After reinstallation controller mysql defines and adds preserved partition with databases and tries to add 'lost+found' directory
The result is:
[ERROR] Invalid (old?) table or database name 'lost+found'
This happen because /var/lib/mysql is ext4 partition where file system created system directory 'lost+found'
Cluster deployed normally only error messages in logs.
The possible solution is additional directory where mysql should be located for example:
/var/lib/mysql/mysql_databases/
or changing format for this partition to xfs (I prefer this solution)
Step to reproduce:
1) Deploy standard openstack cluster
2) reinstall one controller node (provision, deploy using fuel CLI)
3) look on mysql logs
VERSION:
feature_groups:
- mirantis
production: "docker"
release: "7.0"
openstack_version: "2015.1.0-7.0"
api: "1.0"
build_number: "920"
build_id: "2015-08-11_14-18-42"
nailgun_sha: "8232ae9175a9e92ae8e0899834401ad81c10b30c"
python-fuelclient_sha: "ec5c02b3848fe1d15e5b50b323c3dda030f8e1ac"
fuel-agent_sha: "57145b1d8804389304cd04322ba0fb3dc9d30327"
fuel-nailgun-agent_sha: "e01693992d7a0304d926b922b43f3b747c35964c"
astute_sha: "e1d3a435e5df5b40cbfb1a3acf80b4176d15a2dc"
fuel-library_sha: "fac1d453885e5f808d5b295830a9dc2b93af54b0"
fuel-ostf_sha: "c7f745431aa3c147f2491c865e029e0ffea91c47"
fuelmain_sha: "826387d1a05a40e4849433598442024bb1e3b7c7"
> tries to add 'lost+found' directory
any chance to configure mysql somehow to skip that directory?