Hot answers tagged

38

I hate to say "don't use SATA" in critical production environments, but I've seen this situation quite often. SATA drives are not generally meant for the duty cycle you describe, although you did spec drives specifically rated for 24x7 operation in your setup. My experience has been that SATA drives can fail in unpredictable ways, often times affecting the ...


37

Your options are: Restoring from backups. You do have backups, don't you? RAID is not a backup. Professional data recovery It's possible, though very expensive and not guaranteed, that a professional recovery service will be able to recover your data. Accepting your data loss and learning from the experience. As noted in the comments, large SATA ...


35

I've found that when I've had to tune for lower latency vs throughput, I've tuned nr_requests down from it's default (to as low as 32). The idea being smaller batches equals lower latency. Also for read_ahead_kb I've found that for sequential reads/writes, increasing this value offers better throughput, but I've found that this option really depends on ...


35

You have a double disk failure. This means your data is gone, and you will have to restore from a backup. This is why we aren't supposed to use raid 5 on large disks. You want to set up your raid so you always have the ability to withstand two disk failures, especially with large slow disks.


35

You actually part answered this in your question. The lowest form of RAID is RAID 1. RAID 0 was added well after RAID was defined (can't find reference to a date for this though) The 0 in RAID 0 is used to signify that actually it isn't considered redundant. Think of it as more True/False where 0 is False.


27

Generally I'm pretty sure the answer is no (in fact I know of no controller that does this). Doing such a synchronization would be incredibly difficult - vibration, temperature, natural power supply fluctuation, etc. all have small effects on the platter rotational speed (and if you want to be REALLY picky, the size of a sector). You would need to ...


21

RAID-5 is a fault-tolerance solution, not a data-integrity solution. Remember that RAID stands for Redundant Array of Inexpensive Disks. Disks are the atomic unit of redundancy -- RAID doesn't really care about data. You buy solutions that employ filesystems like WAFL or ZFS to address data redundancy and integrity. The RAID controller (hardware or ...


21

RAID controllers did not (and could not) synchronize disk spindles, but it was an option on some drives. Given a set of identical drives with spindle sync connectors you could ensure a set of disks were all synchronized. I happened to own some Seagate Elite 3 (ancient, obsolete SCSI-2 drives) which I remembered having such a connector so I found the ...


17

It allows the raid card to remember what is in its buffers ( that hasnt been sync'd to disk ) Its very important for people who need high data integrity.. Or to save your DB from certain types of corruption.. (Basically whats on disk, is on disk - so thats safe.. The problem is when the OS thinks its on disk but its actually not and in a RAID card buffer) ...


17

How can a single disk bring down the array? The answer is that it shouldn't, but it kind of depends on what is causing the outage. If the disk were to die in a way that behaved, it shouldn't take it down. But it's possible that it's failing in an "edge case" way that the controller can't handle. Are you naive to think this shouldn't happen? No, I don't ...


17

S.M.A.R.T. is not the final word in disk or storage monitoring!! It's a component, but modern RAID controllers use it along with other methods to determine drive and array health. I'm assuming this is a PERC controller in a Dell PowerEdge server. The normal Linux-friendly approach to health monitoring of Dell hardware is to install the Dell OMSA agents for ...


17

Why would you ask the internet about this? There's so much WTF here, that I don't understand where to start!! This question shows a fundamental lack of understanding of hardware, RAID arrays, storage, monitoring, and general IT best-practices. I read this question and can't help but think: Who is actually responsible for this server hardware? Where is ...


16

It doesn't power the disks, it just keeps the data in the cache for (in this case) up to 72 hours until you bring the machine back on line. When you power the machine back up it will write the contents of the cache back out to the disks. All it does is protect against a power failure. If (for some reason) the machine loses power without cleanly flushing ...


16

It's a waste of time. You won't be able to induce failure or stress the drives in a meaningful manner. You have RAID, and that's a good start. Just make sure you have monitoring in place to actually detect failures as they occur and backups to protect against disaster.


13

The idea with ZFS is to let it known as much as possible how is disks are behaving. Then, from worst to better: Hardware raid (ZFS has absolutely no clue about the real hardware), JBOD mode (The issue being more about any potential expander: less bandwidth), HBA mode being the ideal (ZFS know everything about the disks) As ZFS is quite paranoid about ...


12

If I could suggest a slight modification of your plans: Put the OS on two smaller disks and mirror them. Create a second array, preferably RAID 6 with a hot spare, and make it a dynamic partition within Windows so you can expand later. Don't dynamically expand volumes that are on the system disk. I've heard bad things about that. Keep the system separate ...


12

RAID is just a name with a meaning that changed over time. The important part is that the underlying technology and mechanisms are the same for the RAID levels, so you use the same controller (or piece of software, e.g. mdraid) to achieve all RAID levels.


11

It works like this: Most operating systems have a system call that allows a so-called "synchronous write". This means that during a write operation, if a write has completed then it's guaranteed that it was committed to disk. Synchronous write is therefore non-cached. It blocks the application until it has completed. This kind of operation is obviously ...


11

My favorite example: This can be run from the shell or within the tool. hpacucli ctrl all show config (use hpacucli.exe for Windows) Or hpacucli ctrl all show config detail But, if you have the HP Management Agents installed anyway, you should have realtime monitoring of RAID status pushing back to email alerts or an external monitoring system. Either ...


11

You're over-thinking this. Of course, this depends slightly on the manufacturer's specific implementation, but having deployed thousands of HP ProLiant servers over 10 years, I've experienced hundreds of RAID controller battery failures. I replaced the bad units, knowing that sudden power-loss or a system crash would result in some level of data corruption ...


11

Just recently I read an article by one of Godaddy's engineers about this very topic: Learning to Deal with Learning On their hardware (Dell PERC cards) battery learning cycle happens every 90 days, but no way to know when exactly it'll happen, ie during peak or off-peak hours. They talked about different solutions: Outright disable Battery Learning. ...


10

BIOS-based RAID is little more than a poorer version of software RAID, from everything I've read and encountered. Plus if your mobo is fried, you usually have to replace it with a similar mobo, due to the way that BIOS will format the disks to track the volumes. This is a DEFINITE consideration for recovery of data, as you can't just slap a mirrored disk ...


10

RAID 6 is the only RAID level that will work working under all possible two-drive failure scenarios. RAID 10 can SOMETIMES handle two failures, except when the 2 failed drives are in the same mirrored pair. So, RAID 6 it is.


10

Neither - R5'ing 24 disks is a recipe for disaster, in fact many RAID controllers won't even let you create a single R5 array with more than around 14 disks, the risk is too great. And the 6 x 4 R5's makes no sense either. You don't mention your definition of performance but if you want the fastest configuration then R10 is the way but presumably you don't ...


10

16TiB is the maximum volume size with 4K clusters. You'll need to do one of: reformat with a larger cluster size change the cluster size to 8K (apparently Acronis can do so) create another NTFS volume so you can use that unallocated space. You can then mount that volume onto a folder in your C drive if you prefer having a single drive, but you'll have to ...


10

This is an HP ProLiant server with a Smart Array P420i RAID controller. My immediate advice is to not change any of the default configuration settings unless you have a very specific reason to... In short, don't worry about it. The concept of sectors/tracks in the context of this controller and disk geometry isn't very useful here. Lots of layers of ...


10

You need a SAS expander and/or a server with a disk backplane that has an embedded expander... Please see: RAID card w/1x mini-SAS connector : how do I physically connect 16 disks? and How exactly does a SAS SFF-8087 breakout cable work? + RAID/connection questions


10

With RAID 5 you can only lose 1 disk and have your data remain available. You have lost 3 so you will need to rebuild the RAID and then restore the data from backup. We have a canonical Q&A about RAID levels that may help your understanding.


10

What you have probably heard is that for RAID5 configurations using large disks problems can arise under some circumstances. These problems are not an issue for RAID 1 as the rebuild I/O is less than that of RAID5. What's the recommended size? - that which will do the job with some elbow room for expansion. we will be running daily backup so we can just ...



Only top voted, non community-wiki answers of a minimum length are eligible