Am Dienstag 07 Dezember 2010 schrieb Martin Steigerwald: > I will restore as much as possible from my backup. Its easily possible > to combine the contents of a backup and a new records file.
I forgot to copy the records file out of my backup before creating the next
backup. And since I do not yet store mutiple revisions in my backup, I
lost lots of entries.
I tried to combine whats left with
cat old-file > records-new
cat old-file2 >> recodrs-new
cat records >> records-new
sort -n -t ":" -f2 records-new | uniq records-sorted
sort -rf -t ":" -f1 records-sorted > records-final
cp -p records records-2010-12-11.bak
mv records-final records
But I lost to much:
mar...@shambhala:~> uprecords | egrep "( up | down |%up)" | cut -c 1-66
up 504 days, 01:46:37 | since Thu Sep 18
down 309 days, 22:35:13 | since Thu Sep 18
%up 61.924 | since Thu Sep 18
Thus I can start from scratch. I had over 92% of uptime before. Hmm, I
don't know how this can happen, since my last backup I could restore was
from oktober. It can't have 200 days downtime. Maybe I try merging again.
Maybe I did something wrong with above merge commands.
I consider loosing all records from time to time to be an epic failure of
either uptimed or the filesystem. But its an Ext4 with those delayed
allocation work-arounds that should help most applications.
I wonder whether I am better of with purging uptimed and be done with it.
Since I do not understand why this happens as the rename case should be
safe and also do not know what to do about it.
I think I will start with a new records file and put my backup cronjobs
everywhere where uptimed is running, for now.
--
Martin 'Helios' Steigerwald - http://www.Lichtvoll.de
GPG: 03B0 0D6C 0040 0710 4AFA B82F 991B EAAC A599 84C7
signature.asc
Description: This is a digitally signed message part.
