Changes

no edit summary
Line 1: Line 1: −
<noinclude>{{TOCright}}</noinclude>
+
== Info ==
 +
Bender and [[Machine/papert | Papert]] are two twin blade servers donated to Sugar Labs in 2021.
 +
 
 +
Bender is our primary KVM host, while Papert is a hot standby and [[Service/backup|backup]] machine.
    
== Hostnames ==
 
== Hostnames ==
* bender.codewiz.org
+
* bender.sugarlabs.org
* bender.sugarlabs.org
+
* papert.sugarlabs.org
    
== Hardware ==
 
== Hardware ==
* Crappy minitower
+
* HPE ProLiant DL360 Gen10 1RU server
* Core 2 Duo 1.8GHz
+
** Dual socket, current configuration has one CPU
* 4GB RAM
+
* [https://ark.intel.com/content/www/us/en/ark/products/199342/intel-xeon-gold-5218r-processor-27-5m-cache-2-10-ghz.html Xeon Gold 5218R]
* 2x1TB RAID1, 2x320GB RAID1
+
** 20 cores/40 threads
* Fedora 11 x86_64
+
** 2.1GHz base frequency, 4.0GHz max turbo frequency
 +
** 27.5MB of cache
 +
* 64GB RAM
 +
* 2TB NVMe SSD
   −
== Location ==
+
[[Image:BenderRacked.jpg|thumb|320px]]
Hosted by [http://www.develer.com/ Develer]
      
== Admins ==
 
== Admins ==
* [[User:Bernie|Bernie Innocenti]], _bernie on #sugar Freenode
+
* [[User:MrBIOS|Alex Perez]], MrBIOS on #sugar Libera.chat IRC network
* [[User:sascha_silbe|sascha_silbe]] (silbe on #sugar Freenode)
+
* [[User:Bernie|Bernie Innocenti]], @bernie:matrix.org on Sugar Systems
* Stefano Fedrigo <aleph AT develer.com>, sometimes _aleph on Freenode (local access, office hours CET)
+
 
* Develer Infrastructure <it AT lists.develer.com> (local access, office hours CET)
+
== Hosting ==
 +
 
 +
[[Image:SonicColo.jpg|thumb|320px]]
 +
 
 +
Hosted by Sonic in Santa Rosa CA
 +
 
 +
== Network configuration ==
 +
 
 +
Network configuration is managed via [https://netplan.io/ netplan]. To modify, do:
 +
 
 +
vi /etc/netplan/bender.yaml
 +
netplan generate
 +
netplan apply
 +
 
 +
=== IPv4 ===
 +
 
 +
Bender and Papert are globally accessible through a public, static IPv4 address
 +
 
 +
Sonic assigned a /28 network to Sugar Labs. IP assignments are managed in our DNS configuration. Search for "Sonic IP pool" in <code>masters/sugarlabs.org.zone</code>.
 +
 
 +
* Usable IPv4 addresses: 192.184.220.210 ~ 192.184.220.222 (13 addresses)
 +
* Subnet Mask:    255.255.255.240
 +
* Default Gateway: 192.184.220.209
 +
* DNS:            8.8.8.8, 8.8.4.4
 +
 
 +
=== IPv6 ===
 +
IPv6 configuration is a bit weird.
 +
 
 +
* Public block: 2001:5a8:601:f::/64
 +
* Sonic gateway: 2001:5a8:5:3a::15:0/127
 +
* Transport IP: 2001:5a8:5:3a::15:1/127
 +
 
 +
The gateway is configured to route all traffic for our netblock to the transport IP, which is currently assigned to bender.
 +
 
 +
=== Bridges ===
 +
The br0 bridge is created at startup and shared with the virtual machines hosted on Bender. It gives the VMs unfiltered access to the external network. There's no DHCP, all machines must define a static IP configuration, taking care not to stomp onto each other.
 +
 
 +
There is also a virbr0 bridge created by libvirt on startup from <code>/etc/libvirt/qemu/networks/default.xml</code>. This is a NAT interface and is not meant for VMs directly serving on the Internet.
 +
 
 +
Guests simply need to be configured to accept IPv6 routing advertisements. The DNS must be assigned manually.
 +
 
 +
== Virtualization ==
 +
 
 +
Virtual machines are KVM guests managed with libvirt.
 +
 
 +
=== Storage ===
 +
 
 +
All virtual disks come from a pool backed by the main LVM VG:
 +
 
 +
  virsh # pool-define-as nvme-pool logical --source-name nvme-pool
 +
  Pool nvme-pool defined
 +
 
 +
  virsh # pool-start nvme-pool
 +
  Pool nvme-pool started
 +
 
 +
  virsh # pool-autostart nvme-pool
 +
  Pool nvme-pool marked as autostarted
 +
 
 +
  virsh # pool-info nvme-pool
 +
  Name:          nvme-pool
 +
  UUID:          5812819f-b8bf-484e-98fb-2e100fe83df2
 +
  State:          running
 +
  Persistent:    yes
 +
  Autostart:      yes
 +
  Capacity:      1.64 TiB
 +
  Allocation:    250.00 GiB
 +
  Available:      1.40 TiB
   −
== Network configuraiton ==
+
Disks assigned to VMs will appear here:
   −
Bender is globally accessible through public, static IPv4 as wall as
+
virsh # vol-list nvme-pool
an IPv6 /64 subnet (6to4).
+
  Name        Path
 +
-----------------------------------------
 +
  aslo1-root  /dev/nvme-pool/aslo1-root
 +
  aslo1-srv    /dev/nvme-pool/aslo1-srv
 +
  backup      /dev/nvme-pool/backup
   −
Develer has a 10mbit up/downlink and asks to limit bandwidth usage from hosted services.
+
Disks can be created, listed and deleted using the vol-* commands:
There is a traffic shaper, but it may have trouble shaping IPv6 connections.
     −
== Hosted VMs ==
+
  virsh # vol-create-as nvme-pool testvm-root 20G
All buildslaves currently run off bender as KVM virtual machines managed by libvirtd.
+
  Vol testvm-root created
 +
 
 +
  virsh # vol-info testvm-root --pool nvme-pool
 +
  Name:          testvm-root
 +
  Type:          block
 +
  Capacity:      20.00 GiB
 +
  Allocation:    20.00 GiB
 +
 
 +
  virsh # vol-delete testvm-root --pool nvme-pool
 +
  Vol testvm-root deleted
   −
The machines are not globally addressable on IPv4 and are only reachable by IPv6.  The
+
'''NOTE''': avoid allocating large VM volumes as image file within the host's root filesystem, as they're slow and hard to manage. It's ok to use images for test VMs.
IPv6 address is dynamically configured by radvd to a subnet of the public 6to4 net.
  −
Please remember to update the sugarlanbs.org zone when adding new build slaves.