Doing

chkconfig

To Do

New checks to write

Puppet
dmidecode
Mail (queue, spooled, configuration)
Storage - replication, show_disks, emcgrab
Filesystems - inodes, size, large directories?
Detailed CPU state - WIO/USR/SYS
VxVM - vxprint, vxstat, vxdisk list
Memory - physical, virtual, high usage, paging
Networking - multipathing, duplex, speed
Explorers/SOS reports
Packages/Patches
Linux LVM
LDOMs/Containers
NTP
Metastat
ipmi query?
*lom query?
NFS client info (nfsstat)
Shared memory
sysdef -i
Lock Contention (mutex, adaptive, read/write)
nscd/nslcd
Cluster - hastatus, llt, gab, check against auto check script, recent events?
Hostinfo output
Patrol - dumphist, config, parameters from threshold site, is it working?

External data

Recent RAT sessions on the box
Open INCs - pull from RePrioritise
SUDO attempts/commands

Infrastructure tasks

Checksums
Create a cvs/git/svn repo
Should we filter the outputs into standard formats?
Recent changes on the hosts
Vendor call specific modes?
Where do we store the files on the local hosts?
How long do we store the files on the local host?
Code a pull system to bring back indivdiual / fleet wide scans
Custom data collection modes for long term performance analysis
How do we handle hanging scripts?
How do we address additional load - heisenburg principle
Packaging of the check
Analytics of how long it takes to run
Patrol recovery action -> log file monitor -> BEM -> pull state centrally
Scheduled mode - 1200 & 0000
Scan the gathered files for potential errors/warnings
Enable process accounting for in depth performance analysis

Done

List which nfs exports are being used on each host
Linux specific checks
Error checking in the scripts
What gets run at what level?
Sort out the ticket creation functions
Make a better *lib
Crontab
Move away from a monolythic file
fmd
Logging and debugging info
At jobs
How does Pete to the server checks for auditing certain configs?
sysctl -a
/etc/system
Semaphores
Copy full configs (/etc)
ARP