3 * Get rid of any services running on the host.
4 * Optional, shut down the machine, then wait a month
7 badblocks -v -s -p 5 -w -f DEVICE_GOES_HERE
9 * Revoke puppet certificate
11 * on handel: puppet node clean $HOST
12 * reinit samhain on handel
15 * including references such as subgroup: foo@hostname or supplementarygid: foo@hostname
16 * rerun puppet on backuphost / the backup host
17 * reinit samhain on backuphost
19 * Remove any references from puppet configuration
20 * Remove references to the host from DNS files, including reverse DNS
21 * Get rid of the HW (tell the hoster to get rid of it or other as appropriate)
22 * Update password file information for the $HOST
23 * Remove MQ queues related to machine (see [[doc/mq]] for details)
25 for queue in $(rabbitmqadmin -N rainier list queues name | \
26 grep unger.debian.org | \
27 awk '{print $2}') ; do
28 rabbitmqadmin -N rainier -V dsa delete queue name=$queue
31 * Remove disks from the MSA
32 * run multipath -ll on the KVM host and find all disks on that multipath
35 dm=$(sudo multipath -ll | grep "finzi-srv " | awk '{print $3}')
36 cd /sys/devices/virtual/block/$dm &&
37 for device in $(ls -1); do "echo 1 > /sys/block/$device/device/delete"; done
38 multipath -f finzi-srv
41 * for each disk in those multipath run
43 echo 1 > /sys/block/sdX/device/delete
45 * once done, remove that multpath-device from multipath
47 multipath -f finzi-srv
49 * remove the paths in the MSA
51 Three months after decomissioning: