779
Comment:
|
2087
|
Deletions are marked like this. | Additions are marked like this. |
Line 1: | Line 1: |
== TODO == | = TODO = |
Line 3: | Line 3: |
This is all that has no yet been done or documented. | == Various Articles == |
Line 5: | Line 5: |
* mailserver * configure apt sources in template * firewall * monitoring diskspace * disable saving .debs af download * configure apt to remove .deb files after download /var/cache/apt/archives/ * network planning * keyscript should try thrice |
* https://serverfault.com/questions/122042/kvm-online-disk-resize * https://serverfault.com/questions/378086/how-to-extend-a-linux-pv-partition-online-after-virtual-disk-growth * https://eioki.eu/2013/07/02/live-resize-libvirt-disk-lvm-multipath * http://www.unixarena.com/2015/12/linux-kvm-how-to-addresize-virtual-disk-on-fly.html * apt-get install libguestfs-tools to get virt-resize * https://forum.proxmox.com/threads/extend-rescan-virtio-disk-in-guest-vm.7989/ * debug io trouble: sysstat package + iosat -x 2 5 http://bencane.com/2012/08/06/troubleshooting-high-io-wait-in-linux/ do a page on debugging iostat and how to solve * http://thomasmullaly.com/2014/11/16/the-list-of-os-variants-in-kvm/ * http://lists.gluster.org/pipermail/gluster-users/2015-February/020798.html * https://www.linux.com/news/carp-your-way-high-availability * https://www.server-world.info/en/note?os=Debian_9&p=mariadb&f=1 * http://galeracluster.com/documentation-webpages/sslconfig.html * http://galeracluster.com/documentation-webpages/sslcert.html |
Line 14: | Line 19: |
{{{ Space used on vanilla Filesystem Size Used Avail Use% Mounted on /dev/mapper/vg1-root 1,9G 791M 954M 46% / /dev/md0 243M 53M 162M 25% /boot }}} |
* http://galeracluster.com/documentation-webpages/sslconfig.html#securing-replication-traffic * http://galeracluster.com/documentation-webpages/sslcert.html |
Line 21: | Line 22: |
mail we will configure this system with a mail server this server will later be configured to forward its mails to another mail account |
== misc == |
Line 25: | Line 24: |
* write a ssection about spice and TLS, check this to set up the TLS certificates on kvm02 /etc/libvirt/qemu.conf | * Mailserver * Firewall * Monitoring diskspace * Network planning * Keyscript should try thrice * NTP servers are getting the time from themselves, thats not optimal. https://ubuntuforums.org/showthread.php?t=1513500 only 1 iburst server may be the solution. * Create a page about manual DNSs updates and nsupdate * Spice and TLS write a ssection about spice and TLS, check this to set up the TLS certificates on kvm02 /etc/libvirt/qemu.conf * KVM guest types are in package osinfo-db * move gluster07 to dedicated hardware, so the arbiter works as expected. * Do some pages on optimizations, this is both network, diskio etc. BUT redundancy and stability comes first. * Do some pages on security, for now redundancy comes first * virsh capabilities to check what the hardware can do |
TODO
Various Articles
https://serverfault.com/questions/122042/kvm-online-disk-resize
https://eioki.eu/2013/07/02/live-resize-libvirt-disk-lvm-multipath
http://www.unixarena.com/2015/12/linux-kvm-how-to-addresize-virtual-disk-on-fly.html
- apt-get install libguestfs-tools to get virt-resize
https://forum.proxmox.com/threads/extend-rescan-virtio-disk-in-guest-vm.7989/
debug io trouble: sysstat package + iosat -x 2 5 http://bencane.com/2012/08/06/troubleshooting-high-io-wait-in-linux/ do a page on debugging iostat and how to solve
http://thomasmullaly.com/2014/11/16/the-list-of-os-variants-in-kvm/
http://lists.gluster.org/pipermail/gluster-users/2015-February/020798.html
https://www.server-world.info/en/note?os=Debian_9&p=mariadb&f=1
http://galeracluster.com/documentation-webpages/sslconfig.html
http://galeracluster.com/documentation-webpages/sslcert.html
* http://galeracluster.com/documentation-webpages/sslconfig.html#securing-replication-traffic * http://galeracluster.com/documentation-webpages/sslcert.html
misc
- Mailserver
- Firewall
- Monitoring diskspace
- Network planning
- Keyscript should try thrice
NTP servers are getting the time from themselves, thats not optimal. https://ubuntuforums.org/showthread.php?t=1513500 only 1 iburst server may be the solution.
- Create a page about manual DNSs updates and nsupdate
- Spice and TLS write a ssection about spice and TLS, check this to set up the TLS certificates on kvm02 /etc/libvirt/qemu.conf
- KVM guest types are in package osinfo-db
- move gluster07 to dedicated hardware, so the arbiter works as expected.
- Do some pages on optimizations, this is both network, diskio etc. BUT redundancy and stability comes first.
- Do some pages on security, for now redundancy comes first
- virsh capabilities to check what the hardware can do