Hi all,
I changed a bad disk (automaticly disabled from software raid1 and system for I/O error) in one of our servers and now have problem with adding new one to system without reboot. Does anybody have an experience with this? Or is it possible? :) We're using hotswap AXX6DRV3G for 6 SATA disks from Intel connected directly to MB (S5520HC from Intel too). There is AHCI as driver (enabled in bios), no HW raid. I found, something like that
echo "0 0 0" >/sys/class/scsi_host/host<n>/scan
but it found only sda disk which is already running.. Using CentOS 5.5, x86_64.
Thanks for your ideas and replies ... and excuse my english please :)
Jakub wrote:
I changed a bad disk (automaticly disabled from software raid1 and system for I/O error) in one of our servers and now have problem with adding new one to system without reboot. Does anybody have an experience with this? Or is it possible? :) We're using hotswap AXX6DRV3G for 6 SATA disks from Intel connected directly to MB (S5520HC from Intel too). There is AHCI as driver(enabled in bios), no HW raid. I found, something like that
echo "0 0 0" >/sys/class/scsi_host/host<n>/scan
but it found only sda disk which is already running..
Not sure what you're supposed to echo, there. I learned, where I'm working now, to use scsi-rescan-bus, which seems to work.
Using CentOS 5.5, x86_64.
Thanks for your ideas and replies ... and excuse my english please :)
There's nothing to excuse - it's better than some folks who allegedly speak it as a native language (we'll leave out a former US president, here....)
mark
Not sure what you're supposed to echo, there. I learned, where I'm working now, to use scsi-rescan-bus, which seems to work.
Using CentOS 5.5, x86_64.
Thanks for your ideas and replies ... and excuse my english
please :)
There's nothing to excuse - it's better than some folks who allegedly speak it as a native language (we'll leave out a former US president, here....)
Never misunderestimate our former president's grasp of English. And if anybody has a problem with your English, write to them in Polish. Suddenly they'll LOVE your English. Problem solved.
Does cat /proc/partitions show the new drive? ******************************************************************* This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. If you have received this email in error please notify the system manager. This footnote also confirms that this email message has been swept for the presence of computer viruses. www.Hubbell.com - Hubbell Incorporated**
At Fri, 21 May 2010 15:38:44 +0200 CentOS mailing list centos@centos.org wrote:
Hi all,
I changed a bad disk (automaticly disabled from software raid1 and system for I/O error) in one of our servers and now have problem with adding new one to system without reboot. Does anybody have an experience with this? Or is it possible? :) We're using hotswap AXX6DRV3G for 6 SATA disks from Intel connected directly to MB (S5520HC from Intel too). There is AHCI as driver (enabled in bios), no HW raid. I found, something like that
echo "0 0 0" >/sys/class/scsi_host/host<n>/scan
but it found only sda disk which is already running.. Using CentOS 5.5, x86_64.
Thanks for your ideas and replies ... and excuse my english please :)
I didn't need to do anything special when inserting disks into my (cheap) 4x 2.5" SATA hot swap bay. Just inserted the drive and the HAL/udev deamon pick it up all on its own. My motherboard is a nVidia-based:
00:09.0 SATA controller: nVidia Corporation MCP78S [GeForce 8200] AHCI Controller (rev a2)
And I am using the AHCI driver. CentOS 5.4, x86_64 (xen).
It could be that the SATA controller on your motherboard does not support hotswap detection (not all AHCI controllers support hotswap detection).
On 05/21/2010 03:12 PM, Robert Heller wrote:
I didn't need to do anything special when inserting disks into my (cheap) 4x 2.5" SATA hot swap bay. Just inserted the drive and the HAL/udev deamon pick it up all on its own. My motherboard is a nVidia-based:
What sata 4 bay cage are you using ?
- KB
At Tue, 25 May 2010 14:22:56 +0100 CentOS mailing list centos@centos.org wrote:
On 05/21/2010 03:12 PM, Robert Heller wrote:
I didn't need to do anything special when inserting disks into my (cheap) 4x 2.5" SATA hot swap bay. Just inserted the drive and the HAL/udev deamon pick it up all on its own. My motherboard is a nVidia-based:
What sata 4 bay cage are you using ?
This one from NewEgg:
SNT SNT-SATA1842B 4 x 2.5" HDD in 1 x 5.25" bay SAS / SATA 2.5" Hot Swap Backplane RAID cage http://www.newegg.com/Product/Product.aspx?Item=N82E16817993017
It takes up a single 5.25" bay, has a 4-pin molex power connector, and holds 4 2.5" SATA drives, with front panel access.
I have two Seagate 180GB (ST9160827AS) drives (RAID1) and a Hitachi 120GB (HTS54161) drive (backup disk). The 4th bay is presently empty.
(my Motherboard is this: ASRock K10N78 AM2+/AM2 NVIDIA GeForce 8200 ATX AMD Motherboard http://www.newegg.com/Product/Product.aspx?Item=N82E16813157159 I have a Semperon 1-core processor and 2gig of RAM. )
Just set the BIOS to 'AHCI' mode, added irqpoll to the kernel params (needed to get the kernel to properly talk to the controller), and it just works, both CentOS 4.8 (32-bit [inherited from my previous PIII system]) and CentOS 5.x (64-bit xen).
- KB
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
On 5/21/2010 8:38 AM, Jakub Jedelský wrote:
Hi all,
I changed a bad disk (automaticly disabled from software raid1 and system for I/O error) in one of our servers and now have problem with adding new one to system without reboot. Does anybody have an experience with this? Or is it possible? :) We're using hotswap AXX6DRV3G for 6 SATA disks from Intel connected directly to MB (S5520HC from Intel too). There is AHCI as driver (enabled in bios), no HW raid. I found, something like that
echo"0 0 0" >/sys/class/scsi_host/host<n>/scan
but it found only sda disk which is already running.. Using CentOS 5.5, x86_64.
Thanks for your ideas and replies ... and excuse my english please :)
SATA hotswaps are up to the controller to notice. Some (most?) will do it automatically. If they don't, I don't think there is a way to force it. 'dmesg' should show the disconnect/reconnect activity and the new device name.
I found, something like that echo "0 0 0" >/sys/class/scsi_host/host<n>/scan but it found only sda disk which is already running..
Just for your light reading on this matter: http://www.linuxjournal.com/article/7321
Good article outlining that usage...
2010/5/21 Jakub Jedelský jakub.jedelsky@gmail.com:
Hi all,
I changed a bad disk (automaticly disabled from software raid1 and system for I/O error) in one of our servers and now have problem with adding new one to system without reboot. Does anybody have an experience with this? Or is it possible? :) We're using hotswap AXX6DRV3G for 6 SATA disks from Intel connected directly to MB (S5520HC from Intel too). There is AHCI as driver (enabled in bios), no HW raid. I found, something like that
echo "0 0 0" >/sys/class/scsi_host/host<n>/scan
but it found only sda disk which is already running.. Using CentOS 5.5, x86_64.
Thanks for your ideas and replies ... and excuse my english please :)
Jakub Jedelský e-mail/jabber: jakub.jedelsky@gmail.com http://dev.stderr.cz
I use the following command to rescan all scsi devices
for f in /sys/class/scsi_host/host*; do echo "- - -" > $f/scan; done
You can also rescan the partitions on the devices with
partprobe -s /dev/sd?
Ryan
Jakub Jedelský schrieb:
Hi all,
I changed a bad disk (automaticly disabled from software raid1 and system for I/O error) in one of our servers and now have problem with adding new one to system without reboot. Does anybody have an experience with this? Or is it possible? :) We're using hotswap AXX6DRV3G for 6 SATA disks from Intel connected directly to MB (S5520HC from Intel too). There is AHCI as driver (enabled in bios), no HW raid. I found, something like that
echo "0 0 0" >/sys/class/scsi_host/host<n>/scan
but it found only sda disk which is already running.. Using CentOS 5.5, x86_64.
Thanks for your ideas and replies ... and excuse my english please :)
Jakub Jedelský e-mail/jabber: jakub.jedelsky@gmail.com mailto:jakub.jedelsky@gmail.com http://dev.stderr.cz
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
same mobo (Intel S5520HC) here, and also BIOS configured such that the AHCI driver is being used. I noted the same as you: hotswap does not seem to work - detecting a new disk appears to need a cold start. (I did not try to play with something like echo "0 0 0"
/sys/class/scsi_host/host<n>/scan )
Hotswap does work on all my non-S5520HC mobos that have AHCI, though.
HTH, Kay
-
Brunner, Brian T. -
Jakub Jedelský -
Joseph L. Casale -
Karanbir Singh -
Kay Diederichs -
Les Mikesell -
m.roth＠5-cent.us -
Robert Heller -
Ryan Wagoner