<?xml version="1.0" encoding="utf-8"?>
<feed xml:lang="en" xmlns="http://www.w3.org/2005/Atom"><title>Recent changes to bugs</title><link href="https://sourceforge.net/p/cpu/bugs/" rel="alternate"/><link href="https://sourceforge.net/p/cpu/bugs/feed.atom" rel="self"/><id>https://sourceforge.net/p/cpu/bugs/</id><updated>2009-09-03T18:45:02Z</updated><subtitle>Recent changes to bugs</subtitle><entry><title>CPU doesn't increment through large GID numbers properly</title><link href="https://sourceforge.net/p/cpu/bugs/38/" rel="alternate"/><published>2009-09-03T18:45:02Z</published><updated>2009-09-03T18:45:02Z</updated><author><name>Anonymous</name><uri>https://sourceforge.net/u/userid-None/</uri></author><id>https://sourceforge.netf36728025c2821f70f45900de57fb36dce0a28c6</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;We have a weird GID setup....&lt;br /&gt;
Some system groups in the 0-1000 range. &lt;br /&gt;
Users primary groups in the 60-80000 range. &lt;br /&gt;
One off groups in the 9000000-910000.&lt;br /&gt;
I have no idea why it was setup like this.&lt;br /&gt;
&lt;/p&gt;
&lt;p&gt;Anyhow, if  MAX_GIDNUMBER and MIN_GIDNUMBER are in any of those ranges, but the last one, with RANDOM set to false when doing a "cpu groupadd foo -w" it will find the first free GID and do its thing like it should.&lt;br /&gt;
&lt;/p&gt;
&lt;p&gt;However, if my cpu.conf looks like this &lt;br /&gt;
MAX_GIDNUMBER = 9100000&lt;br /&gt;
MIN_GIDNUMBER = 9000040&lt;br /&gt;
it will create groups with GIDs in the 10000 range.&lt;br /&gt;
&lt;/p&gt;
&lt;p&gt;Any thoughts on why it doesn't handle numbers this large?&lt;/p&gt;
&lt;p&gt;Using 1.4.3 on Cent OS 5.3 btw.&lt;br /&gt;
Let me know if you need to know anything else.&lt;br /&gt;
&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>Segfault on Solaris</title><link href="https://sourceforge.net/p/cpu/bugs/37/" rel="alternate"/><published>2007-03-29T08:05:37Z</published><updated>2007-03-29T08:05:37Z</updated><author><name>Eugene E Panfyorov</name><uri>https://sourceforge.net/u/userid-581969/</uri></author><id>https://sourceforge.net2bfa4f1b48860d06a5eaddc0b8c1861784bd9409</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;&amp;gt; uname -a&lt;br /&gt;
SunOS 5.10 Generic_118844-26 i86pc i386 i86pc&lt;/p&gt;
&lt;p&gt;&amp;gt; cc -V&lt;br /&gt;
cc: Sun C 5.7 Patch 117837-04 2005/05/11&lt;/p&gt;
&lt;p&gt;&amp;gt; CC=cc ./configure --with-ldap=/opt/csw&lt;br /&gt;
...&lt;br /&gt;
config successful&lt;/p&gt;
&lt;p&gt;&amp;gt; make; make install&lt;br /&gt;
...&lt;br /&gt;
success&lt;/p&gt;
&lt;p&gt;&amp;gt; cpu cat&lt;br /&gt;
Segmentation Fault (core dumped)&lt;/p&gt;
&lt;p&gt;The config.log is attached to this message&lt;/p&gt;
&lt;p&gt;(gcc produced the same result)&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>compilation is failed on FreeBSD</title><link href="https://sourceforge.net/p/cpu/bugs/36/" rel="alternate"/><published>2007-03-27T13:31:26Z</published><updated>2007-03-27T13:31:26Z</updated><author><name>Anonymous</name><uri>https://sourceforge.net/u/userid-None/</uri></author><id>https://sourceforge.net88539343252dccd7f1158043733496a8fb3a51a9</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;FreeBSD  6.2&lt;/p&gt;
&lt;p&gt;gcc version 3.4.6 [FreeBSD] 20060305&lt;/p&gt;
&lt;p&gt;&amp;gt; cd cpu-1.4.3&lt;br /&gt;
&amp;gt; ./configure --with-ldap=/usr/local&lt;br /&gt;
&amp;gt; make &lt;br /&gt;
ld.c: In function `getlGid':&lt;br /&gt;
ld.c:806: error: subscripted value is neither array nor pointer&lt;/p&gt;
&lt;p&gt;and the following patch makes all things fine:&lt;/p&gt;
&lt;p&gt;--- ld.c.orig   Tue Mar 27 17:14:05 2007&lt;br /&gt;
+++ ld.c        Tue Mar 27 15:15:41 2007&lt;br /&gt;
@@ -771,6 +771,7 @@&lt;br /&gt;
NULL&lt;br /&gt;
};&lt;br /&gt;
char *gid = NULL;&lt;br /&gt;
+  char **gid_ = NULL;&lt;/p&gt;
&lt;p&gt;cn = cfg_get_str ("LDAP", "GROUP_CN_STRING");&lt;br /&gt;
if (cn == NULL)&lt;br /&gt;
@@ -803,9 +804,9 @@&lt;br /&gt;
return -10;&lt;br /&gt;
if (gid != NULL)&lt;br /&gt;
{&lt;br /&gt;
-      gid = ldap_get_values (ld, pos, gid)[0];&lt;br /&gt;
-      if (gid != NULL)&lt;br /&gt;
-       return atoi (gid);&lt;br /&gt;
+      gid_ = ldap_get_values (ld, pos, gid);&lt;br /&gt;
+      if (gid_ != NULL &amp;amp;&amp;amp; gid_[0] != NULL)&lt;br /&gt;
+       return atoi (gid_[0]);&lt;br /&gt;
else&lt;br /&gt;
return -10;&lt;br /&gt;
}&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>compilation fix for fedora 6 (maybe earlier)</title><link href="https://sourceforge.net/p/cpu/bugs/35/" rel="alternate"/><published>2006-12-18T16:27:24Z</published><updated>2006-12-18T16:27:24Z</updated><author><name>Devin Reade</name><uri>https://sourceforge.net/u/glynreade/</uri></author><id>https://sourceforge.net424edcf9619ebd2bb2d5b81b66c89ae8f9cbb4fb</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;This is a change that was necessary to get CPU to compile under fedora core 6.  I seem to recall that I also had to apply it to an earlier fedora release, but don't recall the details.&lt;/p&gt;
&lt;p&gt;In src/plugins/ldap/commandline.c around line 86 there is the following line:&lt;/p&gt;
&lt;p&gt;(int) passent-&amp;gt;sp_lstchg = passent-&amp;gt;sp_min = passent-&amp;gt;sp_max = -10;&lt;/p&gt;
&lt;p&gt;which results in "commandline.c:86: error: invalid lvalue in assignment"&lt;/p&gt;
&lt;p&gt;Changing it to be the following fixed the problem:&lt;/p&gt;
&lt;p&gt;passent-&amp;gt;sp_lstchg = -10;&lt;br /&gt;
passent-&amp;gt;sp_min = -10;&lt;br /&gt;
passent-&amp;gt;sp_max = -10;&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>error cpu useradd</title><link href="https://sourceforge.net/p/cpu/bugs/34/" rel="alternate"/><published>2006-10-14T18:23:31Z</published><updated>2006-10-14T18:23:31Z</updated><author><name>Anonymous</name><uri>https://sourceforge.net/u/userid-None/</uri></author><id>https://sourceforge.net7cfd935987bc3d8e40821a2311c8709a76360f0a</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;I get an error when i want to add a user &lt;br /&gt;
cpu useradd -g users toto&lt;/p&gt;
&lt;p&gt;CPU: ldapUserAdd: ldap_add_s: Internal (implementation&lt;br /&gt;
specific) error&lt;br /&gt;
The error number was 80, please add an appropriate&lt;br /&gt;
entry to ldap_errors.c.&lt;br /&gt;
If you are unable, please email&lt;br /&gt;
cpu-users@lists.sourceforge.net with the error number&lt;br /&gt;
and what operation was being performed at the time.&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>Compilation error ld.c line 806</title><link href="https://sourceforge.net/p/cpu/bugs/33/" rel="alternate"/><published>2006-10-14T18:20:33Z</published><updated>2006-10-14T18:20:33Z</updated><author><name>Anonymous</name><uri>https://sourceforge.net/u/userid-None/</uri></author><id>https://sourceforge.net9fe09bcd17f3b650f66900764d0b58dd5da1e6b1</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;cpu compilation failed on&lt;br /&gt;
gid = ldap_get_values (ld, pos, gid)[0];&lt;br /&gt;
line 806&lt;br /&gt;
it works like this &lt;br /&gt;
gid = ldap_get_values (ld, pos, gid);&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>Broken link on project home page</title><link href="https://sourceforge.net/p/cpu/bugs/32/" rel="alternate"/><published>2005-03-14T15:55:55Z</published><updated>2005-03-14T15:55:55Z</updated><author><name>fghdgh</name><uri>https://sourceforge.net/u/gcc111/</uri></author><id>https://sourceforge.net47825ccb9cc090c1f21adad4545c72e42269065c</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;The link to the Changelog &lt;br /&gt;
(http://www.dbaseiv.net/code/changelog.cpu) on the home &lt;br /&gt;
page (http://cpu.sourceforge.net/) no longer works, and takes &lt;br /&gt;
you to an advertising site. &lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>User added to initial group</title><link href="https://sourceforge.net/p/cpu/bugs/31/" rel="alternate"/><published>2005-02-21T17:00:13Z</published><updated>2005-02-21T17:00:13Z</updated><author><name>Carlos</name><uri>https://sourceforge.net/u/carpecon/</uri></author><id>https://sourceforge.net3357bfc9b7de1680bb838cd489d27d64ed9e1af9</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;With cpu 1.4.3 and the following configuration: &lt;br /&gt;
USERGROUPS = no &lt;br /&gt;
USERS_GID = 1000 &lt;br /&gt;
when doing: &lt;br /&gt;
cpu useradd testuser &lt;br /&gt;
testuser gets explicitly added to group 1000: &lt;br /&gt;
$ cpu cat passwd | grep testuser &lt;br /&gt;
testuser:x:100:1000::/home/testuser:/bin/bash &lt;br /&gt;
testgroup:x:1000:testuser &lt;/p&gt;
&lt;p&gt;I've modified CPU's source to try to avoid this behaviour and it &lt;br /&gt;
seems to work for me: &lt;br /&gt;
$ diff orig/cpu-1.4.3/src/plugins/ldap/user.c &lt;br /&gt;
cpu-1.4.3/src/plugins/ldap/user.c &lt;br /&gt;
425c425 &lt;br /&gt;
&amp;lt;       if (ldapres &amp;gt; 0) &lt;br /&gt;
--- &lt;br /&gt;
&amp;gt; /*    if (ldapres &amp;gt; 0) &lt;br /&gt;
428c428 &lt;br /&gt;
&amp;lt;       } &lt;br /&gt;
--- &lt;br /&gt;
&amp;gt;       }*/ &lt;/p&gt;
&lt;p&gt;I haven't sent this in as a patch because I'm not confident &lt;br /&gt;
enough with CPU source, but to me the problem seems to be &lt;br /&gt;
that ldapUserCheck() also adds the user to the group when &lt;br /&gt;
globalLdap-&amp;gt;gid is not NULL. AFAIK, this gid is the ascii name &lt;br /&gt;
of the user's initial group, to which he doesn't need to be &lt;br /&gt;
added. &lt;/p&gt;
&lt;p&gt;Best regards, &lt;br /&gt;
Carlos &lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>Unable to perform operations on users with $ char in the nam</title><link href="https://sourceforge.net/p/cpu/bugs/30/" rel="alternate"/><published>2005-01-02T11:51:51Z</published><updated>2005-01-02T11:51:51Z</updated><author><name>Fabian</name><uri>https://sourceforge.net/u/mr-meltdown/</uri></author><id>https://sourceforge.neteafc796e2bc2122b28f78e306a40ca4c709faa04</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;I tried using cpu to manage my LDAP with samba accounts&lt;br /&gt;
in it.  I know samba can do LDAP itself, but for some&lt;br /&gt;
historical legacy reasons my current setup just doesn't&lt;br /&gt;
use LDAP itself.  So I replaced in the machine add&lt;br /&gt;
script /usr/sbin/useradd with cpu useradd to let samba&lt;br /&gt;
add machine accounts to the LDAP server.&lt;/p&gt;
&lt;p&gt;This, however, does not work as samba adds an $ to the&lt;br /&gt;
end of machine account user names.  This way I cannot&lt;br /&gt;
remove such accounts using cpu as well.  If I remove&lt;br /&gt;
the $ in a manual test all works well.  As soon as&lt;br /&gt;
there is a $ in the login cpu acts as if it was not&lt;br /&gt;
given at all.  I think this is incorrect, as the&lt;br /&gt;
original useradd unix tool is able to add such user.&lt;/p&gt;
&lt;p&gt;example output:&lt;br /&gt;
# /usr/sbin/cpu useradd -d /dev/null -g machines -c&lt;br /&gt;
'Machine Account' -s /bin/false test_machine&lt;br /&gt;
User test_machine successfully added!&lt;br /&gt;
# /usr/sbin/cpu useradd -d /dev/null -g machines -c&lt;br /&gt;
'Machine Account' -s /bin/false test_machine$&lt;br /&gt;
usage: cpu user{add,del,mod} [options] login&lt;br /&gt;
usage: cpu group{add,del,mod} [options] group&lt;br /&gt;
usage: cpu cat&lt;/p&gt;
&lt;p&gt;usage: cpu useradd [options] login&lt;br /&gt;
-c comment --gecos=COMMENT       : Gecos&lt;br /&gt;
Information&lt;br /&gt;
-d home_dir --directory=home_dir : Users home&lt;br /&gt;
directory&lt;br /&gt;
-g initial_group --gid=initial   : The group id&lt;br /&gt;
or name of the user's&lt;br /&gt;
initial&lt;br /&gt;
login group&lt;br /&gt;
-G group,[...] --sgroup=group,[] : A list of&lt;br /&gt;
supplementary groups&lt;br /&gt;
-k[skeleton_dir] --skel[=dir]    : The skeleton&lt;br /&gt;
directory&lt;br /&gt;
-m --makehome                    : The user's&lt;br /&gt;
home directory will be&lt;br /&gt;
created if&lt;br /&gt;
it does not exist&lt;br /&gt;
-p[passwd] --password[=password] : The&lt;br /&gt;
unencrypted password&lt;br /&gt;
-s shell --shell=shell           : The name of&lt;br /&gt;
the user's login shell&lt;br /&gt;
-u uid --uid=uid                 : The&lt;br /&gt;
numerical value of the user's&lt;br /&gt;
ID.&lt;br /&gt;
-X script --exec=script          : Post add script&lt;/p&gt;
&lt;p&gt;some (relevant?) version information:&lt;br /&gt;
# cpu --version&lt;br /&gt;
CPU 1.4.3&lt;br /&gt;
Written by Blake Matheny&lt;br /&gt;
Copyright 2001, 2002, 2003&lt;/p&gt;
&lt;p&gt;2.6.9 linux kernel Gentoo&lt;/p&gt;&lt;/div&gt;</summary></entry><entry><title>bad behaviour when doing cpu usermod</title><link href="https://sourceforge.net/p/cpu/bugs/29/" rel="alternate"/><published>2004-10-23T12:19:34Z</published><updated>2004-10-23T12:19:34Z</updated><author><name>Sven Linstaedt</name><uri>https://sourceforge.net/u/tzwoenn/</uri></author><id>https://sourceforge.netf0736a11780d0fc90c330798b0c1f63dcdd4092d</id><summary type="html">&lt;div class="markdown_content"&gt;&lt;p&gt;i am using CPU 1.4.3 on debian sarge and i dont know&lt;br /&gt;
wether this is intentionally or a bug:&lt;br /&gt;
when i do a "cpu usermod" all of my secondary group&lt;br /&gt;
memberships are resetted.&lt;br /&gt;
when specifiing the -G option with an empty list this&lt;br /&gt;
would be normal behaviour in my oppinion. but even if i&lt;br /&gt;
dont specify this option (e.g. i just want want to&lt;br /&gt;
change users passwd, gecos or something else), the&lt;br /&gt;
secondary group memberships are resetted to be empty.&lt;br /&gt;
as a workaround i had to look up the group memberships&lt;br /&gt;
before i modify a user in any way (doing a "cpu cat").&lt;br /&gt;
then i could do a "cpu usermod" giving the -G option&lt;br /&gt;
with the desired memberships and any other modifiing&lt;br /&gt;
option.&lt;/p&gt;
&lt;p&gt;i cant imagine that behaviour is wanted :)&lt;/p&gt;&lt;/div&gt;</summary></entry></feed>