Hello all, so I had a bit of an issue with my blade server, which made me realize the setup I have is not ideal for redundancy. (two 146gb hard drives in raid 0 for the os, four 300gb hard drives in raid 0 for storage)
Now, I am still on the shelf about keeping the four 300gb drives in a raid, it's much more convenient for use as NAS, but of course if one drive fails I lose it all. It saves the hassle of splitting up folders with lots of files/sub folders, but anyways, that's a side point.
The battery for my raid controller died, causing my server to be dead in the water, I ignored it for a long time and finally the battery was around half a volt, instead of 3 volts, stopping me from being able to boot up at all.
I have a TS server and some others with configuration data I do not want to lose, so now that it's back up and running (as well as some issues with windows update, and it's just time to refresh the OS, this thing has been running non stop with few reboots for a long time, and the os install itself is pretty old), I plan on copying this critical data, splitting up the raid, and installing the OS again, with a smaller amount of space, which won't be an issue as I am not using up that much as it is.
My question is, if I set the hard drives up in raid 1, so that I do not have to deal with scheduling windows backup, or risking it being interrupted with a power outage, does this offer instant redundancy?
If a hard drive fails the computer will lock up/crash, obviously, but, if I then remove the damaged drive while waiting for a replacement, or in the case one can not be found, when I reboot will windows be able to run from the redundant drive without issue?
I have a feeling that is the intent, however I am not sure if that is the case. If it helps I am running Windows Server 2008 R2 on an HP Proliant DL380 G4
Side note: I'm also curious if there are any systems that allow a single server to be self redundant, which I suppose would essentially mean running the operating system twice, so that if a hard drive failed, it would continue running with minimal error, or if this would be too much of a shock for the server, and it would just be simpler to have two servers mirrored to eachother to serve this purpose. It would be interesting, although I'm sure any company with the need for that type of redundancy would have the budget for multiple servers.
Now, I am still on the shelf about keeping the four 300gb drives in a raid, it's much more convenient for use as NAS, but of course if one drive fails I lose it all. It saves the hassle of splitting up folders with lots of files/sub folders, but anyways, that's a side point.
The battery for my raid controller died, causing my server to be dead in the water, I ignored it for a long time and finally the battery was around half a volt, instead of 3 volts, stopping me from being able to boot up at all.
I have a TS server and some others with configuration data I do not want to lose, so now that it's back up and running (as well as some issues with windows update, and it's just time to refresh the OS, this thing has been running non stop with few reboots for a long time, and the os install itself is pretty old), I plan on copying this critical data, splitting up the raid, and installing the OS again, with a smaller amount of space, which won't be an issue as I am not using up that much as it is.
My question is, if I set the hard drives up in raid 1, so that I do not have to deal with scheduling windows backup, or risking it being interrupted with a power outage, does this offer instant redundancy?
If a hard drive fails the computer will lock up/crash, obviously, but, if I then remove the damaged drive while waiting for a replacement, or in the case one can not be found, when I reboot will windows be able to run from the redundant drive without issue?
I have a feeling that is the intent, however I am not sure if that is the case. If it helps I am running Windows Server 2008 R2 on an HP Proliant DL380 G4
Side note: I'm also curious if there are any systems that allow a single server to be self redundant, which I suppose would essentially mean running the operating system twice, so that if a hard drive failed, it would continue running with minimal error, or if this would be too much of a shock for the server, and it would just be simpler to have two servers mirrored to eachother to serve this purpose. It would be interesting, although I'm sure any company with the need for that type of redundancy would have the budget for multiple servers.
Facebook
Twitter
Instagram