Verify rpms

[michalgutowski]

• added additional -V option to verify rpms (using both rpm -qa and rpm -Va) before and after the switch
• updated README with debugging section and new -V script option

[mark-au]

Why do we store this output to disk? If we don't need it after script completion should we store it in a temporary directory (mktemp -d)?

[michalgutowski]

The -V option is for users that would like to review and compare rpm info. Since the script recomends rebooting at the end of the procedure it migth happen that a user reboots before reviewing these files and in that case the tmpfs filesystem will be cleared.

[michalgutowski]

/root would probably be the safest option but from my experience /var/tmp/ is a slightly better location because it allows users to also easily fetch these files using scp from a remote location.
In all my switch to OL activites we switched instances in batches (few servers at once) and later run a script that will collect (via scp) all generated verification files to a one cental location for reviewing and comparisions. I know that by default there is a tmpwatch running but it will only delete files that haven't been accessed for more than 30days. What do you think ?

[Djelibeybi]

Merged. In future, please remember to --signoff your commits.