force rescan scsi disks

[flexyz]

Hi

How can I force scsi rescan - new disk size

Thanks
Felix

 

[mir]

See: https://blogs.it.ox.ac.uk/oxcloud/2013/03/25/rescanning-your-scsi-bus-to-see-new-storage/

 

[flexyz]

I tried it, and it does not work - disks remain. If I reboot proxmox the disk is "gone" as expected - I am testing on vmware 5.5

poolraid5 ONLINE 0 0 0
raidz1-0 ONLINE 0 0 0
sdo ONLINE 0 0 0
sdn ONLINE 0 0 0
sdp ONLINE 0 0 0


after reboot and zfs import

poolraid5 DEGRADED 0 0 0
raidz1-0 DEGRADED 0 0 0
sdn ONLINE 0 0 0
11525340032380998529 FAULTED 0 0 0 was /dev/sdn1
sdo ONLINE 0 0 0

Felix

 

[mir]

So you are running virtualized proxmox nodes?
If this is true I think your only option is to reboot the server. You might will be able to do what you want if you install vmware guest utils on the proxmox node.

 

[LnxBil]

I tried it, and it does not work - disks remain. If I reboot proxmox the disk is "gone" as expected - I am testing on vmware 5.5

"gone"? So, you want to remove the disk, not scan for new ones?

If so, just do a

echo 1 > /sys/block/sdXX/device/delete

 

[flexyz]

If I delete it, how can I add a new disk if the rescan does not work?

I am testing basic disk stuff and issues

 

[flexyz]

maybe it works anyway - let me come back

echo "- - -" > /sys/class/scsi_host/host2/scan
echo "- - -" > /sys/class/scsi_host/host3/scan

 

[flexyz]

It works on and off - but if I remove a disk and add a new on with the same SCSI ID from vmware it will not be detected / updated - only if I reboot

If I create a new disk with an previously not used SCSI ID, it works automatically

any clues?

 

[flexyz]

well it don't seems to be very consistent what I see - if I reboot the machine disk changes are detected also on the ZFS pool

It might be related to vmware? - but should be the same behavior as a physical server

Felix