sander815 said:
what are the steps to be taken to replace a faulty drive on a softraid1? and how to rebuild the raid again
Great question, will soon face the same problem here, need to take our primairy server down to replace the /dev/hdb disk (or 2nd on the raid 1 software array). But I must admit, I've done some simulations ages ago, but I can't seem to find the documents I made about it back than.
Also a problem I'm facing is this:
I want to make daily SQL/mail backups, the rest may be a once every 2 or 3 day's backup, but you can only configure 1 cron in the sysbk system.
I've already taken sysbk apart to avoid an SQL crash caused by it.
Last night was the first time the SQL backup ran without having to be manually restarted by me, looking forward to the extra sleep
.
My last question, DA backups both databases and the raw data files, which does it use to restore ?
I would really prefer it to use the dumped databases (individual .sql) files, due to the fact I am unsure if the raw data files are backupped without severe corruption, when sysbk shuts down MySQLd on the specific server, it doesn't come back. I have to issue a killall mysqld && service mysqld start command before it works again, so I took the shutting down part out...
edit:
Never only use a raid 1 array as backup, Murphy's law will get to it...
And it isn't safe, what if for some reason a fire breaks out in your server, both drives may be lost. (ok, my solution of storing it on the server that's directly hanging above the master server is not the best solution...)