Ubuntu Servers :: WRITE Performance Down On RAID 1 Array
Sep 7, 2010
I'm currently experiencing some serious issues with WRITE performance on a RAID-1 array. I'm running Ubuntu 10.04 64 bit server with the latest updates. To evaluate the performance ran the following test: [URL]... (great article btw!) Using dd to measure, write performance is only at 8.7 MB/s. Read is great though at 74.5 MB/s. The tests were ran straight after rebooting and I have not (YET!) done any kernel tuning or customization, running the default server package of the Ubuntu kernel. Here's the motherboard in the server: [URL]... with a beta bios to support drives over 300GB.
[code]...
As you can see from the bo column there is definitely something stalling. As per top output, the %wa (waiting for i/o) is always around %75 however as per above, writes are stalling. CPU is basically idle all the time. Hard drives are quite new and smartctl (smartmontools) does not detect any faults.
View 4 Replies
ADVERTISEMENT
Sep 27, 2010
I have a NETGEAR ReadyNAS NV+ with four 1TB drives in a RAID-5 array. This is our primary file storage. This has previously been backed up to a hardware RAID-0 array directly attached to our Windows server. The capacity of this backup array is no longer sufficient. So the plan was, take a bunch of 200GB to 320GB drives (And a 750) I had kicking around, chuck them in a couple of old SCSI drive enclosures I have collecting dust, attach them via IDA/SATA-to-USB adaptors to a USB hub, attach that to the server, create a JBOD array spanning the disks, and back up the NAS to that. Performance is not an issue as this is just to be used for backup, with the idea being as near to zero cost as possible (Spend so far = NZ$100�ish).
The first hurdle I struck was Windows not supporting Dynamic Disks on USB drives (Required to create a spanned volume). At first I resisted using another machine (i.e. a machine running Ubuntu) as I didn't want to dedicate a piece of hardware to backing up the NAS. I then decided it would be acceptable to do this via a VM, which is what I've done.So I have 10.04 running under VMWare Server 2.0.2 under Windows Server 2008 R2. The disks are all presented to the VM. I wasn't sure if I was going to end up creating the array under LVM or something else, but I noticed Disk Utility has an option to create an array, so I tried that. When I add two 250GB drives, the array size is 500GB. When I then add a 160GB drive, the array size drops to 480GB. Huh? If I keep adding disks (Regardless of order) the final array size comes out at 1.8 TB, as per the attached screenshot. Now with the following drives, I expected something more like:
160 + 250 + 250+ 750 + 250 +200 + 200 + 250 + 320 + 250 + 320 = 3.2TB
Am I missing something or making a false assumption somewhere?
View 4 Replies
View Related
Nov 16, 2010
I wasn't sure where to post this question so administrators, feel free to move it.I have a media server I set up running Ubuntu 10.4 Server, and I set up a software raid 5 using 5 Western Digital Caviar Green 2TB 7200RPM 64MB drives. Individually they benchmark (using the Ubuntu's mdadm GUI (pali?somthing...) at about 100-120mb/s read write.I set the raid 5 up with a stripe size of 256kb, and then I waited the 20 hours it took to synchronize. My read speeds in raid are up to 480mb/s, but my write max is just under 60mb/s. I knew my write performance would be quite a bit lower than my read, but I was also expecting at least single drive performance. I have seen other people online with better results in software, but have been unable to achieve the results they have gotten.
My bonnie++ results are more or less identical (I used mkfs.ext4 and set the stride and stripe-width).The PC has 2048mb of RAM and a 2.93Ghz Dual Core Pentium (Core 2 Architecture), so I doubt think that's the bottle neck. These drives are on the P55 (P45*) South Bridge SATA controller.
View 7 Replies
View Related
Jan 5, 2011
I have recently migrated my file server over to a HP Microserver. The server has two 1TB disks, in a software RAID-1 array, using MDADM. When I migrated simply moved the mirrored disks over, from the old server Ubuntu 9.10 (server) to the new one 10.04.1 (server).I Have recently noticed that write speed to the RAID array is *VERY* slow. In the order of 1-2MB/s order of magnitude (more info below). Now obviously this is not optimal performance to say the least. I have checked a few things, CPU utilisation is not abnormal (<5%) nor is memory / swap. When I took a disk out and rebuilt the array, with only one disk (tried both) performance was as to be expected (write speed >~70MB/s) The read speed seems to be unaffected however!
I'm tempted to think that there is something funny going on with the storage subsystem, as copying from the single disk to the array is slower than creating a file from /dev/zero to the array using DD..Either way I can't try the array in another computer right now, so I though I was ask to see if people have seen anything like this!At the moment I'm not sure if it is something strange to do with having simply chucked the mirrored array into the new server, perhaps a different version of MDADM? I'm wondering if it's worth backing up and starting from scratch! Anyhow this has really got me scratching my head, and its a bit of a pain! Any help here would be awesome, e-cookies at the ready! Cheers
View 6 Replies
View Related
Oct 27, 2010
So I have been doing some RAID 5 performance testing and am getting some bad write performance when configuring the RAID with an even number of drives. I'm running kernel 2.6.30 with software based RAID 5. This seems rather odd and doesn't make much since to me. For RAID 0 my performance consistently increases as I add more drives, but this is not the case for RAID 5. Does anyone know why I might be seeing lower performance when constructing my RAID 5 with 4 or 6 drives rather than 3 or 5?
View 2 Replies
View Related
Mar 3, 2010
I have a 4 drive RAID 5 array set up using mdadm. The system is stored on a seperate physical disk outside of the array. When reading from the array its fast but when writing to the array its extremely slow, down to 20MB/Sec compared to 125MB/Sec reading. It does a bit then pauses, then writes a bit more and then pauses again and so on.The test i did was to copy a 5GB file from the RAID to another spare non-raid disk on the system average speed 126MB/s. Copying it back on to the RAID (in another folder) the speed was 20MB/s.The other thing is very slow several KB/s write speed copying from eSATA drive to the RAID.
View 9 Replies
View Related
Mar 26, 2011
I have an Areca hardware RAID array that I'm trying to format & partition on a fresh Ubuntu 10.04 LTS installation. The OS drive is not on the RAID card, it's entirely separate. The RAID is a 6TB volume so I realize I have to use parted to format it, not fdisk (which I've always relied on).
My problem is that I can't figure out how to get parted to like my settings. It seems like everything I try gives me the warning "Warning: The resulting partition is not properly aligned for best performance." Here's what I'm doing:
Code:
(parted) p
Model: Areca ARC-1280-VOL#00 (scsi)[code].....
What start/end settings should I use to get a properly aligned partition? How do I know?I have tried a mix and match of 0, 0s, 1, 1s, -0, -0s, -1, -1s, 100% for my start/end with no success.
View 8 Replies
View Related
Jun 4, 2010
I just restarted my server (Ubuntu 9.04 server, running on ESXi 4.0) and while copying files onto the server using samba I got strange problems and the connection was lost. When I rebooted the total system, so ESXi as well as Ubuntu Server I did find problems on my RAID disk.
The directory, where the new files were added I have a lot of files, but a lot of them do not have any info except their name:
1304 -rw-rw-rw- 1 spoorhobby spoorhobby 1327274 2010-05-15 22:10 DSCF1895.JPG
? -????????? ? ? ? ? ? DSCF1896.JPG
? -????????? ? ? ? ? ? DSCF1897.JPG
? -????????? ? ? ? ? ? DSCF1898.JPG
[Code].....
Both mirror disks are still functioning and I can still add/delete files, from the server, from other LINUX systems and from other Windows systems via samba.
I did make a full backup on a different server.
View 9 Replies
View Related
Jun 11, 2010
so my servers 7 hds in raid 5 all was working well until one of them died. The HD that died sort of works it can read like half a file also freezes on the benchmark test in disk utility. Unfortunate when i take it out on boot it says. The drive for /media_kbt is not ready or present press s to skip or m for manual recovery. I hit s and then go to disk utility. But i can't start or add disks to the array.
Here is me trying to do random stuff
Code:
administrator@3dslice-host:~$ sudo mdadm --stop /dev/md0
[sudo] password for administrator:
mdadm: metadata format 00.90 unknown, ignored.
mdadm: stopped /dev/md0
administrator@3dslice-host:~$ sudo mdadm --add /dev/md0 /dev/sda1
mdadm: metadata format 00.90 unknown, ignored.
[Code]...
View 2 Replies
View Related
Jun 6, 2011
I have an ubuntu 10.04 machine that I use primarily as a file server. I have a RAID5 array built with mdadm from 3 component disks that worked properly until a recent upgrade (I'm not sure exactly what broke it though). The array is /dev/md0 and is set to mount at /var/media on bootup. *Now*, when the system cold boots it hangs partway through the bootup sequence and throws the following error:
The disk drive for /var/media is not ready yet Press S to skip ... Once I "S"kip this manually, I can see that LOWER in the boot sequence mdadm gets called and assembles the drive, and once fully booted into the system I can then simply do a "mount -a" and the array mounts properly. SO... my gut feeling is that some portion of one of the upgrades changed the order in which things are called, and now the "mdadm assemble" is not triggered until AFTER the system tries to mount the drives. My problem is that I don't know the stuff that controls the boot sequence well enough to dig in the right place.
As a workaround I can remove that entry from /etc/fstab, but then (of course) the system won't auto-mount the array. It's better than the boot process completely hanging because as least THIS I can fix remotely, but I'd really like to know
1) why this broke in an upgrade and is it a known problem?
2) how to get it back to where it auto-assembles and then auto-mounts the array on bootup.
View 9 Replies
View Related
Jun 6, 2011
I have 10.04 server with a linkstation raid 5 attached via usb. What is the best way to monitor the drives for a failure? Its at a remote site
View 2 Replies
View Related
Jun 15, 2011
I am trying to use 3 3TB Western Digital drives in a raid 5 software array. The trouble seems to be that the array is created with only 1.5 TB of capacity, rather then the expected 6 TB.
Here are the commands and output:
$ sudo dmraid -f isw -C BackupFull6 --type 5 --disk /dev/sde,/dev/sdf,/dev/sdg --size=5589G
Create a RAID set with ISW metadata format
RAID name: BackupFull6
RAID type: RAID5
RAID size: 5589G (11720982528 blocks)
RAID strip: 64k (128 blocks)
DISKS: /dev/sde, /dev/sdf, /dev/sdg
About to create a RAID set with the above settings. Continue ? [y/n] :y
$ sudo dmraid -s
*** Group superset isw_cdjhcaegij
--> Subset
name: isw_cdjhcaegij_BackupFull6
size : 3131048448
stride : 128
type : raid5_la
status : ok
subsets: 0
devs : 3
spares : 0
So I cannot understand why the size of the created array is only 3131048448 or about 1.5 TB. The first command seemed to imply it was going to create an array with 5589GB.
System is:
Description: Ubuntu 10.04.2 LTS
Release: 10.04
Codename: lucid
View 8 Replies
View Related
Jun 26, 2011
Ubuntu Server 11.04 i386. I've used linux on and off for years but only in small doses, so I'm really just at newbie level. I was running an Openfiler NAS, but decided to give Ubuntu+Webmin a try. And up 'til now I've been happy with progress. I have set up a RAID-6 array using 5 x 1TB SATA drives. I've ensured that the array is in a "clean" state, and now I want to do some failure testing. The problem occurs when I remove one of the drives in the array. I shutdown, remove a drive, then boot up. The array wont start at all, and comes up with this error during boot:
Quote:
the disk drive for /mnt/raidvol1 is not ready yet or not present
Continue to wait; or Press S to skip mounting or M for manual recovery
If I wait, nothing happens. Obviously the RAID array should start in degraded mode, but it fails to mount at all. When I press "M" to go into manual recovery and type "mount -a" I get the response:
Quote:
mount: special device /dev/RAIDVG1/RAIDLV1 does not exist
I have set BOOT_DEGRADED=true in /etc/initramfs-tools/conf.d/mdadm without success. If I reconnect the disconnected drive, the array works fine, and is in a clean state.
View 9 Replies
View Related
Feb 20, 2011
I've got a couple of new hard disks that I have partitioned (3 partitions per disk) and set up in a mirrored software raid array using mdadm. They've synced, I've put file systems on them (1 x ext4, 2 x luks + ext4) and I can mount them. I've checked the partitions using fdisk. I've checked the filesystems using fsck. So far so good. Next step is that I'd like mdadm to automatically assemble them on boot. (Not bothered about mounting and crypttabing yet.)
I've used sudo /usr/share/mdadm/mkconf to generate a new mdadm.conf with the appropriate UUIDs for the new partitions. I've checked that this matches the output of sudo mdadm --detail --scan
The new lines in this file are:
ARRAY /dev/md9 level=raid1 num-devices=2 UUID=470fb8a6:45561fe0:ebda4a02:9ba7a1ed
ARRAY /dev/md10 level=raid1 num-devices=2 UUID=f351fbba:c704a4b2:ebda4a02:9ba7a1ed
ARRAY /dev/md8 level=raid1 num-devices=2 UUID=c6ccec17:2274588e:ebda4a02:9ba7a1ed
To check that the mdadm.conf is fine I have stopped the new arrays:
[Code].....
View 7 Replies
View Related
Mar 7, 2011
Short story: I have a problem with one of my services (mediatomb) - it requires an md RAID array to be mounted in order to start, because it uses files from it. $remote_fs is added by default to the "Required-Start" line of the init script, so I thought that this should be enough. However, the mediatomb service fails to start on boot, but starts just fine when I execute "service mediatomb start" later. The array is entered in /etc/fstab and is automatically mounted on boot.
Long story...
This is my file server (Ubuntu Server 10.10), which has a raid array created with mdadm (mounted on /z), and the root filesystem is located on an USB thumb drive. I've installed mediatomb, but I wanted to put its database files on the raid array instead of the root fs, so I've symlinked /var/lib/mediatomb (the default path) to /z/mediatomb on the array. This is because the mediatomb DB is supposed to be updated fairly often, so I didn't want it to stay on the flash drive.
Problem is, the mediatomb service can't start on boot - in /var/log/mediatomb.log, it says "2011-03-07 19:22:47 ERROR: /var/lib/mediatomb : 20 x No such file or directory". As I said, it works fine when manually started later...
This is the fstab entry for the raid array code...
View 1 Replies
View Related
Mar 12, 2010
I've recently started having an issue with an mdadm RAID 6 array that been operational for about 2500 hours.
Intermittently during write operations the array stalls, dropping to almost 0 write speed for 10-30 seconds. When this occur one or both of the 2 drives attached to a 2 port Silicon Image si3132 SATA-II controller "locks up" with its activity light locked on. This just started occurring within the last week and didn't seem to coincide with any update that i noticed. The array has just recently passed 12.5% full. The size of the write does not seem to make any difference and it seems completely random. Some times copying a 5 GB dataset results in no slow down other times a torrent downloading to the array at 50kb/sec does cause a slow down and vise versa.
The array consists of 8 WD 1.5TB drives, 6 attached to the ICH9R south bridge, and 2 attached to a si3132 based PCI express card. The array is formatted as a single ext4 partition.
Checking SMART data for all drives shows no errors. Testing read speed with hdparm reports what i would expect (100mb/sec for each drive, ~425mb/sec for the array).
The only thing i did notice is that udma6 is enabled for all the ICH9R drives while only udma5 is enabled for the si3132 drives. Write cache is enabled for all the disks. Attempting to set the si3132 drive to udma6 results in an IO error from hdparm.
The si3132 drive is using the sata_sil24 driver. Nothing of interest appears in the kern or syslog. During this time top shows very high wait time.
The s13132 controller appears to have the original firmware from 2006 loaded, there are some firmware updates available on the Silicon Image website for this controller that now appear to offer separate firmwares for RAID operation (some sort of hybrid controller/software thing the controller supports) and a separate firmware for standard IDE use.
Has anyone had similar issues with this controller? Is a firmware update a reasonable course of action? If so which firmware is best supported by the linux driver?
I know i'm not using its raid features but i've dealt with controllers that needed to be in raid mode for ahci to be active and for linux to work well with them. I'm bit ify at the idea of just trying it and finding out as it could knock 2 disks of my array out of action.
View 2 Replies
View Related
Feb 26, 2011
Using a fresh copy of server 10.04 im trying to simulate a failed raid array on a pair of 2tb disks. Here is the procedure i have been following so far:
- Remove the dead disk partitions from each of the raid 1 arrays (substitute the correct md devices and partitions)
- mdadm /dev/md0 -r /dev/sdb2
- mdadm /dev/md1 -r /dev/sdb3
[code]....
I get an error here that sfdisk does not support gpt (guid partition table). I thought sfdisk did support gpt? It says to use parted, but i cant find a command that copies a partition table over from another disk in parted documentation. Any suggestions? I suppose i could make the partitions manually, but im writing a procedure for people who arent that technical and i need it to be simple enough to be run in my absence. manually building the partitions would be too hard for them.
View 2 Replies
View Related
Aug 1, 2011
I'm running 10.04 x86 server with a really simple installation on a single 250GB boot disk. I then have a RAID5 array as /dev/md0 (set up using mdadm with x4 2TB disks). All is working well. My mdadm.conf file looks like this
Code:
# mdadm.conf
#
# Please refer to mdadm.conf(5) for information about this file.
[code]....
if I was to lose the boot disk and need to remount the RAID array on a fresh installation, what steps do I need to go through. My assumption is that the superblocks on the RAID disks will be used and I don't need to keep any additional information - is this right?
View 6 Replies
View Related
May 31, 2011
I've been having some problems w/ a my RAID 5 array, and after extensive investigation, I'm fairly sure that my last resort is rebuilding the array. I'd tried --assemble, b/c it's a previously created array, but it didn't seem to like that. So, I checked into --create, and it will re-create the array w/out destroying the data, if the superblocks are persistent, which they seem to be. However, here's what I get:
[Code]....
My question is: why do /dev/sdb1 and /dev/sdi1 show as both ext2fs and also as part of a RAID array?
View 3 Replies
View Related
Feb 16, 2010
I have a large RAID array of 12 TB attached to one of my Ubuntu server machines. The RAID volume is formatted with NTFS. The problem is that I can not mount this volume in Ubuntu. I can read it normally if I attach it to windows machine.This is the output from "sudo fdisk -l":
sudo fdisk -l
Disk /dev/sda: 164.7 GB, 164696555520 bytes
255 heads, 63 sectors/track, 20023 cylinders[code]........
View 2 Replies
View Related
Feb 1, 2011
Could any RAID gurus kindly assist me on the following RAID-5 issue?I have an mdadm-created RAID5 array consisting of 4 discs. One of the discs was dropping out, so I decided to replace it. Somehow, this went terribly wrong and I succeeded in marking two of the drives as faulty, and the re-adding them as spare.
Now the array is (logically) no longer able to start:
mdadm: Not enough devices to start the array.Degraded and can't create RAID ,auto stop RAID [md1]
I was able to examine the disks though:
Code:
root@127.0.0.1:/etc# mdadm --examine /dev/sdb2
/dev/sdb2:
Magic : a92b4efc
Version : 00.90.00
code....
Code:
mdadm --create --assume-clean --level=5 --raid-devices=4 /dev/sda2 /dev/sdb2 /dev/sdc2 /dev/sdd2
As I don't want to ruin the maybe small chance I have left to rescue my data, I would like to hear the input of this wise community.
View 4 Replies
View Related
Jun 24, 2009
I've tried to install Fedora 11, both 32 and 64 on my main machine.It could not install as it stops on the first install window. I've already filed a bug but really haven't seen any feed back yet.The bug has something to do with Anaconda and the Raid array but I really can't tell.
I have an Intel Board (see signature). I am running intel raid software under W7 currently.It works fine. But, I'm wondering, when I attempt to install F!!, is my current raid set-up causing problems? Do I need to get rid of the intel raid software and use a Fedor/Linux raid program to manage the raid array??
View 2 Replies
View Related
Jul 16, 2011
I built a RAID5 storage array using 'mdadm' on 3x WD Green 1TB hard drives. I used the Disk Utility GUI to create the array and it took about 24 hours to build. When I started copying files to it I noticed it performed at decent speeds for a while, then got really slow, the sped back up. Just for laughs I ran the Read-Only Benchmark function in the Disk Utility and got a graph that would even confuse stock brokers.Any thought on what the issue might be? I tried searching around for an answer, but most people are only affected by wirte performance issues not reading.
View 5 Replies
View Related
Sep 12, 2010
I've been having some problems copying files to USBs. If I'm copying a large (100MB+) amount of data, at random points the transfer will just stop for 30+ seconds before continuing. Sometimes it doesn't start up again at all. Consequently, the write speed drops to less than 1MB/sec, sometimes as low as 100 KB/sec. I do not have these problems on Windows 7, where I achieve speeds of ~16 MB/sec easily. I have had the same results with several USBs (2-32 GB) on several file systems (fat32, ext2) with several different computers running fully patched versions of Ubuntu 10.04, which suggests the problem is related to the way the OS accesses the hardware.
View 6 Replies
View Related
Aug 28, 2010
I am using Ubuntu 10.04 x64. I am not trying to install Ubuntu on a RAID 1 drive like all of the guides are for. I have a RAID 1 array that I am using for data storage. In windows it shows as a single array just fine. In linux it shows as 2 separate drives. I don't care how they show up to be honest I just have to data written to one drive written to the other automatically as well so my RAID isn't screwed up. Looking through different articles and forums I find a lot of stuff saying that it should show up under /dev/mapper/dxxx or something under /dev/mapper. All that shows up there for me is a device called control which doesn't seem to do something.
View 1 Replies
View Related
Nov 4, 2010
I just installed Ubuntu 10.10 64bit and wanted to get access to my nvidia RAID array. This array is working, and is NTFS formatted. But wasn't showing up through normal means in Ubuntu. (for example the NTFS Configuration Tool didn't display it) Here's what the system showed.
Code:
root@hermes:~# ls -l /dev/mapper/
total 0
crw------- 1 root root 10, 59 2010-11-03 22:39 control
lrwxrwxrwx 1 root root 7 2010-11-03 22:42 nvidia_dadijiag -> ../dm-0
[code]....
Is my mirror still in effect, or did i just mount one of the specific drives from the mirror?
View 1 Replies
View Related
Jan 24, 2011
I'm going a little bit crazy. I can't seem to remove my RAID 1 arrays. Any suggestions? I don't need to save data. The drives are empty. I'm upgrading to 4 2TB drives.Running Lucid Lynx server
Code:
jessica@nas:~$ cat /proc/mdstat
Personalities : [linear] [multipath] [raid0] [raid1] [raid6] [raid5] [raid4] [raid10]
md2 : active raid1 sdb1[1]
976759936 blocks [2/1] [_U]
md0 : active raid1 sdd1[0]
[Code]...
View 5 Replies
View Related
Oct 19, 2010
Consider the following setup: Ubuntu system installed on a separate SSD for speed. An ubuntu software RAID array consisting of X number of physical HDD's for storage (RAID6 or RAID10). RAID setup is done during system install. If I suffer a total crash of the SSD and loose my system, will I be able to, using a new system disk, "reconnect" to the RAID array even if the "mothersystem" of the software RAID is lost? If yes, are there any particular config- or system files I need to backup to be able to rescue the array or will it just be recognized "out-of-the-box" when reinstalling ubuntu?
View 4 Replies
View Related
Mar 17, 2010
I want to make a RAID5 array with 4 2TB hard drives. One of the drives is full of data so I will need to start with a 3 disks and then once I copy the data from the 4th onto the array, I will then add the 4th drive. This will be my first experience with RAID. I've spent a few hours searching for info but most of what I have found is a bit over my head.
View 9 Replies
View Related
Jun 29, 2010
I'm just about to be given a Power Mac G5 (Late 2005) Dual 2.0GHz. I think this was the last G5 produced.I plan on using it as file server/NAS and will probably run 10.04 LTS (or maybe 8.04 LTS). I would install a SATA RAID controller and run 4 1TB drives in a RAID 5 configuration. The only thing I'm unsure about is choosing a compatible RAID controller. I need to find a RAID controller that
- Is PCIe
- Is compatible with both the Power Mac and Ubuntu PPC
- Does true hardware RAID
- Doesn't cost a fortune!
Am I right in thinking that the card might need to be open firmware compatible? If it makes any difference, I plan on running the OS from a separate 5th drive. I've found this on eBay. I asked the seller and he claims it supports true hardware RAID and says the chipset is a Silicon Image SIL3124. I does seem suspiciously cheap though...
View 2 Replies
View Related
