hi there, I followed a slightly different how to for setting up a root server at strato than this: http://www.howtoforge.com/perfect_setup_debian_sarge_strato but it is basically unimportant for my question. What do I do in case one of the harddisks of my software raid fails? How do I recover? what do I have to do? I guess the server support staff will exchange the hard disk with another one of the same size and brand so what is the next step I have to do?
I never used SW Raid, I preder HW RAid even with cheap controllers without an own CPU because of the ease of use in the OS... but here you can find two shellscripts that manage this harddisk replacement in the sw raid: http://www.sysop-forum.de/thread.php?threadid=2061&hilight=software+raid
thx, its not about what I prefer but about what I can afford the system I could afford and that suits my needs best, offers 2 HDs so I used them to make a software raid... I was just curious what I should do in case it fails... seems the system will go on using only one HD, but how will mdam tell me that one HD failed? sending mail to root?
Hi, On my Dell PowerEdge 1425 SC and Debian Sarge (unofficial 64 bit), i've software raid too. I have configured RAID 1 using the Debian CD. When one HDD crashes for example, i will receive an email from the system with the message that the RAID array is broken. So i can take action. After that i can replace the crashed HDD for a new one. Then Software RAID will rebuild & synchronise both HDD's. Synchronising costs some time, but that is only after setting up RAID 1. That's all.
Ok, thx. thats exactly the info I was looking for: if I have raid1 set-up via software raid, I will be notified if a HD breaks down and when the HD is replaced it will automatically be rebuilt.
I am not sure if the reconstruction is done automagically in software-raid. All I know is that after your replace the defective HDD, you have to do something like this (depending on what drive you have, e.g. sdx for SATA): raidhotadd /dev/md0 /dev/sda Take a look at this: http://www.tldp.org/HOWTO/Software-RAID-HOWTO-8.html#ss8.1 Also, if you have partitions in your drive, maybe you should create them with fdisk before you run the above command. If you need perfect Linux compatibility and good performance, after A LOT OF SEARCH I found out that if you are on a tight budget host-based SATA RAID controllers with Silicon Image 3114 chipset is the way to go, and if money is not an issue full-hardware controllers like LSI Logic MegaRAID SATA 150-4 (4 channels) or LSI Logic MegaRAID SATA 150-6 (6 channels) is the best option. How do you setup the system to notify you with email in case of a drive failure?
