matt's questions

I have a pool of ESXi-5.5.0 servers that I have configured a single vApp on. This vApp is configured in transient mode to configure IP addresses from an IP pool I specified. I assigned a single virtual machine to this vApp and am trying to get IP addresses, to no avail. I don't really understand how vApp works, the hosts are configured to use DHCP but always get apipa (169.254.0.0/16) addresses in the case of windows hosts, when then dhcp client times out. The same happens on linux guests. I confirmed that the pool is on the correct virtual switch, and the hosts are on the same network. When I manually assign addresses, I can ping the gateway and access network resources.

What am I doing wrong here?

My issue I think boils down to OpenLDAP. I want my users to be able to change their passwords using the Pidgin chat client. At the moment, I have to set their passwords, and it's a pain. OpenLDAP works fine, my config is below.

When users try to change password, they get a 403 forbidden from Pidgin. Nothing happens in the OpenLDAP log when they get this error, which makes me think that the request never gets there. This is the log from Pidgin:

(14:08:01) jabber: Sending (ssl) ([email protected]/b660f5a6): <iq type='set' id='purple711c2157' to='xx.com'><query xmlns='jabber:iq:register'><username>test.account</username><password>22</password></query></iq>

(14:08:01) jabber: Recv (ssl)(302): <iq type="error" id="purple711c2157" from="xx.com" to="[email protected]/b660f5a6"><query xmlns="jabber:iq:register"><username>test.account</username><password>22</password></query><error code="403" type="auth"><forbidden xmlns="urn:ietf:params:xml:ns:xmpp-stanzas"/></error></iq>

This is my slapd.conf:

# This is the main slapd configuration file. See slapd.conf(5) for more
# info on the configuration options.

# Features to permit
#allow bind_v2

# Schema and objectClass definitions
include         /etc/ldap/schema/core.schema
include         /etc/ldap/schema/cosine.schema
include         /etc/ldap/schema/nis.schema
include         /etc/ldap/schema/inetorgperson.schema

# Where the pid file is put. The init.d script
# will not stop the server if you change this.
pidfile         /var/run/slapd/slapd.pid

#     List of arguments that were passed to the server
argsfile        /var/run/slapd/slapd.args

# Read slapd.conf(5) for possible values
loglevel       0

# Where the dynamically loaded modules are stored
modulepath  /usr/lib/ldap
moduleload  back_hdb

# The maximum number of entries that is returned for a search operation
sizelimit 500

# The tool-threads parameter sets the actual amount of cpu's that is used
# for indexing.
tool-threads 1

#######################################################################
# Specific Backend Directives for hdb:
# Backend specific directives apply to this backend until another
# 'backend' directive occurs
backend     hdb

#######################################################################
# Specific Directives for database #1, of type hdb:
# Database specific directives apply to this databasse until another
# 'database' directive occurs
database        hdb

# The base of your directory in database #1
suffix          "dc=xx,dc=com"

# rootdn directive for specifying a superuser on the database. This is needed
# for syncrepl.
# rootdn          "cn=admin,dc=vo,dc=srfarm,dc=net"

rootdn  "cn=admin,dc=xx,dc=com"
rootpw  xXxXxXxX

# Where the database file are physically stored for database #1
directory       "/var/lib/ldap"

# The dbconfig settings are used to generate a DB_CONFIG file the first
# time slapd starts.  They do NOT override existing an existing DB_CONFIG
# file.  You should therefore change these settings in DB_CONFIG directly
# or remove DB_CONFIG and restart slapd for changes to take effect.

# For the Debian package we use 2MB as default but be sure to update this
# value if you have plenty of RAM
dbconfig set_cachesize 0 2097152 0

# Number of objects that can be locked at the same time.
dbconfig set_lk_max_objects 1500
# Number of locks (both requested and granted)
dbconfig set_lk_max_locks 1500
# Number of lockers
dbconfig set_lk_max_lockers 1500

# Indexing options for database #1
index default eq,pres
index           objectClass eq
index       uid,cn,member

# Save the time that the entry gets modified, for database #1
lastmod         on

# Checkpoint the BerkeleyDB database periodically in case of system
# failure and to speed slapd shutdown.
checkpoint      512 30

# The userPassword by default can be changed
# by the entry owning it if they are authenticated.
# Others should not be able to see it, except the
# admin entry below
# These access lines apply to database #1 only
access to attrs=userPassword,shadowLastChange
        by dn="cn=admin,dc=xx,dc=com" write
        by anonymous auth
        by self write
        by * none

# Ensure read access to the base for things like
# supportedSASLMechanisms.  Without this you may
# have problems with SASL not knowing what
# mechanisms are available and the like.
# Note that this is covered by the 'access to *'
# ACL below too but if you change that as people
# are wont to do you'll still need this if you
# want SASL (and possible other things) to work 
# happily.
access to dn.base="" by * read

# The admin dn has full write access, everyone else
# can read everything.
access to *
    by self write
        by dn="cn=admin,dc=xx,dc=com" write
        by * read

Anyone have any ideas?

I installed ruby from source, and everything ruby related seems to work just fine. I can't install anything that depends on ruby using apt because I don't want 2 versions of ruby installed. After I install mcollective from dpkg using --ignore-depends, apt-get stops working entirely.

root@puppet:~# apt-get install ganglia-monitor
Reading package lists... Done
Building dependency tree       
Reading state information... Done
You might want to run 'apt-get -f install' to correct these:
The following packages have unmet dependencies:
 ganglia-monitor : Depends: libconfuse0 (>= 2.5) but it is not going to be installed
                   Depends: libganglia1 (= 3.1.7-1+b1) but it is not going to be installed
 mcollective : Depends: ruby (>= 1.8.1)
 mcollective-client : Depends: ruby (>= 1.8.1)
 mcollective-common : Depends: ruby (>= 1.8.1)
E: Unmet dependencies. Try 'apt-get -f install' with no packages (or specify a solution).

I don't want to remove mcollective, but when I try that, everything works again. Does anyone know a workaround to tell apt to stop looking for ruby? Thanks, matt.

We have an application that is running on a few (5 or so and will grow) boxes. The hardware is identical in all the machines, and ideally the software would be as well. I have been managing them by hand up until now, and don't want to anymore (static ip addresses, disabling all necessary services, installing required packages...) . Can anyone balance the pros and cons of the following options, or suggest something more intelligent?

1: Individually install centos on all the boxes and manage the configs with chef/cfengine/puppet. This would be good, as I have wanted an excuse to learn to use one of applications, but I don't know if this is actually the best solution.

2: Make one box perfect and image it. Serve the image over PXE and whenever I want to make modifications, I can just reboot the boxes from a new image. How do cluster guys normally handle things like having mac addresses in the /etc/sysconfig/network-scripts/ifcfg* files? We use infiniband as well, and it also refuses to start if the hwaddr is wrong. Can these be correctly generated at boot?

I'm leaning towards the PXE solution, but I think monitoring with munin or nagios will be a little more complicated with this. Anyone have experience with this type of problem?

All the servers have SSDs in them and are fast and powerful.

Thanks, matt.

We have a CentOS 5.3 x64 server that by default runs kernel version 2.6.18-164.11.1 and we are attempting to upgrade the box to 2.6.31.12 The drive is LVM +ext3, and the problem I'm having is when I upgrade the kernel and attempt to boot from it, no matter what version of the kernel I use, I get /dev/root not found towards the end of the boot process, and the kernel panics, and than reboots.

I'm installing the kernel exactly as it says in this doc. I've tried it "The centOS way " using make rpm and than installing that. I've updated my mkinitrd. The most interesting part of this problem is that it has been so frustrating that I decided to try and clean install centos on an identical machine without LVM, and the result is EXACTLY the same. After upgrading the kernel, I get /dev/root not found. Does anyone know how to fix this, or what information would be relevant to remedy it? I'm open to try anything at this point.

One more interesting thing about this problem is that in the new version of the kernel, during boot it complains that dm-mapper is started twice, than panics right after that. I've tried this with other kernel versions, and the result is the same. What am I missing here?

If you need any more files, please just ask.

 Linux cg 2.6.18-164.11.1.el5 #1 SMP Wed Jan 20 07:32:21 EST 2010 x86_64 x86_64 x86_64 GNU/Linux

  /dev/VolGroup00/LogVol00 /                       ext3    defaults        1 1
    LABEL=/boot             /boot                   ext3    defaults        1 2
    tmpfs                   /dev/shm                tmpfs   defaults        0 0
    devpts                  /dev/pts                devpts  gid=5,mode=620  0 0
    sysfs                   /sys                    sysfs   defaults        0 0
    proc                    /proc                   proc    defaults        0 0
    /dev/VolGroup00/LogVol01 swap                    swap    defaults        0 0

default=1
timeout=5
splashimage=(hd0,0)/grub/splash.xpm.gz
hiddenmenu
title CentOS (2.6.31.12-rt20) //NOT WORKING!!!!
        root (hd0,0)
        kernel /vmlinuz-2.6.31.12-rt20 ro root=/dev/VolGroup00/LogVol00 isolcpus=8,9,10,11,12,13,14,15 panic=10
        initrd /initrd-2.6.31.12-rt20.img
title CentOS (2.6.18-164.11.1.el5) //WORKING!!
        root (hd0,0)
        kernel /vmlinuz-2.6.18-164.11.1.el5 ro root=/dev/VolGroup00/LogVol00 isolcpus=8,9,10,11,12,13,14,15 panic=10
        initrd /initrd-2.6.18-164.11.1.el5.img

Can anyone recommend a good book or set of documents specifically about what OpenSolaris or Solaris is doing internally? I was reading this post and saw a comment that said

"On Solaris it was/is even more involved. The swap file is mirrored in a ram disk like tmpfs so the memory is always almost full - but it is apparently provable that this is the optimal strategy. – mgb Jun 25 at 19:40"

Can anyone verify that?

I then checked my OpenSolaris fileserver and saw this:

load averages:  0.05,  0.04,  0.04;               up 6+15:41:36        09:48:02
118 processes: 117 sleeping, 1 on cpu
CPU states: 99.4% idle,  0.1% user,  0.4% kernel,  0.0% iowait,  0.0% swap
Kernel: 1025 ctxsw, 6 trap, 1120 intr, 916 syscall
Memory: 16G phys mem, 3816M free mem, 8143M total swap, 8143M free swap

This server is doing nothing at the moment. Why is the swap completely empty? Where is this filesystem?

Basically, I want to learn more about how OpenSolaris is managing memory, and the intricacies of the operating system in general.

Thanks! matt.