116
117 &lt;blockquote&gt;&lt;pre&gt;
118 passwd: files ldap
119 group: files ldap
120 shadow: files ldap
121 hosts: files mdns4_minimal [NOTFOUND=return] dns mdns4
122 networks: files
123 protocols: files
124 services: files
125 ethers: files
126 rpc: files
127 netgroup: files ldap
128 &lt;/pre&gt;&lt;/blockquote&gt;
129
130 &lt;p&gt;The important parts are that ldap is listed last for passwd, group,
131 shadow and netgroup.&lt;/p&gt;
132
133 &lt;p&gt;With these changes in place, any user in LDAP will be able to log
134 in locally on the machine using for example kdm, get a local home
135 directory created and have the password as well as user and group
136 attributes cached.
137
138 &lt;h2&gt;LDAP/Kerberos + nss-updatedb + libpam-ccreds +
139 libpam-mklocaluser/pam_mkhomedir&lt;/h2&gt;
140
141 &lt;p&gt;Because nscd have had its share of problems, and seem to have
142 problems doing proper caching, I&#39;ve seen suggestions and recipes to
143 use nss-updatedb to copy parts of the LDAP database locally when the
144 LDAP database is available. I have not tested such setup, because I
145 discovered sssd.&lt;/p&gt;
146
147 &lt;h2&gt;LDAP/Kerberos + sssd + libpam-mklocaluser&lt;/h2&gt;
148
149 &lt;p&gt;A more flexible and robust setup than the nscd combination
150 mentioned earlier that has shown up recently, is the
151 &lt;a href=&quot;https://fedorahosted.org/sssd/&quot;&gt;sssd&lt;/a&gt; package from Redhat.
152 It is part of the &lt;a href=&quot;http://www.freeipa.org/&quot;&gt;FreeIPA&lt;/A&gt; project
153 to provide a Active Directory like directory service for Linux
154 machines. The sssd system combines the caching of passwords and user
155 information into one package, and remove the need for nscd and
156 libpam-ccreds. It support LDAP and Kerberos, but not NIS. Version
157 1.2 do not support netgroups, but it is said that it will support this
158 in version 1.5 expected to show up later in 2010. Because the
159 &lt;a href=&quot;http://packages.qa.debian.org/s/sssd.html&quot;&gt;sssd package&lt;/a&gt;
160 was missing in Debian, I ended up co-maintaining it with Werner, and
161 version 1.2 is now in testing.
162
163 &lt;p&gt;These packages need to be installed and configured to get the
164 roaming setup I want&lt;/p&gt;
165
166 &lt;blockquote&gt;&lt;pre&gt;
167 libpam-sss libnss-sss libpam-mklocaluser
168 &lt;/pre&gt;&lt;/blockquote&gt;
169
170 The complete setup of sssd is done by editing/creating
171 &lt;tt&gt;/etc/sssd/sssd.conf&lt;/tt&gt;.
172
173 &lt;blockquote&gt;&lt;pre&gt;
174 [sssd]
175 config_file_version = 2
176 reconnection_retries = 3
177 sbus_timeout = 30
178 services = nss, pam
179 domains = INTERN
180
181 [nss]
182 filter_groups = root
183 filter_users = root
184 reconnection_retries = 3
185
186 [pam]
187 reconnection_retries = 3
188
189 [domain/INTERN]
190 enumerate = false
191 cache_credentials = true
192
193 id_provider = ldap
194 auth_provider = ldap
195 chpass_provider = ldap
196
197 ldap_uri = ldap://ldap
198 ldap_search_base = dc=skole,dc=skolelinux,dc=no
199 ldap_tls_reqcert = never
200 ldap_tls_cacert = /etc/ssl/certs/ca-certificates.crt
201 &lt;/pre&gt;&lt;/blockquote&gt;
202
203 &lt;p&gt;I got the same problem here with certificate checking. Had to set
204 &quot;ldap_tls_reqcert = never&quot; to get it working.&lt;/p&gt;
205
206 &lt;p&gt;With the libnss-sss package in testing at the moment, the
207 nsswitch.conf file is update automatically, so there is no need to
208 modify it manually.&lt;/p&gt;
209
210 &lt;p&gt;If you want to help out with implementing this for Debian Edu,
211 please contact us on debian-edu@lists.debian.org.&lt;/p&gt;
212 </description>
213 </item>
214
215 <item>
216 <title>Lenny-&gt;Squeeze upgrades, apt vs aptitude with the Gnome desktop</title>
217 <link>http://people.skolelinux.org/pere/blog/Lenny__Squeeze_upgrades__apt_vs_aptitude_with_the_Gnome_desktop.html</link>
218 <guid isPermaLink="true">http://people.skolelinux.org/pere/blog/Lenny__Squeeze_upgrades__apt_vs_aptitude_with_the_Gnome_desktop.html</guid>
219 <pubDate>Sat, 3 Jul 2010 23:55:00 +0200</pubDate>
220 <description>&lt;p&gt;Here is a short update on my &lt;a
221 href=&quot;http://people.skolelinux.org/~pere/debian-upgrade-testing/&quot;&gt;my
222 Debian Lenny-&gt;Squeeze upgrade testing&lt;/a&gt;. Here is a summary of the
223 difference for Gnome when it is upgraded by apt-get and aptitude. I&#39;m
224 not reporting the status for KDE, because the upgrade crashes when
225 aptitude try because of missing conflicts
226 (&lt;a href=&quot;http://bugs.debian.org/584861&quot;&gt;#584861&lt;/a&gt; and
227 &lt;a href=&quot;http://bugs.debian.org/585716&quot;&gt;#585716&lt;/a&gt;).&lt;/p&gt;
228
229 &lt;p&gt;At the end of the upgrade test script, dpkg -l is executed to get a
230 complete list of the installed packages. Based on this I see these
231 differences when I did a test run today. As usual, I do not really
232 know what the correct set of packages would be, but thought it best to
233 publish the difference.&lt;/p&gt;
234
235 &lt;p&gt;Installed using apt-get, missing with aptitude&lt;/p&gt;
236
237 &lt;blockquote&gt;&lt;p&gt;
238 at-spi cpp-4.3 finger gnome-spell gstreamer0.10-gnomevfs
239 libatspi1.0-0 libcupsys2 libeel2-data libgail-common libgdl-1-common
240 libgnomeprint2.2-data libgnomeprintui2.2-common libgnomevfs2-bin
241 libgtksourceview-common libpt-1.10.10-plugins-alsa
242 libpt-1.10.10-plugins-v4l libservlet2.4-java libxalan2-java
243 libxerces2-java openoffice.org-writer2latex openssl-blacklist p7zip
244 python-4suite-xml python-eggtrayicon python-gtkhtml2
245 python-gtkmozembed svgalibg1 xserver-xephyr zip
246 &lt;/p&gt;&lt;/blockquote&gt;
247
248 &lt;p&gt;Installed using apt-get, removed with aptitude&lt;/p&gt;
249
250 &lt;blockquote&gt;&lt;p&gt;
251 bluez-utils dhcdbd djvulibre-desktop epiphany-gecko
252 gnome-app-install gnome-mount gnome-vfs-obexftp gnome-volume-manager
253 libao2 libavahi-compat-libdnssd1 libavahi-core5 libbind9-50
254 libbluetooth2 libcamel1.2-11 libcdio7 libcucul0 libcurl3
255 libdirectfb-1.0-0 libdvdread3 libedata-cal1.2-6 libedataserver1.2-9
256 libeel2-2.20 libepc-1.0-1 libepc-ui-1.0-1 libexchange-storage1.2-3
257 libfaad0 libgd2-noxpm libgda3-3 libgda3-common libggz2 libggzcore9
258 libggzmod4 libgksu1.2-0 libgksuui1.0-1 libgmyth0 libgnome-desktop-2
259 libgnome-pilot2 libgnomecups1.0-1 libgnomeprint2.2-0
260 libgnomeprintui2.2-0 libgpod3 libgraphviz4 libgtkhtml2-0
261 libgtksourceview1.0-0 libgucharmap6 libhesiod0 libicu38 libisccc50
262 libisccfg50 libiw29 libkpathsea4 libltdl3 liblwres50 libmagick++10
263 libmagick10 libmalaga7 libmtp7 libmysqlclient15off libnautilus-burn4
264 libneon27 libnm-glib0 libnm-util0 libopal-2.2 libosp5
265 libparted1.8-10 libpisock9 libpisync1 libpoppler-glib3 libpoppler3
266 libpt-1.10.10 libraw1394-8 libsensors3 libsmbios2 libsoup2.2-8
267 libssh2-1 libsuitesparse-3.1.0 libswfdec-0.6-90 libtalloc1
268 libtotem-plparser10 libtrackerclient0 libvoikko1 libxalan2-java-gcj
269 libxerces2-java-gcj libxklavier12 libxtrap6 libxxf86misc1 libzephyr3
270 mysql-common swfdec-gnome totem-gstreamer wodim
271 &lt;/p&gt;&lt;/blockquote&gt;
272
273 &lt;p&gt;Installed using aptitude, missing with apt-get&lt;/p&gt;
274
275 &lt;blockquote&gt;&lt;p&gt;
276 gnome gnome-desktop-environment hamster-applet python-gnomeapplet
277 python-gnomekeyring python-wnck rhythmbox-plugins xorg
278 xserver-xorg-input-all xserver-xorg-input-evdev
279 xserver-xorg-input-kbd xserver-xorg-input-mouse
280 xserver-xorg-input-synaptics xserver-xorg-video-all
281 xserver-xorg-video-apm xserver-xorg-video-ark xserver-xorg-video-ati
282 xserver-xorg-video-chips xserver-xorg-video-cirrus
283 xserver-xorg-video-dummy xserver-xorg-video-fbdev
284 xserver-xorg-video-glint xserver-xorg-video-i128
285 xserver-xorg-video-i740 xserver-xorg-video-mach64
286 xserver-xorg-video-mga xserver-xorg-video-neomagic
287 xserver-xorg-video-nouveau xserver-xorg-video-nv
288 xserver-xorg-video-r128 xserver-xorg-video-radeon
289 xserver-xorg-video-radeonhd xserver-xorg-video-rendition
290 xserver-xorg-video-s3 xserver-xorg-video-s3virge
291 xserver-xorg-video-savage xserver-xorg-video-siliconmotion
292 xserver-xorg-video-sis xserver-xorg-video-sisusb
293 xserver-xorg-video-tdfx xserver-xorg-video-tga
294 xserver-xorg-video-trident xserver-xorg-video-tseng
295 xserver-xorg-video-vesa xserver-xorg-video-vmware
296 xserver-xorg-video-voodoo
297 &lt;/p&gt;&lt;/blockquote&gt;
298
299 &lt;p&gt;Installed using aptitude, removed with apt-get&lt;/p&gt;
300
301 &lt;blockquote&gt;&lt;p&gt;
302 deskbar-applet xserver-xorg xserver-xorg-core
303 xserver-xorg-input-wacom xserver-xorg-video-intel
304 xserver-xorg-video-openchrome
305 &lt;/p&gt;&lt;/blockquote&gt;
306
307 &lt;p&gt;I was told on IRC that the xorg-xserver package was
308 &lt;a href=&quot;http://git.debian.org/?p=pkg-xorg/xserver/xorg-server.git;a=commit;h=9c8080d06c457932d3bfec021c69ac000aa60120&quot;&gt;changed
309 in git&lt;/a&gt; today to try to get apt-get to not remove xorg completely.
310 No idea when it hits Squeeze, but when it does I hope it will reduce
311 the difference somewhat.
312 </description>
313 </item>
314
315 <item>
316 <title>MS Word krøller det til for politiet?</title>
317 <link>http://people.skolelinux.org/pere/blog/MS_Word_kr_ller_det_til_for_politiet_.html</link>
318 <guid isPermaLink="true">http://people.skolelinux.org/pere/blog/MS_Word_kr_ller_det_til_for_politiet_.html</guid>
319 <pubDate>Thu, 8 Jul 2010 14:00:00 +0200</pubDate>
320 <description>&lt;p&gt;De siste dagene har Aftenposten
321 &lt;a href=&quot;http://www.aftenposten.no/nyheter/iriks/article3718597.ece&quot;&gt;fortalt&lt;/a&gt;
322 &lt;a href=&quot;http://www.aftenposten.no/nyheter/iriks/article3724249.ece&quot;&gt;hvordan&lt;/a&gt;
323 politet har brukt skriveverktøy som ikke håndterer arabisk tekst og
324 tekst som skal skrives fra høyre mot venstre når de har laget
325 løpeseddel for å be om informasjon fra publikum. Resultatet har vært
326 en uleselig arabisk-bit på løpeseddelen. Feilen har oppstått når
327 teksten har blitt &quot;kopiert inn i programvare som ikke har støtte for
328 språk som skrives fra høyre mot venstre&quot;, og jeg er ganske sikker på
329 at det er snakk om Microsoft Office i dette tilfellet. Er det slik at
330 MS Office i norsk språkdrakt ikke har støtte for tekst som skal
331 skrives fra høyre mot venstre? Jeg tror alle utgaver av
332 OpenOffice.org har slik støtte, og det er jo ikke veldig vanskelig å
333 la slik støtte finnes i alle utgaver av et program hvis støtten først
334 er utviklet. Aftenpostens melding får meg til å undre om problemet
335 ville vært unngått hvis politiet brukte OpenOffice.org i stedet for MS
336 Office.&lt;/p&gt;
337
338 &lt;p&gt;Mon tro om det er flere eksempler på at MS Office har ødelagt for
339 offentlig myndighet?&lt;/p&gt;
340 </description>
341 </item>
342
343 <item>
344 <title>jXplorer, a very nice LDAP GUI</title>
345 <link>http://people.skolelinux.org/pere/blog/jXplorer__a_very_nice_LDAP_GUI.html</link>
346 <guid isPermaLink="true">http://people.skolelinux.org/pere/blog/jXplorer__a_very_nice_LDAP_GUI.html</guid>
347 <pubDate>Fri, 9 Jul 2010 12:55:00 +0200</pubDate>
348 <description>&lt;p&gt;Since
349 &lt;a href=&quot;http://people.skolelinux.org/pere/blog/LUMA__a_very_nice_LDAP_GUI.html&quot;&gt;my
350 last post&lt;/a&gt; about available LDAP tools in Debian, I was told about a
351 LDAP GUI that is even better than luma. The java application
352 &lt;a href=&quot;http://jxplorer.org/&quot;&gt;jXplorer&lt;/a&gt; is claimed to be capable of
353 moving LDAP objects and subtrees using drag-and-drop, and can
354 authenticate using Kerberos. I have only tested the Kerberos
355 authentication, but do not have a LDAP setup allowing me to rewrite
356 LDAP with my test user yet. It is
357 &lt;a href=&quot;http://packages.qa.debian.org/j/jxplorer.html&quot;&gt;available in
358 Debian&lt;/a&gt; testing and unstable at the moment. The only problem I
359 have with it is how it handle errors. If something go wrong, its
360 non-intuitive behaviour require me to go through some query work list
361 and remove the failing query. Nothing big, but very annoying.&lt;/p&gt;
362 </description>
363 </item>
364
365 <item>
366 <title>Idea for storing LTSP configuration in LDAP</title>
367 <link>http://people.skolelinux.org/pere/blog/Idea_for_storing_LTSP_configuration_in_LDAP.html</link>
368 <guid isPermaLink="true">http://people.skolelinux.org/pere/blog/Idea_for_storing_LTSP_configuration_in_LDAP.html</guid>
369 <pubDate>Sun, 11 Jul 2010 22:00:00 +0200</pubDate>
370 <description>&lt;p&gt;Vagrant mentioned on IRC today that ltsp_config now support
371 sourcing files from /usr/share/ltsp/ltsp_config.d/ on the thin
372 clients, and that this can be used to fetch configuration from LDAP if
373 Debian Edu choose to store configuration there.&lt;/p&gt;
374
375 &lt;p&gt;Armed with this information, I got inspired and wrote a test module
376 to get configuration from LDAP. The idea is to look up the MAC
377 address of the client in LDAP, and look for attributes on the form
378 ltspconfigsetting=value, and use this to export SETTING=value to the
379 LTSP clients.&lt;/p&gt;
380
381 &lt;p&gt;The goal is to be able to store the LTSP configuration attributes
382 in a &quot;computer&quot; LDAP object used by both DNS and DHCP, and thus
383 allowing us to store all information about a computer in one place.&lt;/p&gt;
384
385 &lt;p&gt;This is a untested draft implementation, and I welcome feedback on
386 this approach. A real LDAP schema for the ltspClientAux objectclass
387 need to be written. Comments, suggestions, etc?&lt;/p&gt;
388
389 &lt;blockquote&gt;&lt;pre&gt;
390 # Store in /opt/ltsp/$arch/usr/share/ltsp/ltsp_config.d/ldap-config
391 #
392 # Fetch LTSP client settings from LDAP based on MAC address
393 #
394 # Uses ethernet address as stored in the dhcpHost objectclass using
395 # the dhcpHWAddress attribute or ethernet address stored in the
396 # ieee802Device objectclass with the macAddress attribute.
397 #
398 # This module is written to be schema agnostic, and only depend on the
399 # existence of attribute names.
400 #
401 # The LTSP configuration variables are saved directly using a
402 # ltspConfig prefix and uppercasing the rest of the attribute name.
403 # To set the SERVER variable, set the ltspConfigServer attribute.
404 #
405 # Some LDAP schema should be created with all the relevant
406 # configuration settings. Something like this should work:
407 #
408 # objectclass ( 1.1.2.2 NAME &#39;ltspClientAux&#39;
409 # SUP top
410 # AUXILIARY
411 # MAY ( ltspConfigServer $ ltsConfigSound $ ... )
412
413 LDAPSERVER=$(debian-edu-ldapserver)
414 if [ &quot;$LDAPSERVER&quot; ] ; then
415 LDAPBASE=$(debian-edu-ldapserver -b)
416 for MAC in $(LANG=C ifconfig |grep -i hwaddr| awk &#39;{print $5}&#39;|sort -u) ; do
417 filter=&quot;(|(dhcpHWAddress=ethernet $MAC)(macAddress=$MAC))&quot;
418 ldapsearch -h &quot;$LDAPSERVER&quot; -b &quot;$LDAPBASE&quot; -v -x &quot;$filter&quot; | \
419 grep &#39;^ltspConfig&#39; | while read attr value ; do
420 # Remove prefix and convert to upper case
421 attr=$(echo $attr | sed &#39;s/^ltspConfig//i&#39; | tr a-z A-Z)
422 # bass value on to clients
423 eval &quot;$attr=$value; export $attr&quot;
424 done
425 done
426 fi
427 &lt;/pre&gt;&lt;/blockquote&gt;
428
429 &lt;p&gt;I&#39;m not sure this shell construction will work, because I suspect
430 the while block might end up in a subshell causing the variables set
431 there to not show up in ltsp-config, but if that is the case I am sure
432 the code can be restructured to make sure the variables are passed on.
433 I expect that can be solved with some testing. :)&lt;/p&gt;
434
435 &lt;p&gt;If you want to help out with implementing this for Debian Edu,
436 please contact us on debian-edu@lists.debian.org.&lt;/p&gt;
437
438 &lt;p&gt;Update 2010-07-17: I am aware of another effort to store LTSP
439 configuration in LDAP that was created around year 2000 by
440 &lt;a href=&quot;http://www.pcxperience.com/thinclient/documentation/ldap.html&quot;&gt;PC
441 Xperience, Inc., 2000&lt;/a&gt;. I found its
442 &lt;a href=&quot;http://people.redhat.com/alikins/ltsp/ldap/&quot;&gt;files&lt;/a&gt; on a
443 personal home page over at redhat.com.&lt;/p&gt;
444 </description>
445 </item>
446
447 <item>
448 <title>Combining PowerDNS and ISC DHCP LDAP objects</title>
449 <link>http://people.skolelinux.org/pere/blog/Combining_PowerDNS_and_ISC_DHCP_LDAP_objects.html</link>
450 <guid isPermaLink="true">http://people.skolelinux.org/pere/blog/Combining_PowerDNS_and_ISC_DHCP_LDAP_objects.html</guid>
451 <pubDate>Wed, 14 Jul 2010 23:45:00 +0200</pubDate>
452 <description>&lt;p&gt;For a while now, I have wanted to find a way to change the DNS and
453 DHCP services in Debian Edu to use the same LDAP objects for a given
454 computer, to avoid the possibility of having a inconsistent state for
455 a computer in LDAP (as in DHCP but no DNS entry or the other way
456 around) and make it easier to add computers to LDAP.&lt;/p&gt;
457
458 &lt;p&gt;I&#39;ve looked at how powerdns and dhcpd is using LDAP, and using this
459 information finally found a solution that seem to work.&lt;/p&gt;
460
461 &lt;p&gt;The old setup required three LDAP objects for a given computer.
462 One forward DNS entry, one reverse DNS entry and one DHCP entry. If
463 we switch powerdns to use its strict LDAP method (ldap-method=strict
464 in pdns-debian-edu.conf), the forward and reverse DNS entries are
465 merged into one while making it impossible to transfer the reverse map
466 to a slave DNS server.&lt;/p&gt;
467
468 &lt;p&gt;If we also replace the object class used to get the DNS related
469 attributes to one allowing these attributes to be combined with the
470 dhcphost object class, we can merge the DNS and DHCP entries into one.
471 I&#39;ve written such object class in the dnsdomainaux.schema file (need
472 proper OIDs, but that is a minor issue), and tested the setup. It
473 seem to work.&lt;/p&gt;
474
475 &lt;p&gt;With this test setup in place, we can get away with one LDAP object
476 for both DNS and DHCP, and even the LTSP configuration I suggested in
477 an earlier email. The combined LDAP object will look something like
478 this:&lt;/p&gt;
479
480 &lt;blockquote&gt;&lt;pre&gt;
481 dn: cn=hostname,cn=group1,cn=THINCLIENTS,cn=DHCP Config,dc=skole,dc=skolelinux,dc=no
482 cn: hostname
483 objectClass: dhcphost
484 objectclass: domainrelatedobject
485 objectclass: dnsdomainaux
486 associateddomain: hostname.intern
487 arecord: 10.11.12.13
488 dhcphwaddress: ethernet 00:00:00:00:00:00
489 dhcpstatements: fixed-address hostname
490 ldapconfigsound: Y
491 &lt;/pre&gt;&lt;/blockquote&gt;
492
493 &lt;p&gt;The DNS server uses the associateddomain and arecord entries, while
494 the DHCP server uses the dhcphwaddress and dhcpstatements entries
495 before asking DNS to resolve the fixed-adddress. LTSP will use
496 dhcphwaddress or associateddomain and the ldapconfig* attributes.&lt;/p&gt;
497
498 &lt;p&gt;I am not yet sure if I can get the DHCP server to look for its
499 dhcphost in a different location, to allow us to put the objects
500 outside the &quot;DHCP Config&quot; subtree, but hope to figure out a way to do
501 that. If I can&#39;t figure out a way to do that, we can still get rid of
502 the hosts subtree and move all its content into the DHCP Config tree
503 (which probably should be renamed to be more related to the new
504 content. I suspect cn=dnsdhcp,ou=services or something like that
505 might be a good place to put it.&lt;/p&gt;
506
507 &lt;p&gt;If you want to help out with implementing this for Debian Edu,
508 please contact us on debian-edu@lists.debian.org.&lt;/p&gt;
509 </description>
510 </item>
511
512 <item>
513 <title>What are they searching for - PowerDNS and ISC DHCP in LDAP</title>
514 <link>http://people.skolelinux.org/pere/blog/What_are_they_searching_for___PowerDNS_and_ISC_DHCP_in_LDAP.html</link>
515 <guid isPermaLink="true">http://people.skolelinux.org/pere/blog/What_are_they_searching_for___PowerDNS_and_ISC_DHCP_in_LDAP.html</guid>
516 <pubDate>Sat, 17 Jul 2010 21:00:00 +0200</pubDate>
517 <description>&lt;p&gt;This is a
518 &lt;a href=&quot;http://people.skolelinux.org/pere/blog/Time_for_new__LDAP_schemas_replacing_RFC_2307_.html&quot;&gt;followup&lt;/a&gt;
519 on my
520 &lt;a href=&quot;http://people.skolelinux.org/pere/blog/Idea_for_a_change_to_LDAP_schemas_allowing_DNS_and_DHCP_info_to_be_combined_into_one_object.html&quot;&gt;previous
521 work&lt;/a&gt; on
522 &lt;a href=&quot;http://people.skolelinux.org/pere/blog/Combining_PowerDNS_and_ISC_DHCP_LDAP_objects.html&quot;&gt;merging
523 all&lt;/a&gt; the computer related LDAP objects in Debian Edu.&lt;/p&gt;
524
525 &lt;p&gt;As a step to try to see if it possible to merge the DNS and DHCP
526 LDAP objects, I have had a look at how the packages pdns-backend-ldap
527 and dhcp3-server-ldap in Debian use the LDAP server. The two
528 implementations are quite different in how they use LDAP.&lt;/p&gt;
529
530 To get this information, I started slapd with debugging enabled and
531 dumped the debug output to a file to get the LDAP searches performed
532 on a Debian Edu main-server. Here is a summary.
533
534 &lt;p&gt;&lt;strong&gt;powerdns&lt;/strong&gt;&lt;/p&gt;
535
536 &lt;a href=&quot;http://www.linuxnetworks.de/doc/index.php/PowerDNS_LDAP_Backend&quot;&gt;Clues
537 on how to&lt;/a&gt; set up PowerDNS to use a LDAP backend is available on
538 the web.
539
540 &lt;p&gt;PowerDNS have two modes of operation using LDAP as its backend.
541 One &quot;strict&quot; mode where the forward and reverse DNS lookups are done
542 using the same LDAP objects, and a &quot;tree&quot; mode where the forward and
543 reverse entries are in two different subtrees in LDAP with a structure
544 based on the DNS names, as in tjener.intern and
545 2.2.0.10.in-addr.arpa.&lt;/p&gt;
546
547 &lt;p&gt;In tree mode, the server is set up to use a LDAP subtree as its
548 base, and uses a &quot;base&quot; scoped search for the DNS name by adding
549 &quot;dc=tjener,dc=intern,&quot; to the base with a filter for
550 &quot;(associateddomain=tjener.intern)&quot; for the forward entry and
551 &quot;dc=2,dc=2,dc=0,dc=10,dc=in-addr,dc=arpa,&quot; with a filter for
552 &quot;(associateddomain=2.2.0.10.in-addr.arpa)&quot; for the reverse entry. For
553 forward entries, it is looking for attributes named dnsttl, arecord,
554 nsrecord, cnamerecord, soarecord, ptrrecord, hinforecord, mxrecord,
555 txtrecord, rprecord, afsdbrecord, keyrecord, aaaarecord, locrecord,
556 srvrecord, naptrrecord, kxrecord, certrecord, dsrecord, sshfprecord,
557 ipseckeyrecord, rrsigrecord, nsecrecord, dnskeyrecord, dhcidrecord,
558 spfrecord and modifytimestamp. For reverse entries it is looking for
559 the attributes dnsttl, arecord, nsrecord, cnamerecord, soarecord,
560 ptrrecord, hinforecord, mxrecord, txtrecord, rprecord, aaaarecord,
561 locrecord, srvrecord, naptrrecord and modifytimestamp. The equivalent
562 ldapsearch commands could look like this:&lt;/p&gt;
563
564 &lt;blockquote&gt;&lt;pre&gt;
565 ldapsearch -h ldap \
566 -b dc=tjener,dc=intern,ou=hosts,dc=skole,dc=skolelinux,dc=no \
567 -s base -x &#39;(associateddomain=tjener.intern)&#39; dNSTTL aRecord nSRecord \
568 cNAMERecord sOARecord pTRRecord hInfoRecord mXRecord tXTRecord \
569 rPRecord aFSDBRecord KeyRecord aAAARecord lOCRecord sRVRecord \
570 nAPTRRecord kXRecord certRecord dSRecord sSHFPRecord iPSecKeyRecord \
571 rRSIGRecord nSECRecord dNSKeyRecord dHCIDRecord sPFRecord modifyTimestamp
572
573 ldapsearch -h ldap \
574 -b dc=2,dc=2,dc=0,dc=10,dc=in-addr,dc=arpa,ou=hosts,dc=skole,dc=skolelinux,dc=no \
575 -s base -x &#39;(associateddomain=2.2.0.10.in-addr.arpa)&#39;
576 dnsttl, arecord, nsrecord, cnamerecord soarecord ptrrecord \
577 hinforecord mxrecord txtrecord rprecord aaaarecord locrecord \
578 srvrecord naptrrecord modifytimestamp
579 &lt;/pre&gt;&lt;/blockquote&gt;
580
581 &lt;p&gt;In Debian Edu/Lenny, the PowerDNS tree mode is used with
582 ou=hosts,dc=skole,dc=skolelinux,dc=no as the base, and these are two
583 example LDAP objects used there. In addition to these objects, the
584 parent objects all th way up to ou=hosts,dc=skole,dc=skolelinux,dc=no
585 also exist.&lt;/p&gt;
586
587 &lt;blockquote&gt;&lt;pre&gt;
588 dn: dc=tjener,dc=intern,ou=hosts,dc=skole,dc=skolelinux,dc=no
589 objectclass: top
590 objectclass: dnsdomain
591 objectclass: domainrelatedobject
592 dc: tjener
593 arecord: 10.0.2.2
594 associateddomain: tjener.intern
595
596 dn: dc=2,dc=2,dc=0,dc=10,dc=in-addr,dc=arpa,ou=hosts,dc=skole,dc=skolelinux,dc=no
597 objectclass: top
598 objectclass: dnsdomain2
599 objectclass: domainrelatedobject
600 dc: 2
601 ptrrecord: tjener.intern
602 associateddomain: 2.2.0.10.in-addr.arpa
603 &lt;/pre&gt;&lt;/blockquote&gt;
604
605 &lt;p&gt;In strict mode, the server behaves differently. When looking for
606 forward DNS entries, it is doing a &quot;subtree&quot; scoped search with the
607 same base as in the tree mode for a object with filter
608 &quot;(associateddomain=tjener.intern)&quot; and requests the attributes dnsttl,
609 arecord, nsrecord, cnamerecord, soarecord, ptrrecord, hinforecord,
610 mxrecord, txtrecord, rprecord, aaaarecord, locrecord, srvrecord,
611 naptrrecord and modifytimestamp. For reverse entires it also do a
612 subtree scoped search but this time the filter is &quot;(arecord=10.0.2.2)&quot;
613 and the requested attributes are associateddomain, dnsttl and
614 modifytimestamp. In short, in strict mode the objects with ptrrecord
615 go away, and the arecord attribute in the forward object is used
616 instead.&lt;/p&gt;
617
618 &lt;p&gt;The forward and reverse searches can be simulated using ldapsearch
619 like this:&lt;/p&gt;
620
621 &lt;blockquote&gt;&lt;pre&gt;
622 ldapsearch -h ldap -b ou=hosts,dc=skole,dc=skolelinux,dc=no -s sub -x \
623 &#39;(associateddomain=tjener.intern)&#39; dNSTTL aRecord nSRecord \
624 cNAMERecord sOARecord pTRRecord hInfoRecord mXRecord tXTRecord \
625 rPRecord aFSDBRecord KeyRecord aAAARecord lOCRecord sRVRecord \
626 nAPTRRecord kXRecord certRecord dSRecord sSHFPRecord iPSecKeyRecord \
627 rRSIGRecord nSECRecord dNSKeyRecord dHCIDRecord sPFRecord modifyTimestamp
628
629 ldapsearch -h ldap -b ou=hosts,dc=skole,dc=skolelinux,dc=no -s sub -x \
630 &#39;(arecord=10.0.2.2)&#39; associateddomain dnsttl modifytimestamp
631 &lt;/pre&gt;&lt;/blockquote&gt;
632
633 &lt;p&gt;In addition to the forward and reverse searches , there is also a
634 search for SOA records, which behave similar to the forward and
635 reverse lookups.&lt;/p&gt;
636
637 &lt;p&gt;A thing to note with the PowerDNS behaviour is that it do not
638 specify any objectclass names, and instead look for the attributes it
639 need to generate a DNS reply. This make it able to work with any
640 objectclass that provide the needed attributes.&lt;/p&gt;
641
642 &lt;p&gt;The attributes are normally provided in the cosine (RFC 1274) and
643 dnsdomain2 schemas. The latter is used for reverse entries like
644 ptrrecord and recent DNS additions like aaaarecord and srvrecord.&lt;/p&gt;
645
646 &lt;p&gt;In Debian Edu, we have created DNS objects using the object classes
647 dcobject (for dc), dnsdomain or dnsdomain2 (structural, for the DNS
648 attributes) and domainrelatedobject (for associatedDomain). The use
649 of structural object classes make it impossible to combine these
650 classes with the object classes used by DHCP.&lt;/p&gt;
651
652 &lt;p&gt;There are other schemas that could be used too, for example the
653 dnszone structural object class used by Gosa and bind-sdb for the DNS
654 attributes combined with the domainrelatedobject object class, but in
655 this case some unused attributes would have to be included as well
656 (zonename and relativedomainname).&lt;/p&gt;
657
658 &lt;p&gt;My proposal for Debian Edu would be to switch PowerDNS to strict
659 mode and not use any of the existing objectclasses (dnsdomain,
660 dnsdomain2 and dnszone) when one want to combine the DNS information
661 with DHCP information, and instead create a auxiliary object class
662 defined something like this (using the attributes defined for
663 dnsdomain and dnsdomain2 or dnszone):&lt;/p&gt;
664
665 &lt;blockquote&gt;&lt;pre&gt;
666 objectclass ( some-oid NAME &#39;dnsDomainAux&#39;
667 SUP top
668 AUXILIARY
669 MAY ( ARecord $ MDRecord $ MXRecord $ NSRecord $ SOARecord $ CNAMERecord $
670 DNSTTL $ DNSClass $ PTRRecord $ HINFORecord $ MINFORecord $
671 TXTRecord $ SIGRecord $ KEYRecord $ AAAARecord $ LOCRecord $
672 NXTRecord $ SRVRecord $ NAPTRRecord $ KXRecord $ CERTRecord $
673 A6Record $ DNAMERecord
674 ))
675 &lt;/pre&gt;&lt;/blockquote&gt;
676
677 &lt;p&gt;This will allow any object to become a DNS entry when combined with
678 the domainrelatedobject object class, and allow any entity to include
679 all the attributes PowerDNS wants. I&#39;ve sent an email to the PowerDNS
680 developers asking for their view on this schema and if they are
681 interested in providing such schema with PowerDNS, and I hope my
682 message will be accepted into their mailing list soon.&lt;/p&gt;
683
684 &lt;p&gt;&lt;strong&gt;ISC dhcp&lt;/strong&gt;&lt;/p&gt;
685
686 &lt;p&gt;The DHCP server searches for specific objectclass and requests all
687 the object attributes, and then uses the attributes it want. This
688 make it harder to figure out exactly what attributes are used, but
689 thanks to the working example in Debian Edu I can at least get an idea
690 what is needed without having to read the source code.&lt;/p&gt;
691
692 &lt;p&gt;In the DHCP server configuration, the LDAP base to use and the
693 search filter to use to locate the correct dhcpServer entity is
694 stored. These are the relevant entries from
695 /etc/dhcp3/dhcpd.conf:&lt;/p&gt;
696
697 &lt;blockquote&gt;&lt;pre&gt;
698 ldap-base-dn &quot;dc=skole,dc=skolelinux,dc=no&quot;;
699 ldap-dhcp-server-cn &quot;dhcp&quot;;
700 &lt;/pre&gt;&lt;/blockquote&gt;
701
702 &lt;p&gt;The DHCP server uses this information to nest all the DHCP
703 configuration it need. The cn &quot;dhcp&quot; is located using the given LDAP
704 base and the filter &quot;(&amp;(objectClass=dhcpServer)(cn=dhcp))&quot;. The
705 search result is this entry:&lt;/p&gt;
706
707 &lt;blockquote&gt;&lt;pre&gt;
708 dn: cn=dhcp,dc=skole,dc=skolelinux,dc=no
709 cn: dhcp
710 objectClass: top
711 objectClass: dhcpServer
712 dhcpServiceDN: cn=DHCP Config,dc=skole,dc=skolelinux,dc=no
713 &lt;/pre&gt;&lt;/blockquote&gt;
714
715 &lt;p&gt;The content of the dhcpServiceDN attribute is next used to locate the
716 subtree with DHCP configuration. The DHCP configuration subtree base
717 is located using a base scope search with base &quot;cn=DHCP
718 Config,dc=skole,dc=skolelinux,dc=no&quot; and filter
719 &quot;(&amp;(objectClass=dhcpService)(|(dhcpPrimaryDN=cn=dhcp,dc=skole,dc=skolelinux,dc=no)(dhcpSecondaryDN=cn=dhcp,dc=skole,dc=skolelinux,dc=no)))&quot;.
720 The search result is this entry:&lt;/p&gt;
721
722 &lt;blockquote&gt;&lt;pre&gt;
723 dn: cn=DHCP Config,dc=skole,dc=skolelinux,dc=no
724 cn: DHCP Config
725 objectClass: top
726 objectClass: dhcpService
727 objectClass: dhcpOptions
728 dhcpPrimaryDN: cn=dhcp, dc=skole,dc=skolelinux,dc=no
729 dhcpStatements: ddns-update-style none
730 dhcpStatements: authoritative
731 dhcpOption: smtp-server code 69 = array of ip-address
732 dhcpOption: www-server code 72 = array of ip-address
733 dhcpOption: wpad-url code 252 = text
734 &lt;/pre&gt;&lt;/blockquote&gt;
735
736 &lt;p&gt;Next, the entire subtree is processed, one level at the time. When
737 all the DHCP configuration is loaded, it is ready to receive requests.
738 The subtree in Debian Edu contain objects with object classes
739 top/dhcpService/dhcpOptions, top/dhcpSharedNetwork/dhcpOptions,
740 top/dhcpSubnet, top/dhcpGroup and top/dhcpHost. These provide options
741 and information about netmasks, dynamic range etc. Leaving out the
742 details here because it is not relevant for the focus of my
743 investigation, which is to see if it is possible to merge dns and dhcp
744 related computer objects.&lt;/p&gt;
745
746 &lt;p&gt;When a DHCP request come in, LDAP is searched for the MAC address
747 of the client (00:00:00:00:00:00 in this example), using a subtree
748 scoped search with &quot;cn=DHCP Config,dc=skole,dc=skolelinux,dc=no&quot; as
749 the base and &quot;(&amp;(objectClass=dhcpHost)(dhcpHWAddress=ethernet
750 00:00:00:00:00:00))&quot; as the filter. This is what a host object look
751 like:&lt;/p&gt;
752
753 &lt;blockquote&gt;&lt;pre&gt;
754 dn: cn=hostname,cn=group1,cn=THINCLIENTS,cn=DHCP Config,dc=skole,dc=skolelinux,dc=no
755 cn: hostname
756 objectClass: top
757 objectClass: dhcpHost
758 dhcpHWAddress: ethernet 00:00:00:00:00:00
759 dhcpStatements: fixed-address hostname
760 &lt;/pre&gt;&lt;/blockquote&gt;
761
762 &lt;p&gt;There is less flexiblity in the way LDAP searches are done here.
763 The object classes need to have fixed names, and the configuration
764 need to be stored in a fairly specific LDAP structure. On the
765 positive side, the invidiual dhcpHost entires can be anywhere without
766 the DN pointed to by the dhcpServer entries. The latter should make
767 it possible to group all host entries in a subtree next to the
768 configuration entries, and this subtree can also be shared with the
769 DNS server if the schema proposed above is combined with the dhcpHost
770 structural object class.
771
772 &lt;p&gt;&lt;strong&gt;Conclusion&lt;/strong&gt;&lt;/p&gt;
773
774 &lt;p&gt;The PowerDNS implementation seem to be very flexible when it come
775 to which LDAP schemas to use. While its &quot;tree&quot; mode is rigid when it
776 come to the the LDAP structure, the &quot;strict&quot; mode is very flexible,
777 allowing DNS objects to be stored anywhere under the base cn specified
778 in the configuration.&lt;/p&gt;
779
780 &lt;p&gt;The DHCP implementation on the other hand is very inflexible, both
781 regarding which LDAP schemas to use and which LDAP structure to use.
782 I guess one could implement ones own schema, as long as the
783 objectclasses and attributes have the names used, but this do not
784 really help when the DHCP subtree need to have a fairly fixed
785 structure.&lt;/p&gt;
786
787 &lt;p&gt;Based on the observed behaviour, I suspect a LDAP structure like
788 this might work for Debian Edu:&lt;/p&gt;
789
790 &lt;blockquote&gt;&lt;pre&gt;
791 ou=services
792 cn=machine-info (dhcpService) - dhcpServiceDN points here
793 cn=dhcp (dhcpServer)
794 cn=dhcp-internal (dhcpSharedNetwork/dhcpOptions)
795 cn=10.0.2.0 (dhcpSubnet)
796 cn=group1 (dhcpGroup/dhcpOptions)
797 cn=dhcp-thinclients (dhcpSharedNetwork/dhcpOptions)
798 cn=192.168.0.0 (dhcpSubnet)
799 cn=group1 (dhcpGroup/dhcpOptions)
800 ou=machines - PowerDNS base points here
801 cn=hostname (dhcpHost/domainrelatedobject/dnsDomainAux)
802 &lt;/pre&gt;&lt;/blockquote&gt;
803
804 &lt;P&gt;This is not tested yet. If the DHCP server require the dhcpHost
805 entries to be in the dhcpGroup subtrees, the entries can be stored
806 there instead of a common machines subtree, and the PowerDNS base
807 would have to be moved one level up to the machine-info subtree.&lt;/p&gt;
808
809 &lt;p&gt;The combined object under the machines subtree would look something
810 like this:&lt;/p&gt;
811
812 &lt;blockquote&gt;&lt;pre&gt;
813 dn: dc=hostname,ou=machines,cn=machine-info,dc=skole,dc=skolelinux,dc=no
814 dc: hostname
815 objectClass: top
816 objectClass: dhcpHost
817 objectclass: domainrelatedobject
818 objectclass: dnsDomainAux
819 associateddomain: hostname.intern
820 arecord: 10.11.12.13
821 dhcpHWAddress: ethernet 00:00:00:00:00:00
822 dhcpStatements: fixed-address hostname.intern
823 &lt;/pre&gt;&lt;/blockquote&gt;
824
825 &lt;/p&gt;One could even add the LTSP configuration associated with a given
826 machine, as long as the required attributes are available in a
827 auxiliary object class.&lt;/p&gt;
828 </description>
829 </item>
830
831 <item>
832 <title>OpenStreetmap one step closer to having routing on its front page</title>
833 <link>http://people.skolelinux.org/pere/blog/OpenStreetmap_one_step_closer_to_having_routing_on_its_front_page.html</link>
834 <guid isPermaLink="true">http://people.skolelinux.org/pere/blog/OpenStreetmap_one_step_closer_to_having_routing_on_its_front_page.html</guid>
835 <pubDate>Sun, 18 Jul 2010 16:45:00 +0200</pubDate>
836 <description>&lt;p&gt;Thanks to
837 &lt;a href=&quot;http://feedproxy.google.com/~r/Opengeodata/~3/wUTCzDZk3lc/project-of-the-week-which-way-home&quot;&gt;todays
838 opengeodata blog entry&lt;/a&gt;, I just discovered that the
839 OpenStreetmap.org site have gotten
840 &lt;a href=&quot;http://nroets.dev.openstreetmap.org/demo/index.html?layers=B000FTFTT&quot;&gt;support
841 for calculating routes&lt;/a&gt;. The support is still experimental and
842 only available from the development server, until more experience is
843 gathered on the user interface and any scalability issues.&lt;/p&gt;
844
845 &lt;p&gt;Earlier, the routing I knew about using the OpenStreetmap.org data
846 was provided by &lt;a href=&quot;http://maps.cloudmade.com/&quot;&gt;Cloudmade&lt;/a&gt;,
847 but having it on the main page is required to make everyone aware of
848 the issue. I&#39;ve had people reject Openstreetmap.org as a viable
849 alternative for them because the front page lacked routing support,
850 and I hope their needs will be catered for when routing show up on the
851 www.openstreetmap.org front page.&lt;/p&gt;
852 </description>
853 </item>
854
855 <item>
856 <title>Digitale restriksjonsmekanismer fikk meg til å slutte å kjøpe musikk</title>
857 <link>http://people.skolelinux.org/pere/blog/Digitale_restriksjonsmekanismer_fikk_meg_til___slutte___kj_pe_musikk.html</link>
858 <guid isPermaLink="true">http://people.skolelinux.org/pere/blog/Digitale_restriksjonsmekanismer_fikk_meg_til___slutte___kj_pe_musikk.html</guid>
859 <pubDate>Thu, 22 Jul 2010 23:50:00 +0200</pubDate>
860 <description>&lt;p&gt;For mange år siden slutte jeg å kjøpe musikk-CDer. Årsaken var at
861 musikkbransjen var godt i gang med å selge platene sine med DRM som
862 gjorde at jeg ikke fikk spilt av musikken jeg kjøpte på utstyret jeg
863 hadde tilgjengelig, dvs. min datamaskin. Det var umulig å se på en
864 plate om den var ødelagt eller ikke, og jeg hadde jo allerede en
865 anseelig samling med plater, så jeg bestemme meg for å slutte å gi
866 penger til en bransje som åpenbart ikke respekterte meg.&lt;/p&gt;
867
868 &lt;p&gt;Jeg har mange titalls dager med musikk på CD i dag. Det meste er
869 lagt i et stort arkiv som kan spilles av fra husets datamaskiner (har
870 ikke rukket rippe alt). Jeg ser dermed ikke behovet for å skaffe mer
871 musikk. De fleste av mine favoritter er i hus, og jeg er dermed godt
872 fornøyd.&lt;/p&gt;
873
874 &lt;p&gt;Hvis musikkbransjen ønsker mine penger, så må de demonstrere at de
875 setter pris på meg som kunde, og ikke skremme meg bort med DRM og
876 antydninger om at kundene er kriminelle.&lt;/p&gt;
877
878 &lt;p&gt;Filmbransjen er like ille, men mens musikk gjerne varer lenge, er
879 filmer mer ferskvare. Har dermed ikke helt sluttet å kjøpe filmer, men
880 holder meg til DVD-filmer som kan spilles av på mine Linuxbokser.
881 Kommer neppe til å ta i bruk Blueray, og ei heller de nye DRM-greiene
882 «Ultraviolet» som be annonsert her om dagen.&lt;/p&gt;
883 </description>
884 </item>
885
886 <item>
887 <title>One step closer to single signon in Debian Edu</title>
888 <link>http://people.skolelinux.org/pere/blog/One_step_closer_to_single_signon_in_Debian_Edu.html</link>
889 <guid isPermaLink="true">http://people.skolelinux.org/pere/blog/One_step_closer_to_single_signon_in_Debian_Edu.html</guid>
890 <pubDate>Sun, 25 Jul 2010 10:00:00 +0200</pubDate>
891 <description>&lt;p&gt;The last few months me and the other Debian Edu developers have
892 been working hard to get the Debian/Squeeze based version of Debian
893 Edu/Skolelinux into shape. This future version will use Kerberos for
894 authentication, and services are slowly migrated to single signon,
895 getting rid of password questions one at the time.&lt;/p&gt;
896
897 &lt;p&gt;It will also feature a roaming workstation profile with local home
898 directory, for laptops that are only some times on the Skolelinux
899 network, and for this profile a shortcut is created in Gnome and KDE
900 to gain access to the users home directory on the file server. This
901 shortcut uses SMB at the moment, and yesterday I had time to test if
902 SMB mounting had started working in KDE after we added the cifs-utils
903 package. I was pleasantly surprised how well it worked.&lt;/p&gt;
904
905 &lt;p&gt;Thanks to the recent changes to our samba configuration to get it
906 to use Kerberos for authentication, there were no question about user
907 password when mounting the SMB volume. A simple click on the shortcut
908 in the KDE menu, and a window with the home directory popped
909 up. :)&lt;/p&gt;
910
911 &lt;p&gt;One step closer to a single signon solution out of the box in
912 Debian Edu. We already had PAM, LDAP, IMAP and SMTP in place, and now
913 also Samba. Next step is Cups and hopefully also NFS.&lt;/p&gt;
914
915 &lt;p&gt;We had planned a alpha0 release of Debian Edu for today, but thanks
916 to the autobuilder administrators for some architectures being slow to
917 sign packages, we are still missing the fixed LTSP package we need for
918 the release. It was uploaded three days ago with urgency=high, and if
919 it had entered testing yesterday we would have been able to test it in
920 time for a alpha0 release today. As the binaries for ia64 and powerpc
921 still not uploaded to the Debian archive, we need to delay the alpha
922 release another day.&lt;/p&gt;
923
924 &lt;p&gt;If you want to help out with implementing Kerberos for Debian Edu,
925 please contact us on debian-edu@lists.debian.org.&lt;/p&gt;
926 </description>
927 </item>
928
929 <item>
930 <title>First Debian Edu test release (alpha0) based on Squeeze is released</title>
931 <link>http://people.skolelinux.org/pere/blog/First_Debian_Edu_test_release__alpha0__based_on_Squeeze_is_released.html</link>
932 <guid isPermaLink="true">http://people.skolelinux.org/pere/blog/First_Debian_Edu_test_release__alpha0__based_on_Squeeze_is_released.html</guid>
933 <pubDate>Tue, 27 Jul 2010 17:45:00 +0200</pubDate>
934 <description>&lt;p&gt;I just posted this announcement culminating several months of work
935 with the next Debian Edu release. Not nearly done, but one major step
936 completed.&lt;/p&gt;
937
938 &lt;blockquote&gt;
939 &lt;p&gt;This is the first test release based on Squeeze. The focus of this
940 release is to test the user application selection. To have a look,
941 install the standalone profile and let the developers know if the set
942 of installed packages i.e. applications should be modified. If some
943 user application is missing, or if there are some applications that no
944 longer make sense to be included in Debian Edu, please let us know.
945 Also, if a useful application is missing the translation for your
946 language of choice, please let us know too.&lt;/p&gt;
947
948 &lt;p&gt;In addition, feedback and help to polish the desktop (menus,
949 artwork, starters, etc.) is appreciated. We would like to ship a nice
950 and handy KDE4 desktop targeted for schools out of the box.&lt;/p&gt;
951
952 &lt;p&gt;The other profiles should be installable, but there is a lot more
953 work left to be done before they are ready, so do not expect to
954 much.&lt;/p&gt;
955
956 &lt;p&gt;Changes compared to the lenny based version&lt;/p&gt;
957
958 &lt;ul&gt;
959 &lt;li&gt;Everything from Debian Squeeze
960 &lt;ul&gt;
961 &lt;li&gt;Desktop environment KDE 4.4 =&gt; the new KDE desktop in
962 combination with some new artwork
963 &lt;li&gt;Web browser Iceweasel 3.5
964 &lt;li&gt;OpenOffice.org 3.2
965 &lt;li&gt;Educational toolbox GCompris 9.3
966 &lt;li&gt;Music creator Rosegarden 10.04.2
967 &lt;li&gt;Image editor Gimp 2.6.10
968 &lt;li&gt;Virtual universe Celestia 1.6.0
969 &lt;li&gt;Virtual stargazer Stellarium 0.10.4
970 &lt;li&gt;3D modeler Blender 2.49.2 (new application)
971 &lt;li&gt;Video editor Kdenlive 0.7.7 (new application)
972 &lt;/ul&gt;&lt;/li&gt;
973 &lt;li&gt;Now using Kerberos for password checking (migration not finished).
974 Enabled for:
975 &lt;ul&gt;
976 &lt;li&gt;PAM
977 &lt;li&gt;LDAP
978 &lt;li&gt;IMAP
979 &lt;li&gt;SMTP (sender verification)
980 &lt;/ul&gt;
981 &lt;/li&gt;
982 &lt;li&gt;New experimental roaming workstation profile for laptops.&lt;/li&gt;
983 &lt;li&gt;Show welcome page to users when they first log in. The URL is
984 fetched from LDAP.&lt;/li&gt;
985 &lt;li&gt;New LXDE desktop option, in addition to KDE (default) and Gnome.&lt;/li&gt;
986 &lt;li&gt;General cleanup (not finished)&lt;/li&gt;
987 &lt;/ul&gt;
988 &lt;p&gt;The following features are not working as they should&lt;/p&gt;
989
990 &lt;ul&gt;
991 &lt;li&gt;No web based administration tool for creating users and groups. The
992 scripts ldap-createuser-krb and ldap-add-user-to-group can be used
993 for testing.&lt;/li&gt;
994 &lt;li&gt;DVD installs are missing debian-installer images for the PXE boot,
995 and do not set up the PXE menu on eth0 because of this. LTSP
996 clients should still boot from eth1 on thin client servers.&lt;/li&gt;
997 &lt;li&gt;The restructured KDE menu is not implemented.&lt;/li&gt;
998 &lt;li&gt;The LDAP server setup need to be reviewed for security.&lt;/li&gt;
999 &lt;li&gt;The LDAP directory structure need to be reworked.&lt;/li&gt;
1000 &lt;li&gt;Different sets of packages are installed when using the DVD and the
1001 netinst CD. More packages are installed using the netinst CD.&lt;/li&gt;
1002 &lt;li&gt;The jackd package fail to install. This is believed to be caused by
1003 some ongoing transition, and hopefully should be solved soon. The
1004 jackd1 package can be installed manually for those that need it.&lt;/li&gt;
1005 &lt;li&gt;Some packages lack translations. See
1006 http://wiki.debian.org/DebianEdu/Status/Squeeze for updated status,
1007 and help out with translations.&lt;/li&gt;
1008 &lt;/ul&gt;
1009
1010 &lt;p&gt;To download this multiarch netinstall release you can use&lt;/p&gt;
1011
1012 &lt;ul&gt;
1013 &lt;li&gt;&lt;a href=&quot;ftp://ftp.skolelinux.org/skolelinux-cd/squeeze-alpha/debian-edu-6.0.0+edua0-CD.iso&quot;&gt;ftp://ftp.skolelinux.org/skolelinux-cd/squeeze-alpha/debian-edu-6.0.0+edua0-CD.iso&lt;/a&gt;&lt;/li&gt;
1014 &lt;li&gt;&lt;a href=&quot;http://ftp.skolelinux.org/skolelinux-cd/squeeze-alpha/debian-edu-6.0.0+edua0-CD.iso&quot;&gt;http://ftp.skolelinux.org/skolelinux-cd/squeeze-alpha/debian-edu-6.0.0+edua0-CD.iso&lt;/a&gt;&lt;/li&gt;
1015 &lt;li&gt;rsync -avzP ftp.skolelinux.org::skolelinux-cd/squeeze-alpha/debian-edu-6.0.0+edua0-CD.iso&lt;/li&gt;
1016 &lt;/ul&gt;
1017 &lt;p&gt;To download this multiarch dvd release you can use&lt;/p&gt;
1018
1019 &lt;ul&gt;
1020 &lt;li&gt;&lt;a href=&quot;ftp://ftp.skolelinux.org/skolelinux-cd/squeeze-alpha/debian-edu-6.0.0+edua0-DVD.iso&quot;&gt;ftp://ftp.skolelinux.org/skolelinux-cd/squeeze-alpha/debian-edu-6.0.0+edua0-DVD.iso&lt;/a&gt;&lt;/li&gt;
1021 &lt;li&gt;&lt;a href=&quot;http://ftp.skolelinux.org/skolelinux-cd/squeeze-alpha/debian-edu-6.0.0+edua0-DVD.iso&quot;&gt;http://ftp.skolelinux.org/skolelinux-cd/squeeze-alpha/debian-edu-6.0.0+edua0-DVD.iso&lt;/a&gt;&lt;/li&gt;
1022 &lt;li&gt;rsync -avzP ftp.skolelinux.org::skolelinux-cd/squeeze-alpha/debian-edu-6.0.0+edua0-DVD.iso&lt;/li&gt;
1023 &lt;/ul&gt;
1024
1025 &lt;p&gt;There is no source DVD available yet. It will be prepared when we
1026 get closer to the final release.&lt;/p&gt;
1027
1028 &lt;p&gt;The MD5SUM of these images are&lt;/p&gt;
1029
1030 &lt;ul&gt;
1031 &lt;li&gt;3dbf45d59f42a53518b6e3c9ec3b5eb6 debian-edu-6.0.0+edua0-CD.iso&lt;/li&gt;
1032 &lt;li&gt;22f2cbfce281d1c6e478be452638675d debian-edu-6.0.0+edua0-DVD.iso&lt;/li&gt;
1033 &lt;/ul&gt;
1034
1035 &lt;p&gt;The SHA1SUM of these images are&lt;/p&gt;
1036 &lt;ul&gt;
1037 &lt;li&gt;c53d1b69b40cf37cd27aefaf33f6f6a3821bedf0 debian-edu-6.0.0+edua0-CD.iso&lt;/li&gt;
1038 &lt;li&gt;2ec29d7db676d59d32197b05c277ffe16348376c debian-edu-6.0.0+edua0-DVD.iso&lt;/li&gt;
1039 &lt;/ul&gt;
1040 &lt;p&gt;How to report bugs:
1041 http://wiki.debian.org/DebianEdu/HowTo/ReportBugsInBugzilla&lt;/p&gt;
1042
1043 &lt;p&gt;Please direct replies to debian-edu@lists.debian.org&lt;/p&gt;
1044 &lt;/blockquote&gt;
1045 </description>
1046 </item>
1047
1048 <item>
1049 <title>Circular package dependencies harms apt recovery</title>
1050 <link>http://people.skolelinux.org/pere/blog/Circular_package_dependencies_harms_apt_recovery.html</link>
1051 <guid isPermaLink="true">http://people.skolelinux.org/pere/blog/Circular_package_dependencies_harms_apt_recovery.html</guid>
1052 <pubDate>Tue, 27 Jul 2010 23:50:00 +0200</pubDate>
1053 <description>&lt;p&gt;I discovered this while doing
1054 &lt;a href=&quot;http://people.skolelinux.org/pere/blog/Automatic_upgrade_testing_from_Lenny_to_Squeeze.html&quot;&gt;automated
1055 testing of upgrades from Debian Lenny to Squeeze&lt;/a&gt;. A few packages
1056 in Debian still got circular dependencies, and it is often claimed
1057 that apt and aptitude should be able to handle this just fine, but
1058 some times these dependency loops causes apt to fail.&lt;/p&gt;
1059
1060 &lt;p&gt;An example is from todays
1061 &lt;a href=&quot;http://people.skolelinux.org/~pere/debian-upgrade-testing//test-20100727-lenny-squeeze-kde-aptitude.txt&quot;&gt;upgrade
1062 of KDE using aptitude&lt;/a&gt;. In it, a bug in kdebase-workspace-data
1063 causes perl-modules to fail to upgrade. The cause is simple. If a
1064 package fail to unpack, then only part of packages with the circular
1065 dependency might end up being unpacked when unpacking aborts, and the
1066 ones already unpacked will fail to configure in the recovery phase
1067 because its dependencies are unavailable.&lt;/p&gt;
1068
1069 &lt;p&gt;In this log, the problem manifest itself with this error:&lt;/p&gt;
1070
1071 &lt;blockquote&gt;&lt;pre&gt;
1072 dpkg: dependency problems prevent configuration of perl-modules:
1073 perl-modules depends on perl (&gt;= 5.10.1-1); however:
1074 Version of perl on system is 5.10.0-19lenny2.
1075 dpkg: error processing perl-modules (--configure):
1076 dependency problems - leaving unconfigured
1077 &lt;/pre&gt;&lt;/blockquote&gt;
1078
1079 &lt;p&gt;The perl/perl-modules circular dependency is already
1080 &lt;a href=&quot;http://bugs.debian.org/527917&quot;&gt;reported as a bug&lt;/a&gt;, and will
1081 hopefully be solved as soon as possible, but it is not the only one,
1082 and each one of these loops in the dependency tree can cause similar
1083 failures. Of course, they only occur when there are bugs in other
1084 packages causing the unpacking to fail, but it is rather nasty when
1085 the failure of one package causes the problem to become worse because
1086 of dependency loops.&lt;/p&gt;
1087
1088 &lt;p&gt;Thanks to
1089 &lt;a href=&quot;http://lists.debian.org/debian-devel/2010/06/msg00116.html&quot;&gt;the
1090 tireless effort by Bill Allombert&lt;/a&gt;, the number of circular
1091 dependencies
1092 &lt;a href=&quot;http://debian.semistable.com/debgraph.out.html&quot;&gt;left in Debian
1093 is dropping&lt;/a&gt;, and perhaps it will reach zero one day. :)&lt;/p&gt;
1094
1095 &lt;p&gt;Todays testing also exposed a bug in
1096 &lt;a href=&quot;http://bugs.debian.org/590605&quot;&gt;update-notifier&lt;/a&gt; and
1097 &lt;a href=&quot;http://bugs.debian.org/590604&quot;&gt;different behaviour&lt;/a&gt; between
1098 apt-get and aptitude, the latter possibly caused by some circular
1099 dependency. Reported both to BTS to try to get someone to look at
1100 it.&lt;/p&gt;
1101 </description>
1102 </item>
1103
1104 </channel>
1105 </rss>