0

I'm trying to check the state of the drives on a remote server running Debian wheezy. I'm using a software RAID10 on top of, I guess, xen, since the entries in /dev are /dev/xvda and /dev/xvdb

But it I try

smartctl -a /dev/xvda

I get

/dev/xvda: Unable to detect device type Smartctl: please specify device type with the -d option.

I've tried various device type guesses, none work

Similar problem with hddtemp, which reports

ERROR: /dev/xvda: can't determine bus type (or this bus type is unknown)

I've searched the smartmontools documentation, but can't find any discussion of virtual disks. . .

How do I get behind the virtualisation to something smart tools or hddtemp can work with?

HST
  • 11
  • 3
  • 2
    What are you trying to check on the "drive"? Nearly anything I can think of to check is meaningless in the context of a virtual HDD - it doesn't have a temperature, the fault count is of necessity zero, and so on. If what you're asking is "*how do I see out of the virtual machine to the hypervisor hardware?*", I doubt that's possible, as it would make the virtualisation pointless. – MadHatter Oct 29 '13 at 10:38
  • Fair enough -- so if I want to know if a temperature problem with my disks is the source of (apparently) random crashes, I'm out of luck? – HST Oct 29 '13 at 11:14
  • You don't **have** any discs, only a chunk of space on someone else's discs. A temperature problem with **those** might well bring down the hypervisor - and thus all those hosted images - but assuming you've been in touch with your virtualisation provider and they've confirmed the machine isn't dying, that's not it. – MadHatter Oct 29 '13 at 13:01
  • 1
    Classic [XY problem](http://meta.stackexchange.com/q/66377/189912). Read that link first, then ask about what's really going on. – Michael Hampton Oct 29 '13 at 16:51
  • My real problem is remote server reboots with no evidence I can see as to why. Following the advice at [a serverfault article about precisely that](http://serverfault.com/questions/276179/troubleshooting-linux-server-reboots), I was trying to check HDD temp. And I've gotten useful advice, which is to check with the hosting company if other machines sharing the same disks that I'm using are crashing, so thanks to @MadHatter – HST Oct 30 '13 at 18:32

1 Answers1

1

How about asking your VM provider what is going on with your VM?

In that service request state the times where your VM did reboot.

Nils
  • 7,657
  • 3
  • 31
  • 71