BRRABill's Field Report With XenServer

olivier

We started to implement a way to catch this message properly to give more feedback when it happens (and ideally one day, unmount automagically all guest tools ISO)

olivier

@BRRABill Maybe but if it's possible it's not trivial. I'm afraid you'll have to do it a bit manually until we work on that.

momurda

@BRRABill
xe vm-cd-eject --multiple
I think unmounts all

BRRABill

@olivier said in BRRABill's Field Report With XenServer:

We started to implement a way to catch this message properly to give more feedback when it happens (and ideally one day, unmount automagically all guest tools ISO)

I can do it pretty easily in XC, but I am trying to stay away from XC and only use XO.

olivier

@momurda Yep that's a solution IIRC.

BRRABill

Sweet, I can stay in XO then!

Thanks @momurda

BRRABill

Do I remember a bug of some sort where you could not use XO to reboot the host?

REBOOT on the update tab does not seem to reboot the host.

olivier

@BRRABill Advanced tab, force reboot.

This is fixed in next-release coming soon in stable (we catch the exception in XenServer and display a modal to confirm force if normal reboot can't be done)

BRRABill

I am getting ...

NO_HOSTS_AVAILABLE(OpaqueRef:e0b425cf-7e9c-76e3-d289-c5dde892e7b6)

@olivier I saw your post come in as I was posting this. OK, thanks

olivier

Explanation: a "soft" reboot will try to live migrate the running VMs on this host to another one.

But if it's not possible, the XenServer exception will be NO_HOSTS_AVAILABLE. Now, we are able to catch it and display a extra message to ask for force.

BRRABill

@olivier said in BRRABill's Field Report With XenServer:

Explanation: a "soft" reboot will try to live migrate the running VMs on this host to another one.

But if it's not possible, the XenServer exception will be NO_HOSTS_AVAILABLE. Now, we are able to catch it and display a extra message to ask for force.

Will choosing a regular reboot from the HOST page do a regular XC-like reboot? Or does it have the same issue?

olivier

Should be the same behavior everywhere (at least since we fixed it)

BRRABill

@olivier What does a 0 byte(s) in the installed updates section indicate?

olivier

@BRRABill Nothing. It means the patch was probably installed via another patch (some patches "embed" previous patches). So this exact patch doesn't use space.

That's a bit strange but that's how XS patches work.

Danp

@olivier said in BRRABill's Field Report With XenServer:

If it stopped (no pending tasks related to patches), there is probably something that stopped it.

Check in settings/logs to see if there is something.

edit: you can re-click on the update button. It won't explode.

FWIW, ran into this today with a new XS install. Here's the error from the XO logs --

Nov 30 15:40:55 xo-5 xo-server[980]: Wed, 30 Nov 2016 21:40:55 GMT xen-api [email protected]: pool_patch.apply(...) [17ms] =!> XapiError: PATCH_ALREADY_APPLIED(OpaqueRef:a34b3e4e-838a-0eb2-3ca4-9b1bf13b4c09)
Nov 30 15:40:55 xo-5 xo-server[980]: Wed, 30 Nov 2016 21:40:55 GMT xo:xapi task destroyed: Patch upload (XS70E017)

scottalanmiller

@BRRABill it's been a full year since hte last follow up on this thread. How about an update after another year of use?