hv02.ovh.rpmfusion.net
Contents
Overview
The hv02.ovh.rpmfusion.net host is a HyperVisor server to be used for RPM Fusion related services.
Server details :
- Hostname : hv02.ovh.rpmfusion.net aka ks313095.kimsufi.com
- Hardware : OVH Kemsirve i7-2T, i7 920 (4 x 2(HT) x 2.66GHz), 8GB DDR3, 2 x 1TB SATA (RAID-1 software)
- Network : 100Mbps switch port, 3TB monthly traffic (switch port limited to 10Mbps when exceeded)
- Main IPv4 address : 188.165.226.50/24 gw 188.165.226.254
- Additional IPv4 failover addresses :
- 46.105.55.71 associated with 02:00:00:28:b9:9d used for se01.ovh.rpmfusion.net
- 46.105.55.72 associated with 02:00:00:73:6a:fd used for se02.ovh.rpmfusion.net
- 46.105.55.73 associated with 02:00:00:aa:ff:cf used for se03.ovh.rpmfusion.net
- IPv6 network : 2001:41d0:2:ad32::/64
Management : OVH Manager with the rf30282-ovh RPM Forge account. This includes network traffic reports, tech support, remote reboot, DNS reverse mappings, etc.
Services
Detail of services running on the host and its virtual servers :
- hv02.ovh.rpmfusion.net :
- Web for static files and reverse proxying (nginx)
- DHCP for virtual server installations
- IPv4 NAT for *.lan servers
- se01.ovh.rpmfusion.net :
- Primary DNS
- SMTP, main MX
- Mailman
- se02.ovh.rpmfusion.net :
- MoinMoin website
- se03.ovh.rpmfusion.net :
- TBD
- puppet.ovh.rpmfusion.lan :
- Puppet master
- builder1.ovh.rpmfusion.lan :
- Plague builder for i386 and x86_64
- bugzilla.ovh.rpmfusion.lan :
- Bugzilla (behind the host's reverse proxy)
(Re)Installation
Installed with CentOS 5 64bit in English. Steps below are mostly raw notes, yet easily understandable.
Post-Install Manual Steps
vi /etc/sysconfig/network # HOSTNAME=hv02.ovh.rpmfusion.net # NETWORKING_IPV6=yes vi /etc/crontab # remove the last line for rtm vi /etc/fstab # remove /home and swap umount /home mdadm --manage --stop /dev/md2 vi /etc/mdadm.conf # remove md2 line swapoff -a fdisk /dev/sda # delete 3 & 2 # n p 2 : first 1306 last 120557 # n p 3 : first 120558 last enter # t 2 fd # t 3 fd fdisk /dev/sdb # same changes partprobe mdadm --create /dev/md2 --metadata=0.90 --level=1 --raid-devices=2 /dev/sda2 /dev/sdb2 mdadm --create /dev/md3 --metadata=0.90 --level=1 --raid-devices=2 /dev/sda3 /dev/sdb3 mkswap -f /dev/md3 vi /etc/fstab # add : /dev/md3 none swap defaults 0 0 swapon -a free
Upgrade to RHEL6 somehow. This requires only updating to an intermediate xz-aware rpm-4.7.1-6.el5 build and its requirements, then pointing to a RHEL6 repository.
Main /srv Mount Point & Repositories
pvcreate /dev/md2 vgcreate vg0 /dev/md2 lvcreate -L512G -nsrv vg0 vi /etc/mdadm.conf # add md2 and md3 with UUID # get with mdadm --detail /dev/md2 mkfs.ext4 -j -m 1 -L srv /dev/vg0/srv tune2fs -c 0 -i 0 /dev/vg0/srv vi /etc/fstab # add : /dev/vg0/srv /srv ext4 noatime,nodiratime,commit=120 0 0 # !!! Add ,commit=120 for all ext* filesystems! mount /srv mkdir -p /srv/repo/redhat/el{5,6}/x86_64 ln -s el5 /srv/repo/redhat/el5Server ln -s el6 /srv/repo/redhat/el6Server # sync iso (loop mount, create os), updates (and optional for el6) from somewhere
Use the local repository :
rm -rf /var/cache/yum/* mv /etc/yum.repos.d/* /tmp/ vi /etc/yum.repos.d/el.repo [os] name=Red Hat Enterprise Linux $releasever - $basearch - OS baseurl=file:///srv/repo/redhat/el$releasever/$basearch/os/ enabled=1 gpgcheck=1 gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-redhat-release [updates] name=Red Hat Enterprise Linux $releasever - $basearch - Updates baseurl=file:///srv/repo/redhat/el$releasever/$basearch/updates/ enabled=1 gpgcheck=1 gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-redhat-release
Tweaks
yum remove bind bind-chroot vi /etc/resolv.conf # remove 127.0.0.1 yum remove 'man-pages-??' specspo # Some obtained by "yum list extras" yum remove 'abrt*' esound Deployment_Guide-en-US xen-libs qpixman htmlview compat-libcap yum remove libvolume_id pkinit-nss qffmpeg-libs etherboot-zroms-kvm e4fsprogs-libs # The EL5 updates are more recent than the EL6 packages yum downgrade dmidecode less pciutils chkconfig iscsi off chkconfig iscsid off chkconfig --level 345 ntpd on service ntpd start
KVM Hypervisor
yum update yum install kvm python-virtinst libvirt.x86_64 virt-top rm -f /etc/libvirt/qemu/networks/autostart/default.xml service libvirtd start chkconfig --level 345 ksm on chkconfig --level 345 ksmtuned on
Network
- /etc/sysconfig/network-scripts/ifcfg-eth0 :
DEVICE=eth0 BRIDGE=br0 ONBOOT=yes
- /etc/sysconfig/network-scripts/ifcfg-br0 :
# Main public Ethernet DEVICE=br0 TYPE=Bridge BOOTPROTO=static IPADDR=188.165.226.50 NETMASK=255.255.255.0 GATEWAY=188.165.226.254 ONBOOT=yes # IPv6 IPV6INIT=yes IPV6ADDR=2001:41d0:2:ad32::cafe/56 IPV6_DEFAULTGW=2001:41d0:2:adff:ff:ff:ff:ff
- /etc/sysconfig/network-scripts/ifcfg-br1 :
# Private Ethernet (IPv4 only) DEVICE=br1 TYPE=Bridge BOOTPROTO=static IPADDR=192.168.230.254 NETMASK=255.255.255.0 ONBOOT=yes # IPv6 IPV6INIT=no
Give IPv4 access to the guests being installed :
vi /etc/sysctl.conf # net.ipv4.ip_forward = 1 sysctl -p iptables -t nat -A POSTROUTING -o br0 -s 192.168.230.0/24 -j MASQUERADE service iptables save
DHCP Server
To make it trivial to install KVM guests using virt-install, the host will have a DHCP server on the private bridge :
yum install dhcp vi /etc/dhcp/dhcpd.conf chkconfig --level 345 dhcpd on service dhcpd start
With /etc/dhcp/dhcpd.conf :
ddns-update-style none; subnet 192.168.230.0 netmask 255.255.255.0 { option routers 192.168.230.254; option subnet-mask 255.255.255.0; option domain-name "ovh.rpmfusion.net"; option domain-name-servers 213.186.33.99; range dynamic-bootp 192.168.230.128 192.168.230.160; default-lease-time 21600; max-lease-time 43200; }
Web Server
In order to serve kickstart files as well as the local yum repositories, we also have a trivial static web server. It will be extended to also serve as a Fedora and RPM Fusion mirror :
rpm -Uvh http://download.fedora.redhat.com/pub/epel/6/x86_64/epel-release-6-5.noarch.rpm yum install nginx vi /etc/nginx/nginx.conf # With the important change being to have : location / { root /srv; } chkconfig --level 345 nginx on service nginx start
KVM Guests
se01.ovh.rpmfusion.net
Kickstart file :
# Kickstart file for RHEL6 - se01.ovh.rpmfusion.net install text reboot url --url=http://192.168.230.254/repo/redhat/el6Server/x86_64/install lang en_US.UTF-8 keyboard us # Configure eth0 in post network --device eth0 --bootproto static --noipv4 network --hostname se01.ovh.rpmfusion.net --device eth1 --bootproto static --ip 192.168.230.1 --netmask 255.255.255.0 --gateway 192.168.230.254 --nameserver 213.186.33.99 rootpw tototo firewall --service=ssh --port=domain:udp --port=domain:tcp services --disabled anacron,auditd --enabled=ntpd authconfig --enableshadow --passalgo=sha512 --enablefingerprint selinux --enforcing timezone --utc Europe/Paris bootloader --location=mbr --append="console=ttyS0,9600 crashkernel=auto" clearpart --all --initlabel part / --asprimary --fstype=ext4 --size=100 --grow --fsoptions="commit=120" part swap --asprimary --recommended repo --name="Red Hat Enterprise Linux 6" --baseurl=http://192.168.230.254/repo/redhat/el6Server/x86_64/install/ --cost=100 %packages --nobase @core -rhn-check -rhn-client-tools -rhn-setup -rhnlib -rhnsd -subscription-manager -yum-rhn-plugin bind-utils elinks iptraf lsof man mtr nmap ntp openssh-clients prelink rsync screen strace sysstat tcpdump telnet vim-enhanced wget yum-utils %end %post --log=/root/ks-post.log wget -O /etc/yum.repos.d/el.repo http://192.168.230.254/install/el.repo # Disable all tty consoles, since we run headless (the serial one is active) sed -i -e 's|RUNLEVEL=\[2345\]|RUNLEVEL=[45]|' /etc/init/start-ttys.conf # Hacks to get the main IPv4 address with its /32 mask to work + add IPv6 # eth0 cat > /etc/sysconfig/network-scripts/ifcfg-eth0 << EOF DEVICE="eth0" BOOTPROTO="static" DNS1="213.186.33.99" HWADDR="02:00:00:28:b9:9d" IPADDR="46.105.55.71" NETMASK="255.255.255.255" NM_CONTROLLED="no" ONBOOT="yes" # IPv6 IPV6INIT="yes" IPV6ADDR="2001:41d0:2:ad32::1/64" IPV6_AUTOCONF="yes" EOF # eth1 (to remove DNS and gateway from here) cat > /etc/sysconfig/network-scripts/ifcfg-eth1 << EOF DEVICE="eth1" BOOTPROTO="static" HWADDR="52:54:00:FF:BB:92" IPADDR="192.168.230.1" NETMASK="255.255.255.0" NM_CONTROLLED="no" ONBOOT="yes" EOF # eth0 routes, to get a default gateway working cat > /etc/sysconfig/network-scripts/route-eth0 << EOF # The HV's directly connected network ADDRESS0=188.165.226.0 NETMASK0=255.255.255.0 # The default gateway, on the network above ADDRESS1=0.0.0.0 NETMASK1=0.0.0.0 GATEWAY1=188.165.226.254 EOF %end
Block device and virt-install run :
lvcreate -L16G -nse01 vg0 virt-install --name=se01 --ram=2048 --vcpus=8 --os-variant=rhel6 --hvm --accelerate \ --location=/srv/repo/redhat/el6Server/x86_64/install --disk=path=/dev/vg0/se01 \ --network=bridge=br0,mac=02:00:00:28:b9:9d --network=bridge=br1,mac=52:54:00:ff:bb:92 \ --nographics --keymap=us \ --extra-args="console=ttyS0,9600 ksdevice=eth1 ks=http://192.168.230.254/install/ks/se01.ks"
se02.ovh.rpmfusion.net
lvcreate -L16G -nse02 vg0 virt-install --name=se02 --ram=2048 --vcpus=8 --os-variant=rhel6 --hvm --accelerate \ --location=/srv/repo/redhat/el6Server/x86_64/install --disk=path=/dev/vg0/se02 \ --network=bridge=br0,mac=02:00:00:73:6a:fd --network=bridge=br1,mac=52:54:00:ff:bb:93 \ --nographics --keymap=us \ --extra-args="console=ttyS0,9600 ksdevice=eth1 ks=http://192.168.230.254/install/ks/se02.ks"
se03.ovh.rpmfusion.net
lvcreate -L16G -nse03 vg0 virt-install --name=se03 --ram=2048 --vcpus=8 --os-variant=rhel6 --hvm --accelerate \ --location=/srv/repo/redhat/el6Server/x86_64/install --disk=path=/dev/vg0/se03 \ --network=bridge=br0,mac=02:00:00:aa:ff:cf --network=bridge=br1,mac=52:54:00:ff:bb:94 \ --nographics --keymap=us \ --extra-args="console=ttyS0,9600 ksdevice=eth1 ks=http://192.168.230.254/install/ks/se03.ks"
Note : Since old02 needed its ssh service facing the outside for the CVS service, se03 redirects port 22 to old02 and has its own ssh service available on port 2222 instead of 22.
puppet.ovh.rpmfusion.lan
lvcreate -L4G -npuppet vg0 virt-install --name=puppet --ram=512 --vcpus=4 --os-variant=rhel6 --hvm --accelerate \ --location=/srv/repo/redhat/el6Server/x86_64/install --disk=path=/dev/vg0/puppet \ --network=bridge=br0,mac=52:54:00:ff:bb:a1 --network=bridge=br1,mac=52:54:00:ff:bb:b1 \ --nographics --keymap=us \ --extra-args="console=ttyS0,9600 ksdevice=eth1 ks=http://192.168.230.254/install/ks/puppet.ks"
Then to start the puppetmaster :
yum install puppet-server policycoreutils-python mkdir -p /etc/selinux/local/puppetmaster cd /etc/selinux/local/puppetmaster/ cat > messages << 'EOF' avc: denied { search } for pid=1368 comm="puppetmasterd" name="/" dev=sysfs ino=1 context=unconfined_u:system_r:puppetmaster_t:s0 tcontext=system_u:object_r:sysfs_t:s0 tclass=dir avc: denied { search } for pid=1910 comm="puppetmasterd" name="/" dev=sysfs ino=1 scontext=unconfined_u:system_r:puppetmaster_t:s0 tcontext=system_u:object_r:sysfs_t:s0 tclass=dir avc: denied { getattr } for pid=1368 comm="puppetmasterd" path="/usr/bin/chage" dev=vda1 ino=4005 context=unconfined_u:system_r:puppetmaster_t:s0 tcontext=system_u:object_r:passwd_exec_t:s0 tclass=file avc: denied { execute } for pid=1368 comm="puppetmasterd" name="chage" dev=vda1 ino=4005 scontext=unconfined_u:system_r:puppetmaster_t:s0 tcontext=system_u:object_r:passwd_exec_t:s0 tclass=file avc: denied { name_bind } for pid=1495 comm="puppetmasterd" src=13808 scontext=unconfined_u:system_r:puppetmaster_t:s0 tcontext=system_u:object_r:port_t:s0 tclass=udp_socket avc: denied { node_bind } for pid=1495 comm="puppetmasterd" src=20571 scontext=unconfined_u:system_r:puppetmaster_t:s0 tcontext=system_u:object_r:node_t:s0 tclass=udp_socket avc: denied { getattr } for pid=1586 comm="puppetmasterd" path="/usr/bin/chage" dev=vda1 ino=4005 scontext=unconfined_u:system_r:puppetmaster_t:s0 tcontext=system_u:object_r:passwd_exec_t:s0 tclass=file avc: denied { read } for pid=1650 comm="puppetmasterd" name="config" dev=vda1 ino=11895 scontext=unconfined_u:system_r:puppetmaster_t:s0 tcontext=system_u:object_r:selinux_config_t:s0 tclass=file avc: denied { relabelfrom } for pid=1650 comm="puppetmasterd" name="serial" dev=vda1 ino=14200 scontext=unconfined_u:system_r:puppetmaster_t:s0 tcontext=unconfined_u:object_r:puppet_var_lib_t:s0 tclass=file avc: denied { relabelto } for pid=1822 comm="puppetmasterd" name="serial" dev=vda1 ino=16975 scontext=unconfined_u:system_r:puppetmaster_t:s0 tcontext=system_u:object_r:puppet_var_lib_t:s0 tclass=file avc: denied { open } for pid=1822 comm="puppetmasterd" name="config" dev=vda1 ino=11895 scontext=unconfined_u:system_r:puppetmaster_t:s0 tcontext=system_u:object_r:selinux_config_t:s0 tclass=file avc: denied { getattr } for pid=1870 comm="puppetmasterd" path="/etc/selinux/config" dev=vda1 ino=11895 scontext=unconfined_u:system_r:puppetmaster_t:s0 tcontext=system_u:object_r:selinux_config_t:s0 tclass=file EOF audit2allow -M puppetmaster -i messages; semodule -i puppetmaster.pp vi /etc/puppet/manifests/site.pp service puppetmaster start
To later work with stored configurations and dashboard, there are also needed :
avc: denied { search } for pid=6183 comm="puppetmasterd" name="mysql" dev=vda1 ino=260574 scontext=unconfined_u:system_r:puppetmaster_t:s0 tcontext=system_u:object_r:mysqld_db_t:s0 tclass=dir avc: denied { name_connect } for pid=6183 comm="puppetmasterd" dest=3000 scontext=unconfined_u:system_r:puppetmaster_t:s0 tcontext=system_u:object_r:ntop_port_t:s0 tclass=tcp_socket avc: denied { write } for pid=6183 comm="puppetmasterd" name="mysql.sock" dev=vda1 ino=260680 scontext=unconfined_u:system_r:puppetmaster_t:s0 tcontext=unconfined_u:object_r:mysqld_var_run_t:s0 tclass=sock_file avc: denied { connectto } for pid=6183 comm="puppetmasterd" path="/var/lib/mysql/mysql.sock" scontext=unconfined_u:system_r:puppetmaster_t:s0 tcontext=unconfined_u:system_r:mysqld_t:s0 tclass=unix_stream_socket
And on nodes :
grep -q puppet.ovh.rpmfusion.lan /etc/hosts || echo "192.168.230.11 puppet.ovh.rpmfusion.lan" >> /etc/hosts yum install puppet ruby-rdoc puppet agent --onetime --no-daemonize --logdest=/var/log/puppet/puppet.log --logdest=console --server=puppet.ovh.rpmfusion.lan # Then the usual puppetca --sign on the puppetmaster
bugzilla.ovh.rpmfusion.lan
lvcreate -L4G -nbugzilla vg0 virt-install --name=bugzilla --ram=512 --vcpus=4 --os-variant=rhel6 --hvm --accelerate \ --location=/srv/repo/redhat/el6Server/x86_64/install --disk=path=/dev/vg0/bugzilla \ --network=bridge=br0,mac=52:54:00:ff:bb:c1 --network=bridge=br1,mac=52:54:00:ff:bb:c2 \ --nographics --keymap=us \ --extra-args="console=ttyS0,9600 ksdevice=eth1 ks=http://192.168.230.254/install/ks/bugzilla.ks"
builder1.ovh.rpmfusion.lan
lvcreate -L 256G -n builder1 vg0 virt-install --name=builder1 --ram=2048 --vcpus=4 --os-variant=rhel6 --hvm --accelerate \ --location=/srv/repo/redhat/el6Server/x86_64/install --disk=path=/dev/vg0/builder1 \ --network=bridge=br0,mac=52:54:00:ff:bb:c3 --network=bridge=br1,mac=52:54:00:ff:bb:c4 \ --nographics --keymap=us \ --extra-args="console=ttyS0,9600 ksdevice=eth1 ks=http://192.168.230.254/install/ks/builder1.ks"
DNS Server
The DNS Server is entirely puppet managed. All zones must be edited on the puppet master then applied on se01.
Note that a glue record has been created using the OVH Manager for the ns1.rpmfusion.net entry, pointing to 46.105.55.71.
DNS Servers for RPM Fusion domains :
- ns1.rpmfusion.net
- ns.kimsufi.com
Local Infra Package Repository
On hv02, under /srv/repo/redhat/el6/x86_64/rpmfusion-infra/ are located both "custom/" (home made and/or rebuilt with tweaks) and "epel/" (binaries copied from EPEL) packages. Any required package not provided by the original Red Hat Enterprise Linux 6 repositories must be copied there, under the correct sub-directory. Always manage the files as the rpmfusion user.
To regenerate the repository's metadata :
[rpmfusion@hv02 ~]$ repo /srv/repo/redhat/el6/x86_64/rpmfusion-infra
All of the servers use this repository automatically, enabled from the el.repo installed by kickstart.
Xen Guests Migrated To KVM
Notes
Existing 32bit RHEL5 Xen guests on the es.rpmfusion.net network, the initial infrastructure servers, have been moved as-is. This is because they have legacy services which are still critical to RPM Fusion, such as CVS and Plague, not worth reconfiguring from scratch. Time will be best spent migrating to newer services such as Git and Koji.
Migration notes :
- Initial sync example (best done from the source HV with the stopped VM instead of the running source VM) :
# Server sync rsync --numeric-ids -avHPSx --exclude=lost+found --delete se02.es.rpmfusion.net:/ /srv/old-es/se02 # NFS data sync rsync --numeric-ids -avHPSx --exclude=lost+found --delete files01.es6.egwn.net:/nfs/rpmf/ /srv/nfs
- Kickstart a clean KVM guest, see each server's virt-install command.
- Mount the new LV to be accessible and sync to it :
kpartx -a /dev/vg0/old02 mount /dev/mapper/vg0-old02p1 /mnt/old02 # We'll keep the x86_64 bit grub and kernel+modules, replace all the rest # That way we have a clean MBR, initrd and grub conf (serial console) rsync --numeric-ids -avHPSx --exclude=lost+found --delete --exclude=/boot --exclude=/lib/modules /srv/old-es/se02/ /mnt/old02
- Inside the copied file tree :
# Replace /dev/xvda with /dev/vda1, /dev/xvdb with /dev/vda2 vi /etc/fstab # Replace NFS server : # files01.es6.egwn.lan:/nfs/rpmf -> 192.168.230.254:/srv/nfs vi /etc/fstab files01.es6.egwn.lan:/nfs/rpmf/repo /srv/cache/ nfs tcp,rsize=8192,wsize=8192,hard,intr 0 0 192.168.230.254:/srv/nfs/repo /srv/cache/ nfs tcp,rsize=8192,wsize=8192,hard,intr 0 0 # HOSTNAME=old...ovh.rpmfusion.lan # GATEWAY=192.168.230.3 vi /etc/sysconfig/network # Replace content entirely (new IP addressing, new MAC addresses) vi /etc/sysconfig/network-scripts/ifcfg-eth0 vi /etc/sysconfig/network-scripts/ifcfg-eth1 # domain ovh.rpmfusion.lan # search ovh.rpmfusion.lan # nameserver 213.186.33.99 vi /etc/resolv.conf # Replace xvc0 with ttyS0 # co:2345:respawn:/sbin/agetty ttyS0 9600 vt100-nav vi /etc/inittab # Add ttyS0 vi /etc/securetty # Replace xennet with virtio_net and xenblk with virtio_blk vi /etc/modprobe.conf # Update hostname and IP address (keep only 127.0.0.1 line) vi /etc/hosts # Remove broken rules (!?) CLOSED, icmp, multiport vi /etc/sysconfig/iptables
old02.ovh.rpmfusion.lan
- Old name : se02.es.rpmfusion.net
- Installation :
lvcreate -L 12G -n old02 vg0 virt-install --name=old02 --ram=512 --vcpus=2 --os-variant=rhel5.4 --hvm --accelerate \ --location=/srv/repo/redhat/el5Server/x86_64/os --disk=path=/dev/vg0/old02 \ --network=bridge=br0,mac=52:54:00:ff:bb:c5 --network=bridge=br1,mac=52:54:00:ff:bb:c6 \ --nographics --keymap=us \ --extra-args="console=ttyS0,9600 ksdevice=eth1 ks=http://192.168.230.254/install/ks/old02.ks"
- Services :
- FAS (with PostgreSQL)
- SSH on 22/tcp for CVS_RSH
- CVS on 2401/tcp
- Web :
- FAS on https://fas.rpmfusion.org/accounts/ reverse proxied to http://localhost:8088/accounts
- Source lookaside cache on /repo/pkgs/ for cvs.rpmfusion.org
- ViewVC on cvs.rpmfusion.org
- Completely changed configuration files :
/etc/sysconfig/network-scripts/ifcfg-eth0 :
# Virtio Network Device DEVICE=eth0 HWADDR=52:54:00:FF:BB:C5 IPV6ADDR= IPV6PREFIX= ONBOOT=no
/etc/sysconfig/network-scripts/ifcfg-eth1 :
# Virtio Network Device DEVICE=eth1 BOOTPROTO=static HWADDR=52:54:00:FF:BB:C6 IPADDR=192.168.230.14 NETMASK=255.255.255.0 ONBOOT=yes
old03.ovh.rpmfusion.lan
- Old name : se03.es.rpmfusion.net
- Installation :
lvcreate -L 12G -n old03 vg0 virt-install --name=old03 --ram=512 --vcpus=2 --os-variant=rhel5.4 --hvm --accelerate \ --location=/srv/repo/redhat/el5Server/x86_64/os --disk=path=/dev/vg0/old03 \ --network=bridge=br0,mac=52:54:00:ff:bb:c7 --network=bridge=br1,mac=52:54:00:ff:bb:c8 \ --nographics --keymap=us \ --extra-args="console=ttyS0,9600 ksdevice=eth1 ks=http://192.168.230.254/install/ks/old03.ks"
- Services :
- Plague Server on 8886/tcp and 8887/tcp with both MySQL and PostgreSQL
- Rsync on 873/tcp (unused, koji testing only)
- Web :
- Plague on buildsys.rpmfusion.org (only http, https broken)
- Completely changed configuration files :
/etc/sysconfig/network-scripts/ifcfg-eth0 :
# Virtio Network Device DEVICE=eth0 HWADDR=52:54:00:FF:BB:C7 IPV6ADDR= IPV6PREFIX= ONBOOT=no
/etc/sysconfig/network-scripts/ifcfg-eth1 :
# Virtio Network Device DEVICE=eth1 BOOTPROTO=static HWADDR=52:54:00:FF:BB:C8 IPADDR=192.168.230.15 NETMASK=255.255.255.0 ONBOOT=yes
- Work around plague's strict binding to its configured name :
- Edit /etc/plague/server/plague-server.cfg to have hostname = plague-any in the [General] section.
- Edit /etc/hosts to add an entry with : 0.0.0.0 plague-any