Tag Info

New answers tagged

0

If it's indeed the same drive/partition, you can use the --re-add switch, like so: mdadm --manage /dev/md1 --re-add /dev/sdc5. See this for more information on how it works. I highly recommend a good backup before messing with it.


0

Yes, mdadm will detect such errors, mark the failed drive as defective and drop it from the working array which will continue to function in degraded mode if redundancy is available. But AFAIK mdadm does this at the 'software' level, based on errors it receives from the drive in response to its generic I/O requests (which works with any drive), not by ...


0

Space out multiple VD add/delete operations on a given host by 60 seconds, as each locks up the HBA for a short time, and you don't want that to happen back to back. Also, why are you still using megacli when storcli is available?


1

No, do not defrag an SSD or SSD cache. If you want to defrag your harddrives, it might not be possible using a RAID array as Windows doesn't know what's where on the actual drives. It can only see the logical drive, which means that unless the FakeRAID driver from Intel is smart enough to understand defragmentation at all, you will get zero benefits from ...


2

What is the sense in defragging an SSD? Using up write cycles - that is actually a good reason. Having too much money to burn and this wanting to prematurely age your SSD drives. The gain of defrag is less head movment on a hard disc. SSD do not have this issue to start with - a defrag makes very little sense for them. This is why the windows defragmenter ...


3

It is highly different on different systems. The linux software raid uses a raid superbock, which is on the last 64kB of the each of the member devices. More exactly, if the size of the device is n byte, the raid superblock is at n&~65535-65536. The raid superblock is on the end, because in mirror-like raid levels it makes easier to use a member ...


3

The metadata is stored in the superblock. To delete, I think you want to look at mdadm --zero-superblock. Make sure you have a backup of the metadata before you start fiddling.


0

A good NAS (Thecus/QNAP) would be a bit simpler to configure. Use RAID1 or 10 for your data. If building a nas yourself use SATA conections, do not use USB/USB3.0 (for reliability). PS: Old lead soldered IT equpment may need a new set of electrolytic/polymer capacitors, but still can have longer lifetime expectancy than leadfree (tin) soldered modern kit ...


1

It depends from the workload, but IMHO yes adding 2 additional disks to existing 2 disks array should give better overall performance. You need to realize where the bottlenecks are: CPU - how much data flow CPU can handle, bus/controller - how much data it can carry, SSD/HDD - how much data it can give/take. Let's assume that there is a Linux software ...


0

I would definitely go for a one single RAID 10. Multiple RAIDs, separate for database/system/whatever are 90s thinking. In the age of LVMs and virtual disks, resizeable partitions, etc., this is obsolete and greatly impacts performance. And there is no real benefit. (I am answering this question despite the fact, that a lot of folks have already done the ...


8

In theory yes, more drives in a raid0 would lead to higher performance because the load is shared over more drives. However in practice you would be limited by the the bandwidth of the raid controller, the CPU and memory performance and similar. The performance increase would not be linear, that is 4 disks is not exactly twice as fast as 2 disks. In any ...


0

LV - Logical Volume PD - Physical Disk I did same thing multiple times with active raid1 LV built from two disk within. No data lose as a result. Moreover megacli will refuse to add new LV when PD is used in other LV.


0

I used that (couple months ago to create raid 50): megacli -CfgSpanAdd -r50 -Array0[252:2,252:3,252:4] Array1[252:5,252:6,252:7] WB RA Direct CachedBadBBU -a0 IBM x3560 and: Product Name : ServeRAID M5015 SAS/SATA Controller Serial No : SV14018726 FW Package Build: 12.13.0-0179


0

You have two drives joined in a raid0 stripe. One drive /dev/sda and raid partition /dev/sda6 looks fine. What happened to the second drive? I suspect the second drive is damaged. Does it show up if you run... fdisk -l


6

Is it possible to plug 10 cat5 (network) cables that come from 10 different DSL modems to an HP server so I have 10 different broadband lines connected to one single server? Yes, it's a terrible idea but yes you can. If so how? Buy some multi-port NICs Any suggestions from geniuses in Networking? A hint of sarcasm there but I'll rise ...


0

Have you tried replacing the 500g degraded disk with the 1t and letting the raid 1 copy over the info to it? Then after it is mirrored replace the good 500g drive with your other 1t drive? I'm not familiar with copywipe so I can't give much guidance on that.


1

If either will work, and you're only connecting 7200RPM SATA drives to it, I doubt you'll saturate a PCIe 2.0 1x lane. http://en.wikipedia.org/wiki/PCI_Express States that it should be able to perform at about 500mb/s, per lane (the 4x connector would offer about 2gb/s). PCI Express 2.0 has about a 20% overhead, so your actual maximum throuput is about 400 ...


1

Seriously, you need help regardless RAID controller state without giving it's name?! You have very big chances to recover the data with just pluging one of the mirrors to another computer, assuming that raid state was consistent. If you worry about it you can do non-destructive copy byte-by-byte to other media to create disk image and operate on that. You ...


1

Both approaches are OK. What you want is for your swap to be continuous and not fragmented, and both approaches will ensure that. LVM can cause fragmented volumes, if you extend/shrink them often, but since swap is defined at creation time, it should stay pretty much continuous. With second approach, you have a downside of not being able to easily extend ...


0

Even if it seems to non-match, aborting the resync should not alter your data that fast. If you are running a large-array (like 1TB disks), the resync normally takes hours, so aborting it should keep your data. In my case it worked, even if it seems to be mismatching. Whenever you get your raid running, keep in mind to backup your data afterwards and ...


2

RAID5 is unsuitable full stop A good explanation (from 5 years ago) is here: http://storagemojo.com/2010/02/27/does-raid-6-stops-working-in-2019/


2

Zoredache answered it quite well there. Answering the last question: You will be increasing the probability of failure in a big way since the failure of just one of the devices will make the whole virtual disk fail. That's why most of us use RAID 10.


1

Don't worry about the result not fitting for addition. Just ignore the overflow. There is no ambiguity. For example, let's imagine buckets that hold a single decimal digit. 5 + 4 + 8 + 3 + 6 => 26. Since it only holds one digit, we'll just write "6" for parity. Now we need to reconstruct the fourth data digit. The inverse for addition is subtraction: ...



Top 50 recent answers are included