Re: Networking under Xen

[Mark Felder]

On Tue, Jul 14, 2015, at 13:54, Colin Percival wrote:
> On 07/14/15 09:15, Mark Felder wrote:
> > On Tue, Jul 14, 2015, at 07:36, Sydney Meyer wrote:
> >> - Dropped/Stalled Connections with TCP Segmentation Offload and pf
> >> enabled.
> > 
> > TSO is a known issue. I've been turning it off for years to get FreeBSD
> > to play nice on Xen.
> 
> More precisely *pf* is a known issue.  It's inserting TCP checksums
> into TSO packets.  I reported this a long time ago and I don't know
> why pf still hasn't been fixed.
> 

I think Kristof may have solved this in HEAD with all of the care and
feeding he has been doing to pf, but I haven't verified.
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: Networking under Xen

[Mark Felder]

On Tue, Jul 14, 2015, at 07:36, Sydney Meyer wrote:
> Hello everybody,
> 
> i have noticed some odd behaviour with networking under Xen with FreeBSD
> 10 as a DomU.
> 
> - IPv6 (TCP) bandwith drops from ~10 Gbit/s IPv4 to around 3 Gbit/s IPv6.
> (measured with iperf)
> 

What is the "before" and "after" here? When is FreeBSD successfully
doing 10Gbit/s and when isn't it? Is pf enabled? Are you scrubbing?

> - Dropped/Stalled Connections with TCP Segmentation Offload and pf
> enabled.
> 

TSO is a known issue. I've been turning it off for years to get FreeBSD
to play nice on Xen.

> - IPSEC-enabled Kernel TCP Performance drops from ~10 Gbit/s to ~200
> Mbit/s (iperf).
> 

Are you saying FreeBSD non-IPSEC kernel can do 10Gbit/s TCP performance,
but IPSEC kernel immediately drops it to 200Mbit/s?
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: Poor performance with FreeBSD 10.1 under Xen 4.2

[Mark Felder]

On Sat, Mar 28, 2015, at 08:16, Roger Pau Monné wrote:
> 
> I'm Ccing feld because IIRC he found something similar on one of his
> boxes, that also had VTx but no EPT (just like yours). Would it be
> possible for you to try the same set of tests on a different hardware?
> 

Sadly I no longer have access to that hardware, or any FreeBSD on Xen
right now.
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: XenServer 6.5 migrate FBSD 10/11 results in clock reset to 1970?

[Mark Felder]

On Wed, Jan 14, 2015, at 09:16, Karl Pielorz wrote:
> 
> Hi,
> 
> This has been seen before e.g. see
> 
>  
> 
> We're now seeing this now we've started using 10.x boxes under XenServer
> 6.5
> 
> Is there any work around for it?
> 
> The 'hit' rate for us seems to be quite high (+70%?) i.e. the clock
> resets 
> most of the time we do even just a storage migration.
> 
> The guests are running NTP - and that continues running after the event, 
> but  is obviously unwilling to make such a big clock adjustment to drag
> the 
> guests time from 1970 to present day.
> 
> Unfortunately - this also causes various other things on the box to break 
> as well :(
> 
> Is there no 'after migration' hook or script I can lodge some code to 
> shutdown NTP, do an ntpdate - then restart NTP again?
> 

When I ran into this I manually stopped NTP, migrated, ran ntpdate,
started NTP again. It was painful.

The problem as I recall is in the PVHVM code and is fixed upstream in
Xen but wasn't pulled into XenServer. Roger will know more details, but
if you have a Citrix support contract you should pressure them to open a
bug / regression on this. Unfortunately I'm no longer in a position to
do so...
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: disk loss

[Mark Felder]

On Mon, Oct 20, 2014, at 02:09, Egoitz Aurrekoetxea wrote:
> Good morning,
> 
> I would recommend you using NFS instead of iSCSI. It’s far more better to
> handle the connection to disk arrays (the FreeNAS in this situation)
> through a mature and stable protocol like NFS
> and not something manipulating blocks directly. I would advise you to
> rely the responsibility of serving the SR to NFS. 
> 

You can't have redundant paths with NFS (in FreeBSD), though. I'm not so
sure everyone would agree that NFS is mature and stable, either :-)

My personal experience with building a Xen+FreeBSD cluster concluded
that NFS was far too slow and unreliable, and a properly configured
iSCSI with multiple paths and proper alignment was extremely fast.
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: disk loss

[Mark Felder]

On Thu, Oct 9, 2014, at 09:08, Jay West wrote:
> We're running lots of FreeBSD10 VM's  under Xenserver 6.2, and using
> FreeNas
> 9.X for iscsi storage. Note, none of the VM's see iSCSI - just the
> hypervisors see it to create their storage repositories.
> 
>  
> 
> We've noticed an ongoing issue with recovery from crashes or issues with
> the
> NAS. This doesn't happen often at all, but the nas will reboot about once
> a
> year.
> 
>  
> 
> Windows VM's seem to recover fine. FreeBSD, not so much. The pattern
> seems
> to be that if the NAS holding the SR's for the hypervisor reboots, when
> the
> freebsd vm's come up that have just one virtual disk (boot), it complains
> about unclean shutdown and fsck needs to be run. No problem, running fsck
> fixes it. However, if the freebsd vm has two virtual disks (boot and
> data),
> the first one (say, ada0) complains about unclean shutdown and fsck will
> fix
> it. But the second disk (say, ada1) is not found. All we get is "can't
> stat
> /dev/ada1p1" or somesuch, and /dev/ad1 is missing from /dev.
> 
>  
> 
> This seems to be a pretty consistent failure mode. We've tried to detach
> the
> data disk and attach it to a different VM - but no joy.
> 
>  
> 
> 1)  Anyone have any idea why FreeBSD seems so affected and Windows
> doesn't?
> 
> 2)  Any ideas as to how to (non-destructively) get /dev/ada1 back so
> we
> can try to access the data?
> 
>  

Are you rebooting the iSCSI NAS while the servers are running? Does
rescanning the repository help at all? (or whatever the right-click
option is on the SR... I don't have access to XenServer anymore and
can't remember what that option is called)
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: Routing/NAT problem on Xenserver 6.2 with virtual firewall

[Mark Felder]

On Fri, Sep 12, 2014, at 05:42, Karl Pielorz wrote:
> 
> --On 12 September 2014 12:33 +0200 Marko Lerota  
> wrote:
> 
> > Can somebody help me in this situation? I don't know what's wrong.
> > The firewall/NAT doesn't work if the virtual hosts are on the same
> > machine where firewall is. The funny thing is that ICMP packets are
> > passing through, but ordinary traffic does not. Do I have to change
> > something on Xenserver dom0 or PF firewall?
> 
> This is a known bug - see:
> 
>   
> 
> It's also an absolute PITA :( - It also affects DHCP (as I found out a 
> while ago).
> 
> You either have to run a separate pool for the 'router' VM's (and setup
> the 
> VM's accordingly balanced between pools) - or you can run the router VM's 
> in HVM mode only, and they will work (i.e. xn0 etc. become re0 etc.) - 
> performance isn't brilliant in that mode, and also as it's HVM they're
> not 
> 'agile' (so no xen motion migration, no moving storage while they're 
> running).
> 

I'm confident you could patch out the HVM xn0 but keep the rest of the
HVM code so you have fast disk, etc, and you can run the xen tools which
then allows you to use XM and XSM :-) I know Roger has given me a patch
that does this while we were troubleshooting a performance issue.
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: Xen/HVM support in stable/9: GENERIC + xenhvm.ko

[Mark Felder]

On May 16, 2014, at 14:29, Colin Percival  wrote:

> Hi all,
> 
> As of r266269, stable/9 can run in Xen/HVM environments using the GENERIC
> kernel configuration plus a new xenhvm.ko.  This will allow FreeBSD 9.3 to
> run in Xen using "official release binaries" as long as xenhvm_load="YES"
> is placed in /boot/loader.conf.
> 


This is FANTASTIC news! I will begin testing as time permits...

:-)

___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: Poor network performance

[Mark Felder]

On Apr 3, 2014, at 4:26, Egoitz Aurrekoetxea  wrote:

> Hi Jay,
> 
> Tso is a nic's feature. Nothing to do with packet filtering.
> 

There is a known bug with TSO and pf on Xen. There is an open PR -- kern/154428

___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: Terrible performance of XenServer 6.2 and FreeBSD 10.0-RELEASE guest

[Mark Felder]

Is all this work being done over SSH? Do you have a pf firewall? Have you 
disabled TSO?

ifconfig xn0 -tso

or

sysctl net.inet.tcp.tso=0


My suspicion is that the slowness of the shell over the network is making the 
machine seem slower than it should be. There are terrible network issues with 
pf and tso on a XenServer environment. I'm not otherwise aware of severe I/O 
issues on XenServer. I have a fairly large cluster running almost all FreeBSD 
guests without any sort of issues related to I/O.
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: FreeBSD 10.0-R as Xen 'guest' - clarification?

[Mark Felder]

On Wed, Jan 29, 2014, at 9:28, Karl Pielorz wrote:
> 
> And for 'xen tools' - we're still correct to install the port of 
> 'sysutils/xe-guest-utilities' - correct?
> 

Yes, I have a few VMs deployed working marvellously with 10.0-RELEASE
and sysutils/xe-guest-utilities
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: kern/171118: [xen] FreeBSD XENHVM guest doesn' t shutdown cleanly

[Mark Felder]

The following reply was made to PR kern/171118; it has been noted by GNATS.

From: Mark Felder 
To: =?ISO-8859-1?Q?Roger=20Pau=20Monn=E9?= ,
 bug-follo...@freebsd.org
Cc:  
Subject: Re: kern/171118: [xen] FreeBSD XENHVM guest doesn't shutdown 
cleanly
Date: Thu, 31 Oct 2013 09:59:18 -0500

 Your patch works. It would be appreciated if this could make its way
 into 10.0-RELEASE.
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: kern/171118: [xen] FreeBSD XENHVM guest doesn

[Mark Felder]

The following reply was made to PR kern/171118; it has been noted by GNATS.

From: Mark Felder 
To: "bug-followup" 
Cc:  
Subject: Re: kern/171118: [xen] FreeBSD XENHVM guest doesn
Date: Mon, 21 Oct 2013 08:25:47 -0500

 Quoting this in a reply to the PR -- it seems Jason's message was
 rejected by spam perhaps and didn't make it into GNATS?
 
 On Tue, Aug 13, 2013, at 21:01, Jason Shi wrote:
 > After we lanuch a "shutdown" operation in Xen Center, Xen Server issues a
 > clean_shutdown operation. 
 > And actually the signal sended to VM is "halt". We can see that from the
 > xen source log on dom0. e.g.:
 > Aug 13 15:03:48 xenserver-lxpwbbml xenopsd:
 > [debug|xenserver-lxpwbbml|7|Async.VM.clean_shutdown
 > R:832f756cad74|xenops] VM =
 > dc401e38-97af-5972-7150-d067d701fa88; domid = 115; Requesting
 > domain
 > halt
 > 
 > FreeBSD VM receives the signal and do shutdown_nice(RB_HALT). This is a
 > halt action which will not lead to
 >  a poweroff ending. So the VM domain will not be destroyed.
 > Xen Server expected the target domain to be destroyed until it times out.
 > Then Xen Center gets a timeout and 
 > it logs "shutdown failed" then.
 > 
 > I think this a xen bug. 
 > But as a workaround, you can modify the xctrl_halt function from:
 > shutdown_nice(RB_HALT); 
 > to
 > shutdown_nice(RB_POWEROFF|RB_HALT);
 > 
 > Though you may not get exactly what you expect when you really want your
 > VMs to "halt".
 > 
 > 
 > 
 > 
 > Jason Shi
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: kern/171118: [xen] FreeBSD XENHVM guest doesn

[Mark Felder]

The following reply was made to PR kern/171118; it has been noted by GNATS.

From: Mark Felder 
To: shidj ,
 "bug-followup" 
Cc:  
Subject: Re: kern/171118: [xen] FreeBSD XENHVM guest doesn
Date: Mon, 21 Oct 2013 08:00:43 -0500

 On Tue, Aug 13, 2013, at 21:01, Jason Shi wrote:
 > 
 > Though you may not get exactly what you expect when you really want your
 > VMs to "halt".
 > 
 
 I'm struggling to think of a scenario where I want a VM to halt and not
 shutdown.
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: FreeBSD 10.0-Beta1 OVA appliance for XenServer 6.2

[Mark Felder]

On Thu, Oct 17, 2013, at 3:11, pekka.pan...@sofor.fi wrote:
> Hi
> 
> As you all know, you cant install FreeBSD 10.0-Beta1 from ISO CD under 
> XenServer 6.2, because cd-rom problem it has. Hope its somehow fixed in 
> future.
>

Hi Pekka! This has been fixed in 10-STABLE and will be part of the next
BETA release :-) There is also a patch for FreeBSD 9 that they will
hopefully backport. Note, this patch is slightly different than the one
required for 10-STABLE. Big thanks to gibbs and Roger for their work on
Xen.

https://feld.me/freebsd/freebsd9-xen-cdrom-fix.patch
 
> One notice regarding Xen Tools, when you do shutdown option from
> XenCenter 
> its not actually seems to be shutdowning, it says its halted and it 
> waiting key press to reboot.
> I am hoping that is modified in future and shutdown actually shutdowns it 
> as now it leaves its running state under XenServer. Not sure if there is
> a 
> bug report open for that.
> 

I do have a bug report for that -- kern/171118. I had a conversation
with someone not too long ago that I'm looking for in my email. It was
suggested that XenServer sends the wrong command.
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: FreeBSD Alpha5 amd64 - Citrix Xen 6.2 problem

[Mark Felder]

On Mon, Oct 14, 2013, at 9:29, Roger Pau Monné wrote:
> On 14/10/13 15:57, Mark Felder wrote:
> > Patching 9-STABLE with this applies cleanly, but I'm guessing the work
> > done with the restructuring so it fits in GENERIC is not permitting it
> > to build.
> > 
> > 
> > /usr/src/sys/dev/xen/blkfront/blkfront.c: In function 'blkfront_probe':
> > /usr/src/sys/dev/xen/blkfront/blkfront.c:405: warning: implicit
> > declaration of function 'xen_hvm_domain'
> > /usr/src/sys/dev/xen/blkfront/blkfront.c:405: warning: nested extern
> > declaration of 'xen_hvm_domain' [-Wnested-externs]
> > *** [blkfront.o] Error code 1
> > 
> > Stop in /usr/obj/usr/src/sys/XENHVM.
> > *** [buildkernel] Error code 1
> > 
> > Stop in /usr/src.
> > *** [buildkernel] Error code 1
> > 
> > Stop in /usr/src.
> 
> Yes, the xen_hvm_domain function is not available in 9-STABLE, you
> should be able to replace the 'if' gate with a '#ifdef XENHVM .. #endif'
> (not pretty, but should do it's job).
> 

I have just tested that patched on FreeBSD 9.1-p7 and it works
perfectly. I am able to boot the OS with no issues and mounting CDROMs
works fine as well. One thing to note is that it does change the disk
from /dev/adX to /dev/adaX which is something that was happening between
FreeBSD 9 and FreeBSD 10 anyway. Anyone who applies this patch should
update their fstab before rebooting.

https://feld.me/freebsd/freebsd9-xen-cdrom-fix.patch

Thanks again Roger and Justin -- you've just made a huge impact in
FreeBSD usability on XenServer.
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: FreeBSD Alpha5 amd64 - Citrix Xen 6.2 problem

[Mark Felder]

Patching 9-STABLE with this applies cleanly, but I'm guessing the work
done with the restructuring so it fits in GENERIC is not permitting it
to build.


/usr/src/sys/dev/xen/blkfront/blkfront.c: In function 'blkfront_probe':
/usr/src/sys/dev/xen/blkfront/blkfront.c:405: warning: implicit
declaration of function 'xen_hvm_domain'
/usr/src/sys/dev/xen/blkfront/blkfront.c:405: warning: nested extern
declaration of 'xen_hvm_domain' [-Wnested-externs]
*** [blkfront.o] Error code 1

Stop in /usr/obj/usr/src/sys/XENHVM.
*** [buildkernel] Error code 1

Stop in /usr/src.
*** [buildkernel] Error code 1

Stop in /usr/src.
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: FreeBSD Alpha5 amd64 - Citrix Xen 6.2 problem

[Mark Felder]

On Sat, Oct 12, 2013, at 21:48, Justin T. Gibbs wrote:
> On Oct 10, 2013, at 3:09 AM, Shanker Balan  wrote:
> 
> > 
> > On 10-Oct-2013, at 1:37 PM, Shanker Balan  wrote:
> > 
> >> On 09-Oct-2013, at 8:06 PM, Mark Felder  wrote:
> >> 
> >>> On Wed, Oct 9, 2013, at 8:53, Roger Pau Monné wrote:
> >>>> On 09/10/13 13:49, Mark Felder wrote:
> >>>>> On Wed, Oct 9, 2013, at 2:46, Roger Pau Monné wrote:
> >>>>>> On 09/10/13 08:18, Shanker Balan wrote:
> >>>>>>> On 08-Oct-2013, at 9:19 PM, Mark Felder  wrote:
> >>>>>>> 
> >>>>>>>> On Tue, Oct 8, 2013, at 9:45, Josias L.G wrote:
> >>>>>>>>> Problem with Citrix Xen 6.2 and install from ISO. The "solution" was
> >>>>>>>>> remove cd-rom drive from virtual machine. Not possible now with xen
> >>>>>>>>> default in GENERIC kernel.
> >>>>>>>>> Message error: 
> >>>>>>>>> run_interrupt_driven_hooks - still waiting after 300 seconds for
> >>>>>>>>> xenbusb_nop_confighook_cb
> >>>>>>>>> panic: run_interrupt_driven_config_hooks: waited too long
> >>>>>>>>> 
> >>>>>>>> 
> >>>>>>>> I was going to test this soon... but you're right -- you probably 
> >>>>>>>> can't
> >>>>>>>> install FreeBSD 10 from ISO on Citrix XenServer because of this bug.
> >>>>>>>> 
> >>>>>>>> Can someone working on the xen bits test and maybe find a workaround?
> >>>>>>> 
> >>>>>>> The "xenbusb_nop_confighook_cb" issue is the only issue which I am 
> >>>>>>> aware
> >>>>>>> of that prevents CloudStack/XenServer IaaS private clouds from 
> >>>>>>> offering
> >>>>>>> FreeBSD 10 as a supported OS template. The "vbd-destroy" workaround 
> >>>>>>> is not
> >>>>>>> possible as the ISO is attached to the VM instance during the 
> >>>>>>> installation.
> >>>>>>> 
> >>>>>>> A "please pretty please" request to @citrix R&D for the hopefully 
> >>>>>>> last fix
> >>>>>>> to get FreeBSD 10 running on XenServer+CloudStack.
> >>>>>>> 
> >>>>>>> The earlier HyperV related panic on XenServer has been fixed in 
> >>>>>>> ALPHA5.
> >>>>>> 
> >>>>>> Hello,
> >>>>>> 
> >>>>>> I've taken a look into this and I'm afraid there's no easy way to
> >>>>>> workaround it from FreeBSD. When Xen is detected all IDE devices are
> >>>>>> disconnected, and there's no fine grained way to only disable IDE disks
> >>>>>> and not cdrom devices.
> >>>>>> 
> >>>>>> Could you please contact your XenServer representative, and/or submit
> >>>>>> this bug to xs-devel (xs-de...@lists.xenserver.org) mailing lists in
> >>>>>> order to get this fixed on XenServer.
> >>>>>> 
> >>>>> 
> >>>>> Citrix is aware of this as I've contacted several people there and this
> >>>>> has been discussed both here and on the xs-devel list. There has to be
> >>>>> something FreeBSD can do to work around this issue since Linux and
> >>>>> NetBSD have no issues.
> >>>> 
> >>>> Linux and NetBSD have no issues because you probably only tried them on
> >>>> PV mode, which doesn't exhibit this issue (also NetBSD doesn't have
> >>>> PVHVM support, so it's quite clear it won't have this issue).
> >>>> 
> >>>>> As far as I'm aware the issue has been tracked
> >>>>> down to badly behaving qemu in XenServer -- they don't use upstream qemu
> >>>>> in XenServer (yet), and instead have their own fork. A future release is
> >>>>> supposed to merge with upstream qemu.
> >>>> 
> >>>> The main problem here is that XenServer announces a PV block device on
> >>>> xenstore (the cdrom), but then it seems lik

Re: FreeBSD Alpha5 amd64 - Citrix Xen 6.2 problem

[Mark Felder]

On Wed, Oct 9, 2013, at 8:53, Roger Pau Monné wrote:
> On 09/10/13 13:49, Mark Felder wrote:
> > On Wed, Oct 9, 2013, at 2:46, Roger Pau Monné wrote:
> >> On 09/10/13 08:18, Shanker Balan wrote:
> >>> On 08-Oct-2013, at 9:19 PM, Mark Felder  wrote:
> >>>
> >>>> On Tue, Oct 8, 2013, at 9:45, Josias L.G wrote:
> >>>>> Problem with Citrix Xen 6.2 and install from ISO. The "solution" was
> >>>>> remove cd-rom drive from virtual machine. Not possible now with xen
> >>>>> default in GENERIC kernel.
> >>>>> Message error: 
> >>>>> run_interrupt_driven_hooks - still waiting after 300 seconds for
> >>>>> xenbusb_nop_confighook_cb
> >>>>> panic: run_interrupt_driven_config_hooks: waited too long
> >>>>>
> >>>>
> >>>> I was going to test this soon... but you're right -- you probably can't
> >>>> install FreeBSD 10 from ISO on Citrix XenServer because of this bug.
> >>>>
> >>>> Can someone working on the xen bits test and maybe find a workaround?
> >>>
> >>> The "xenbusb_nop_confighook_cb" issue is the only issue which I am aware
> >>> of that prevents CloudStack/XenServer IaaS private clouds from offering
> >>> FreeBSD 10 as a supported OS template. The "vbd-destroy" workaround is not
> >>> possible as the ISO is attached to the VM instance during the 
> >>> installation.
> >>>
> >>> A "please pretty please" request to @citrix R&D for the hopefully last fix
> >>> to get FreeBSD 10 running on XenServer+CloudStack.
> >>>
> >>> The earlier HyperV related panic on XenServer has been fixed in ALPHA5.
> >>
> >> Hello,
> >>
> >> I've taken a look into this and I'm afraid there's no easy way to
> >> workaround it from FreeBSD. When Xen is detected all IDE devices are
> >> disconnected, and there's no fine grained way to only disable IDE disks
> >> and not cdrom devices.
> >>
> >> Could you please contact your XenServer representative, and/or submit
> >> this bug to xs-devel (xs-de...@lists.xenserver.org) mailing lists in
> >> order to get this fixed on XenServer.
> >>
> > 
> > Citrix is aware of this as I've contacted several people there and this
> > has been discussed both here and on the xs-devel list. There has to be
> > something FreeBSD can do to work around this issue since Linux and
> > NetBSD have no issues.
> 
> Linux and NetBSD have no issues because you probably only tried them on
> PV mode, which doesn't exhibit this issue (also NetBSD doesn't have
> PVHVM support, so it's quite clear it won't have this issue).
> 
> > As far as I'm aware the issue has been tracked
> > down to badly behaving qemu in XenServer -- they don't use upstream qemu
> > in XenServer (yet), and instead have their own fork. A future release is
> > supposed to merge with upstream qemu.
> 
> The main problem here is that XenServer announces a PV block device on
> xenstore (the cdrom), but then it seems like there's no backend to
> handle it, so it hangs on the connection phase. IMHO the problem is not
> with the device model (Qemu), but with the backend that should handle
> this PV device.
> 
> Xen only allows you to either disable all IDE devices or none, so the
> only possible solution I can think of is to not disable anything at all
> and use the emulated devices, which will leave us with very poor
> performance (unless I'm missing something, there's no way to only
> disable disks but not cdroms).
> 
> > But the fact remains that this is a non-issue on Linux and NetBSD who
> > handle this buggy virtual CDROM without any problems. There has to be
> > some way we can add a quirk on our side so this device doesn't stop the
> > entire boot process. If FreeBSD 10 is released without out-of-the-box
> > support on the premier commercial Xen platform we'll be shooting
> > ourselves in the foot and all of this work will be for naught. Amazon
> > isn't the only Xen platform people use.
> 
> You can always use the pre-build VM images I guess (I have not tested
> those, but I expect they should work fine under Xen).
> 
> ftp://ftp.nl.freebsd.org/pub/FreeBSD/snapshots/VM-IMAGES/20131007/10.0-ALPHA5/amd64/
> 

Thanks for the reply and further details. I'm hoping I can import the VM
image onto a 6.2 server today and see what happens.
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: FreeBSD Alpha5 amd64 - Citrix Xen 6.2 problem

[Mark Felder]

On Wed, Oct 9, 2013, at 2:46, Roger Pau Monné wrote:
> On 09/10/13 08:18, Shanker Balan wrote:
> > On 08-Oct-2013, at 9:19 PM, Mark Felder  wrote:
> > 
> >> On Tue, Oct 8, 2013, at 9:45, Josias L.G wrote:
> >>> Problem with Citrix Xen 6.2 and install from ISO. The "solution" was
> >>> remove cd-rom drive from virtual machine. Not possible now with xen
> >>> default in GENERIC kernel.
> >>> Message error: 
> >>> run_interrupt_driven_hooks - still waiting after 300 seconds for
> >>> xenbusb_nop_confighook_cb
> >>> panic: run_interrupt_driven_config_hooks: waited too long
> >>>
> >>
> >> I was going to test this soon... but you're right -- you probably can't
> >> install FreeBSD 10 from ISO on Citrix XenServer because of this bug.
> >>
> >> Can someone working on the xen bits test and maybe find a workaround?
> > 
> > The "xenbusb_nop_confighook_cb" issue is the only issue which I am aware
> > of that prevents CloudStack/XenServer IaaS private clouds from offering
> > FreeBSD 10 as a supported OS template. The "vbd-destroy" workaround is not
> > possible as the ISO is attached to the VM instance during the installation.
> > 
> > A "please pretty please" request to @citrix R&D for the hopefully last fix
> > to get FreeBSD 10 running on XenServer+CloudStack.
> > 
> > The earlier HyperV related panic on XenServer has been fixed in ALPHA5.
> 
> Hello,
> 
> I've taken a look into this and I'm afraid there's no easy way to
> workaround it from FreeBSD. When Xen is detected all IDE devices are
> disconnected, and there's no fine grained way to only disable IDE disks
> and not cdrom devices.
> 
> Could you please contact your XenServer representative, and/or submit
> this bug to xs-devel (xs-de...@lists.xenserver.org) mailing lists in
> order to get this fixed on XenServer.
> 

Citrix is aware of this as I've contacted several people there and this
has been discussed both here and on the xs-devel list. There has to be
something FreeBSD can do to work around this issue since Linux and
NetBSD have no issues. As far as I'm aware the issue has been tracked
down to badly behaving qemu in XenServer -- they don't use upstream qemu
in XenServer (yet), and instead have their own fork. A future release is
supposed to merge with upstream qemu.

But the fact remains that this is a non-issue on Linux and NetBSD who
handle this buggy virtual CDROM without any problems. There has to be
some way we can add a quirk on our side so this device doesn't stop the
entire boot process. If FreeBSD 10 is released without out-of-the-box
support on the premier commercial Xen platform we'll be shooting
ourselves in the foot and all of this work will be for naught. Amazon
isn't the only Xen platform people use.
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: FreeBSD Alpha5 amd64 - Citrix Xen 6.2 problem

[Mark Felder]

On Tue, Oct 8, 2013, at 9:45, Josias L.G wrote:
> Problem with Citrix Xen 6.2 and install from ISO. The "solution" was
> remove cd-rom drive from virtual machine. Not possible now with xen
> default in GENERIC kernel.
> Message error: 
> run_interrupt_driven_hooks - still waiting after 300 seconds for
> xenbusb_nop_confighook_cb
> panic: run_interrupt_driven_config_hooks: waited too long
> 

I was going to test this soon... but you're right -- you probably can't
install FreeBSD 10 from ISO on Citrix XenServer because of this bug.

Can someone working on the xen bits test and maybe find a workaround?
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: XenServer 6.2 tools for FreeBSD 9 'Guest'?

[Mark Felder]

On Thu, Aug 29, 2013, at 3:36, Karl Pielorz wrote:
> 
> 
> --On 28 August 2013 13:02 -0400 Adam McDougall 
> wrote:
> 
> > Try the xe-guest-utilities which uses xen-tools.  You may have to hack
> > /usr/local/sbin/xe-update-guest-attrs near the bottom to provide a
> > MinorVersion and/or MicroVersion higher than the default to get the
> > hypervisor to accept it as recent enough.
> 
> I installed the xe-guest-utilities port, and ran up 
> /usr/local/etc/rc.d/xenguest.
> 
> Without changing anything - XenServer has now 'lit' the 'Suspend' options
> - 
> and will let me take snapshots, 'including memory' - so I'd guess it's 
> worked,
> 
> Thanks!
> 

Let me know if you run into any issues as I'm the one who cobbled
xe-guest-utilities together from several other people's previous work.
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: XenServer 6.2 tools for FreeBSD 9 'Guest'?

[Mark Felder]

On Wed, Aug 28, 2013, at 10:53, Roger Pau Monné wrote:
> On 28/08/13 17:31, Karl Pielorz wrote:
> > 
> > Hi,
> > 
> > We've got XenServer now running a couple of FreeBSD 9.2-RC2 VM's
> > (running the XENHVM kernel).
> > 
> > Is there any up to date 'xen tools' setup guide or anything anywhere we
> > can look at?
> > 
> > I've seen 'xen-tools' in the ports collection - but once installed it
> > doesn't seem to do a lot. I'm looking to enable things like suspending
> > VM's and migration etc.
> 
> Using the XENHVM kernel should already provide you with everything
> needed for migration (suspension and resume). From a quick look at the
> xen-tools package, it contains utilities to interact with xenstore,
> which you probably don't need.
> 

Roger,

XenServer and XCP require guest utilities to talk to something in the
hypervisor before they permit you to do migrations. Maybe you could ask
someone at Citrix in the XenServer group to disable that ridiculous
requirement in future releases?
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: CFT: replacing XENHVM kernel config with GENERIC + xenhvm.ko

[Mark Felder]

On Wed, Aug 28, 2013, at 13:28, Colin Percival wrote:
> Hi Roger,
> 
> On 08/28/13 08:44, Roger Pau Monné wrote:
> > On 27/08/13 07:08, Colin Percival wrote:
> >> I've attached a patch which eliminates the XENHVM kernel configuration and
> >> instead allows FreeBSD to run under Xen/HVM with PV drivers by loading a
> >> new xenhvm.ko module from the boot loader.
> > 
> > I'm sorry to say this, but I'm not sure this is the best way to move
> > forward, I would prefer to just have the files merged directly into the
> > GENERIC kernel rather than having to load a module. Also with the PVHVM
> > changes I've posted earlier I'm modifying some common files, which
> > cannot be put into a module, so it makes me wonder if it's worth it to
> > have some Xen specific code into a module while the rest of it is
> > already integrated into GENERIC.
> 
> Are your changes going to be integrated into HEAD in time for
> 10.0-RELEASE?
> (i.e., before the code freeze starts, 10 days from now?)
> 

This is important to me as well. I need to know what to help test if we
have 10 days to nail this down. It will be embarrassing to not have
XENHVM in GENERIC by 10.0.
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: XENHVM and XenServer 6.2

[Mark Felder]

Hi Michał,

To my knowledge this is strictly a qemu issue and should be solved when
XenServer starts shipping upstream qemu instead of their own fork. There
may be a way for FreeBSD to work around it, but I am not aware of anyone
trying.

The current solution is to remove the entire CDROM/DVDROM drive from the
VM to boot a XENHVM kernel. There is no workaround, unfortunately. I
have filed this bug with a couple Citrix contacts of mine but have not
heard back on an ETA for a solution on the XenServer side.
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: kern/154428: [xen] xn0 network interface and PF - Massive performance drop

[Mark Felder]

On Mon, Jul 15, 2013, at 23:20, a...@ahhyes.net wrote:
>  
>  Hi Mark,
>  
>  You're certain TSO is enabled for the NIC? ie: Not disabled via 
>  ifconfig or sysctl?
>  

Ok, I just did some further testing again. I left TSO off in production
because I wanted this to match our other servers for now.

With TSO on there are no more speed issues. Before network operations
were slower than dialup, and this is not the case anymore. However,
there is an added bit of latency to all network operations. For example,
a large text output over ssh via "dmesg" with TSO off is instant. With
TSO on there's a 1 second lag before it reaches my terminal.

It appears it's not fully fixed, but it's getting closer. We'll have to
see if the recent XENHVM work in head has made any further progress. I
haven't had time to test it yet.
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: kern/154428: [xen] xn0 network interface and PF - Massive performance drop

[Mark Felder]

The following reply was made to PR kern/154428; it has been noted by GNATS.

From: "Mark Felder" 
To: bug-follo...@freebsd.org, a...@ahhyes.net
Cc:  
Subject: Re: kern/154428: [xen] xn0 network interface and PF - Massive
 performance drop
Date: Mon, 15 Jul 2013 08:43:30 -0500

 I wasn't able to replicate this on an 8.4 XENHVM kernel -- perhaps this  
 has now been fixed?
 
 When 9.2-RELEASE drops we should test there as well before closing this  
 out.
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

XenServer 6.2 released, going fully open source

[Mark Felder]

Looks like Citrix finally got their lawyers to sign off on it --
XenServer is now fully open source. XCP will go away and users will be
upgrading to XenServer 6.2.

http://blog.xen.org/index.php/2013/06/25/xenserver-org-and-the-xen-project/


I have not yet tested XenServer 6.2 to see if it resolves the
"unable-to-boot-with-cdrom-on-FreeBSD-guest" bug yet, but I've heard the
upgrade path will definitely require you jump from XCP 1.6 to XenServer
6.2, so anyone stuck on XCP 1.5beta will have to remove their virtual
cdrom drives before attempting the upgrade process.

Also, it looks like XenCenter is being released under a BSD-like
license. Hopefully someone gets this ported to *nix soon

http://blog.xen.org/index.php/2013/06/25/xenserver-org-and-the-xen-project/
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: FreeBSD PVHVM call for testing

[Mark Felder]

On Thu, 23 May 2013 12:49:03 -0500, Outback Dingo   
wrote:
actually this tells you how to remove/eject the virtual iso, not the  
cdrom

device from the VM itself in XCP



gah! you're completely right! However, the bottom picture with the red box  
showing "click here to create a dvd drive" is what you should see when you  
DO destroy the dvd drive device.


It should be something like

xe vbd-list vm-name-label=VM_NAME

the device saying "empty (RO): TRUE is the dvd rom, get the uuid

xe vbd-destroy uuid=uuid_of_that_device
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: FreeBSD PVHVM call for testing

[Mark Felder]

On Thu, 23 May 2013 12:29:34 -0500, Outback Dingo   
wrote:




Thanks, but this is XCP not a generic XEN server where there are vm  
config

files under /etc/xen/ in XCP they dont exists


This should cover it

http://support.citrix.com/article/CTX132411
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: kern/176471: [xen] xn driver crash on detach

[Mark Felder]

The following reply was made to PR kern/176471; it has been noted by GNATS.

From: "Mark Felder" 
To: bug-follo...@freebsd.org, p...@helenius.fi
Cc:  
Subject: Re: kern/176471: [xen] xn driver crash on detach
Date: Tue, 21 May 2013 09:52:28 -0500

 I've been running this in production since March 1st with no problems. I  
 also would like to see this committed ASAP.
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: FreeBSD PVHVM call for testing

[Mark Felder]

On Thu, 16 May 2013 05:45:35 -0500, Roger Pau Monné   
wrote:




Thanks for testing this, I'm afraid I'm not really familiar with
XenServer or XenCenter. It looks like Linux users are also running into
problems when trying to setup Linux PVHVMs on XenServer 6.1:


He's hitting a well-known bug that's had a lot of discussion on this list  
and a little bit on one of the xen lists. I'm waiting for someone at  
Citrix to fix this, but it doesn't seem to be a priority. Considering your  
email is @citrix.com, it's a little bit disheartening that you're not  
familiar with XenServer or XCP -- the only Xen platform where a business  
can get support direct from the vendor. :( Please use whatever resources  
you can to prioritize this issue as we'd love to upgrade our  
virtualization environment to 1.6 ASAP.


From all the information I've gathered it appears there's a regression in  
qemu used in XenServer/XCP that prevents FreeBSD XENHVM from booting on  
XenServer 6.1 / XCP 1.6. Earlier versions it works fine. NetBSD and  
Linux-based Xen installs don't tend to have this problem because they use  
upstream qemu directly. I'm waiting for the next release (XenServer 6.2?)  
to come out so I can test again as they're supposed to be using upstream  
directly in future releases.


Sergey, you can work around this by removing the CDROM/DVDROM device from  
the VM. This will allow you to boot the OS and use it. You could also step  
back down to XenServer 6.0 or XCP 1.5beta to work around it, but I'm sure  
you're relying on some of the new features 6.1 has to offer.

___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: misc. questions

[Mark Felder]

On Wed, 06 Mar 2013 16:40:38 -0600, Colin Percival   
wrote:


You'll want to turn off tso, since it produces long mbuf chains which  
most
xn netbacks choke on.  (I have a very ugly workaround patch for this  
which I
use on EC2, but simply turning off tso is enough unless you need Gbps+  
speeds).


Can you link me to this patch? I have an environment that might warrant  
using it for now.



Thanks!
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: misc. questions

[Mark Felder]

On Wed, 06 Mar 2013 14:17:02 -0600, Jay West  wrote:


1)One or two of our freebsd 9.1 HVM (with PVM drivers) under
Xenserver 6.1 advanced are fairly frequently generating this message on  
the
console: xn_txeof: WARNING: response is -1! Any ideas what this may be  
and

what should be done? It does seem to only occur on the machines that have
higher network load than the others.


You're having success with Xenserver 6.1? That's good news. I wonder why  
XCP 1.6 is unable to boot FreeBSD if there's an emulated DVDROM? :(



2)We did a pilot project of about 8 VM's (the above mentioned
environment) and all went well. Now that we've moved it into production  
with

many more VM's, I'm wondering about recommended tuning. I seem to recall
from watching this list that there are a few sysctl's and the like that  
are
highly recommended, I think they had to do with network settings and  
turning

off "offloading" or somesuch. Does anyone have a quick & dirty list of
"here's the first things you should always change" with regards to  
FreeBSD

HVM (pvm drivers) under XenServer?


If you're using pf, you will certainly need to set net.inet.tcp.tso=0

As far as the other offloading -- I put this on my NICs:

ifconfig_xn0="(your stuff here) -txcsum -rxcsum -lro -tso"

3)When migrating bare-metal non-VM FreeBSD machines  
(primarily

webservers and mailservers) to the above Xenserver environment, we have
always just created VM's from an ISO, installed apache, sendmail, etc.  
and
then migrated the websites, mailboxes, etc. manually "across the wire"  
from

the non-VM machines to the replacement VM's. Xenconvert does not seem to
support FreeBSD/gpart/ufs. Does anyone know of a way (software or  
procedure)
to take a bare-metal FreeBSD 9.1 install and turn it into a VDI or  
OVF/OVA

that can be imported to Xenserver?


We usually install the same version on the VM and rsync the entire OS  
over. Then replace the kernel with the XENHVM one, do a bit of tweaking  
(rc.conf, pf.conf, fstab), and move on.


FYI, when we build kernels we do:

make buildkernel KERNCONF=XENHVM
make installkernel KERNCONF=XENHVM DESTDIR=/boot/ KODIR=9.1-XENHVM

This puts the kernel and modules in /boot/9.1-XENHVM and leaves the  
GENERIC kernel alone (as an emergency fallback)


/boot/loader.conf then contains kernel="9.1-XENHVM"

Note: for 9.0 it doesn't install the modules when you do it this way. We  
just set module_path="/boot/kernel;/boot/modules" and piggyback on the  
GENERIC kernel's modules.

___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: kern/176471: [xen] xn driver crash on detach

[Mark Felder]

The following reply was made to PR kern/176471; it has been noted by GNATS.

From: Mark Felder 
To: bug-follo...@freebsd.org, p...@helenius.fi
Cc:  
Subject: Re: kern/176471: [xen] xn driver crash on detach
Date: Fri, 1 Mar 2013 08:37:40 -0600

 ZKr8kQLgp1eZBwnNXnyCtT
 Content-Type: text/plain; format=flowed; delsp=yes
 
 I had to reformat the patch (the one in this PR didn't come through well)  
 but I've tested this on FreeBSD 9.0 and our Xen environment and I can  
 successfully add/remove network interfaces and change their properties  
 without crashes.
 
 I still have to attempt this on the FreeBSD 9.1 VMs but I don't expect  
 different results.
 
 
 Thank you very much Roger!
 
 ZKr8kQLgp1eZBwnNXnyCtT
 Content-Disposition: attachment; filename=xen-netfront.patch
 Content-Type: application/octet-stream; name=xen-netfront.patch
 Content-Transfer-Encoding: base64
 
 LS0tIHN5cy9kZXYveGVuL25ldGZyb250L25ldGZyb250LmMub3JpZwkyMDEzLTAzLTAxIDA2
 OjUyOjQ2LjAwMDAwMDAwMCAtMDYwMAorKysgc3lzL2Rldi94ZW4vbmV0ZnJvbnQvbmV0ZnJv
 bnQuYwkyMDEzLTAzLTAxIDA2OjU0OjE5LjAwMDAwMDAwMCAtMDYwMApAQCAtMjE5OSw3ICsy
 MTk5LDE0IEBACiBzdGF0aWMgdm9pZAogbmV0aWZfZnJlZShzdHJ1Y3QgbmV0ZnJvbnRfaW5m
 byAqaW5mbykKIHsKKwlYTl9MT0NLKGluZm8pOworCXhuX3N0b3AoaW5mbyk7CisJWE5fVU5M
 T0NLKGluZm8pOworCWNhbGxvdXRfZHJhaW4oJmluZm8tPnhuX3N0YXRfY2gpOwogCW5ldGlm
 X2Rpc2Nvbm5lY3RfYmFja2VuZChpbmZvKTsKKwlpZm1lZGlhX3JlbW92ZWFsbCgmaW5mby0+
 c2NfbWVkaWEpOworCWV0aGVyX2lmZGV0YWNoKGluZm8tPnhuX2lmcCk7CisJaWZfZnJlZShp
 bmZvLT54bl9pZnApOwogI2lmIDAKIAljbG9zZV9uZXRkZXYoaW5mbyk7CiAjZW5kaWYK
 
 ZKr8kQLgp1eZBwnNXnyCtT--
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: kern/171138: [xen] [panic] Deactivating network interface produces panic on XENHVM

[Mark Felder]

The following reply was made to PR kern/171138; it has been noted by GNATS.

From: Mark Felder 
To: bug-follo...@freebsd.org
Cc:  
Subject: Re: kern/171138: [xen] [panic] Deactivating network interface
 produces panic on XENHVM
Date: Fri, 1 Mar 2013 08:37:40 -0600

 This is a duplicate of pr 176471 and should be closed.
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: kern/176471: [xen] xn driver crash on detach

[Mark Felder]

On 28.02.2013 12:00, Roger Pau Monné wrote:


 Also, what do you mean by "changing configuration of xn network"?


He means changing the native VLAN of the interface attached to the Xen 
VM. I have also hit this bug and will gladly test this patch tomorrow 
when I'm in the office. It's probably the exact same bug, so if you 
fixed the add/remove interface panic I'd assume it fixes this other one.


Thanks for looking into this, Roger!
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: kern/175954: [xen] XENHVM xn network driver extreme packet loss during a slow download

[Mark Felder]

The following reply was made to PR kern/175954; it has been noted by GNATS.

From: Mark Felder 
To: bug-follo...@freebsd.org, m8r-67o...@mailinator.com
Cc:  
Subject: Re: kern/175954: [xen] XENHVM xn network driver extreme packet loss
 during a slow download
Date: Wed, 13 Feb 2013 12:22:26 -0600

 Is this server running a pf firewall? And does setting net.inet.tcp.tso=0  
 resolve it?
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: kern/175822: [xen] FreeBSD 9.1 does not work with Xen 4.0

[Mark Felder]

On Tue, 05 Feb 2013 05:49:01 -0600, Jeroen van der Ham   
wrote:


The system is then not respsonsive, with just the message scrolling by  
with the number increasing.


If memory serves me this is a well known Xen regression that's solved by  
upgrading to Xen 4.1 or 4.2. I don't believe this is something that  
FreeBSD can fix.

___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: Upgrading XENHVM 9.0 to 9.1

[Mark Felder]

On Wed, 23 Jan 2013 15:48:29 -0600, Jeroen van der Ham   
wrote:



Hi,

Today I started with an upgrade process of my VPS which has been running  
9.0 XENHVM almost a full year. I followed the process as outlined on the  
Handbook[1]. I may have made a mistake in placing back the right old  
kernel, but a new GENERIC kernel did not get installed.


I was able to boot into the 9.1 system using my old 9.0 XENHVM kernel. I  
have been trying to fix this, but I keep running into problems:


- I manually downloaded and extracted the 9.1 GENERIC kernel. Booting  
from that results in a rootmount mismatch (ada0 vs ad0). Just renaming  
it in /etc/fstab does not seem to be enough, what else would I need to  
do?


Yes, the GENERIC kernel uses adaX and the XENHVM kernel uses adX. It  
shouldn't be a problem to just change it in the fstab though. I've not yet  
had problems running 9.1 with XENHVM yet.




- I manually built a new 9.1 XENHVM kernel. Booting from that results in  
the message: "pudna: fpcurthread == curthread xx times" scrolling by.  
What can I do to fix that?




I've not seen this message yet either. Can you provide more details of the  
Xen environment you're hosted on?


Is there a problem in continuing to run the 9.0 XENHVM kernel with the  
9.1 system?




This would not be advised. It *may* work, but be warned that you may run  
into strange issues.

___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: [Xen-devel] Fwd: xenbusb_nop_confighook_cb timeout and cd issue

[Mark Felder]

On Thu, 17 Jan 2013 09:56:33 -0600, Steven Chamberlain  
 wrote:



On 17/01/13 15:24, Mark Felder wrote:
I no longer have that environment that I was using to test the  
viability of upgrading to XCP 1.6 but it was reproducible every boot.


Just to clarify - you rebooted the dom0/hypervisor also?

I know the bug is reproducible on every FreeBSD domU boot, at least.



We did reboot the dom0 more than once but I didn't test that specifically  
to see if the problem would subside after a few reboots.

___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: Fwd: xenbusb_nop_confighook_cb timeout and cd issue

[Mark Felder]

For anyone running XCP 1.6 please try to use these RPMs which are patched.  
I'll be trying this myself later today.


http://downloads.xen.org/XCP/freebsd-xen-fix/


Cheers!
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: Fwd: xenbusb_nop_confighook_cb timeout and cd issue

[Mark Felder]

I no longer have that environment that I was using to test the viability of 
upgrading to XCP 1.6 but it was reproducible every boot.
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: Fwd: xenbusb_nop_confighook_cb timeout and cd issue

[Mark Felder]

Citrix has an open internal bug for this because of my incessant nagging.  
I'm going to direct them to this patch. If I can get their build manager  
to spin up a copy of XCP 1.6 with this patch included I'll make sure I get  
the ISO linked onto this mailing list.

___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: 9.1R?

[Mark Felder]

On Sun, 23 Dec 2012 18:36:54 +0900
moto kawasaki  wrote:

> It might be better to disable rxcsum, txcsum, and lro, too.
> My understanding is the point is the 'NIC hardware' is not the real
> one but emulation by Xen Host, so that those 'off loading' makes
> context switch storm between domU and dom0.
> # please kindly correct me if wrong.

You're actually quite correct as this is a source of network problems on many 
different domU OSes.
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: 9.1R?

[Mark Felder]

On Wed, 26 Dec 2012 10:17:25 +0900
moto kawasaki  wrote:

> Official Support of FreeBSD as domU.
> # Didn't someone talk about it on this ML ?? or my imagination ??

I've mentioned it. The domU tools scripts are being redesigned to be more 
platform agnostic and a guy on that team at Citrix has been having me assist 
him with figuring out how to pull data in the format he needs as there are new 
features coming to XenServer. It looks like Citrix has interest in FreeBSD 
because a client wants to do a large deployment, and NetBSD will be easy too. 
They want these tools to be in FreeBSD ports and NetBSD's pkgsrc, so it will be 
well integrated with the community and the way the OS already is maintained. It 
just sounds like Citrix will be pushing patches to the port maintainer in the 
future (unless they take it over themselves).

Overall, things are looking bright for FreeBSD/Xen users.
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: 9.1R?

[Mark Felder]

You're not running XCP or XenServer. That's why you don't have those bugs.
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: 9.1R?

[Mark Felder]

On Fri, 21 Dec 2012 12:26:36 -0600
Jay West  wrote:

> All our FreeBSD use is now under Xenserver 6.1. I was curious, given the
> port of xe-guest-utilities that allows online migrations and such. after
> installing those tools (thus PVM/64), should we be changing our mount points
> to refer to the xen hd device (xbdX) or continue to use adX inside the VM
> for filesystem mounts (kinda like we need to change the network device {ex.
> bgeX} to xnX)? Just looking for a sanity check J

You'll notice that /dev/xbd* doesn't actually exist :-)

Also, if you aren't aware, here is my personal list of Xen bugs:

- Shutdown via Xen fails to poweroff. It doesn't issue any normal shutdown 
commands,
  but tells the Xen code in the kernel to start a shutdown. It's broken on the 
kernel
  side. kern/171118 
  Also, NetBSD's workaround for this is to enable powerd, but we can't
  when we're running XENHVM and I suspect their powerd is a completely 
different beast.

- If you use PF with Xen HVM your performance will suck. Known bug.
  kern/154428
  Set net.inet.tcp.tso=0 as a workaround.

- "Deactivating" a network interface is an option after enabling the 
xe-guest-utilities scripts.
  Please don't do that. It seems to cause a panic every time. kern/171138
  Also, this happens when you try to change the VLAN assigned to an
  interface.
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: XENHVM ada -> ad ?

[Mark Felder]

On this topic, is there any way to *definitively* know which Xen block device 
corresponds to which adaX device? The guys at Citrix really want to know 
instead of just assuming that xbd.0 is ada0. Perhaps adding a sysctl value like 
"dev.xbd.0.%block: /dev/ada0" would be nice... Who would I talk to about this?

Hint: they're adding official tools support for FreeBSD soon.


fake edit: perhaps they can work with the "dev.xbd.0.%parent: xenbusb_front0" 
type output. I'll see...
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: xenbusb_nop_confighook_cb timeout

[Mark Felder]

Hi guys,

I've been in contact with Bob Ball at Citrix and he's passed on this
information after confirming the bug we're seeing. I was hoping
someone on the list could help track this down.

> I experienced the CDROM issue on XenServer 6.0 and 6.1.  I've seen 
> http://www.mail-archive.com/freebsd-xen@freebsd.org/msg01369.html which
> suggests that there was a fix in Xen 4.1.2, which is surprising since
> XenServer 6.1 uses Xen 4.1.3 and still exhibits the problem.
> http://xenbits.xen.org/gitweb/?p=qemu-xen-4.2-testing.git;a=commit;h=effd5676225761abdab90becac519716515c3be4
> refers to another regression from 4.1.2 to 4.1.3 relating to the CDROM,
> but it appears as though this regression was not merged into XenServer
> (i.e. the Xen 4.1.3 used in XenServer did not require this further patch).
> 
> http://xen.1045712.n5.nabble.com/Xen-4-1-1-HVM-guest-cdrom-trouble-lost-interrupts-ata-failed-commands-frozen-td4953147.html#a5028910
> gives one potential fix, which I believe is the one referenced in the
> release notes for 4.1 - however, I've applied this to my Xen and it
> does not appear to enable FreeBSD guests to boot with CDs attached.
> Could you confirm this from the other perspective and try the
> equivalent guest kernel patch?
> 
> diff --git a/arch/x86/pci/xen.c b/arch/x86/pci/xen.c
> index 8214724..6b57f90 100644
> --- a/arch/x86/pci/xen.c
> +++ b/arch/x86/pci/xen.c
> @@ -308,8 +308,7 @@ int __init pci_xen_init(void)
>  
>  int __init pci_xen_hvm_init(void)
>  {
> -   if (!xen_feature(XENFEAT_hvm_pirqs))
> -   return 0;
> +   return 0;
>  
>  #ifdef CONFIG_ACPI
> /*
> 
> This appears to be a patch for Linux, and I've not located the 
> equivalent file in FreeBSD (I'm assuming this is down to my 
> inexperience in FreeBSD!)
> 
> I believe that this probably means we have a problem in Xen with 
> booting FreeBSD guests that hasn't been tracked down, would you agree?
> Have either of you seen this working on later versions of OpenSource 
> Xen? (i.e. not XenServer?)  
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: xenbusb_nop_confighook_cb timeout

[Mark Felder]

On Wed, 10 Oct 2012 16:18:57 -0500, Gót András   
wrote:


The patched worked for me also. I hope somehow it'll make to into 9.1rc  
if send a reply to the previous PR.


Can we get a confirmation from any devs about whether or not this patch is  
the "right way" to fix it? If so, I'd also appreciate seeing this slip  
into 9.1 so the latest stable FreeBSD release works reliably with the  
latest stable XenServer/XCP release...

___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: xenbusb_nop_confighook_cb timeout

[Mark Felder]

This is also preventing my XCP 1.5beta to 1.6 testing :-(


Any suggestions are appreciated!
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: kern/171118: [xen] FreeBSD XENHVM guest doesn't shutdown cleanly

[Mark Felder]

The following reply was made to PR kern/171118; it has been noted by GNATS.

From: Mark Felder 
To: bug-follo...@freebsd.org
Cc:  
Subject: Re: kern/171118: [xen] FreeBSD XENHVM guest doesn't shutdown
 cleanly
Date: Mon, 8 Oct 2012 16:35:26 -0500

 NetBSD's documentation mentions a similar issue and their workaround is to  
 enable the NetBSD powerd daemon in rc.conf. Perhaps we can figure out what  
 their powerd daemon is doing so we can have a workaround? We can't run  
 powerd as a domU because it's looking for cpufreq settings that aren't in  
 sysctl.
 
 http://www.netbsd.org/ports/xen/howto.html
 
 > It is also desirable to add
 >  powerd=YES
 >
 > in rc.conf. This way, the domain will be properly shut down if xm  
 > shutdown -Ror xm shutdown -H is used on the domain0.
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: kern/164450: [xen] Failed to install FreeeBSD 9.0-RELEASE from CD in a Citrix XenServer VM

[Mark Felder]

The following reply was made to PR kern/164450; it has been noted by GNATS.

From: Mark Felder 
To: bug-follo...@freebsd.org, yeho...@gmail.com
Cc:  
Subject: Re: kern/164450: [xen] Failed to install FreeeBSD 9.0-RELEASE from CD
 in a Citrix XenServer VM
Date: Mon, 27 Aug 2012 06:34:21 -0500

 Was this Citrix XenServer 5.x? I've successfully done installs from a real  
 CD and an ISO on XenServer 6.x.
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: kern/162677: [xen] FreeBSD not compatible with "Current Stable Xen"

[Mark Felder]

The following reply was made to PR kern/162677; it has been noted by GNATS.

From: Mark Felder 
To: bug-follo...@freebsd.org, k...@ionvz.com
Cc:  
Subject: Re: kern/162677: [xen] FreeBSD not compatible with "Current
 Stable Xen"
Date: Mon, 27 Aug 2012 06:29:29 -0500

 This forum thread seems to indicate that it's mostly an Intel hardware  
 issue.
 
 This pfsense thread seems to have more details:
 
 http://forum.pfsense.org/index.php?topic=3294.15
 
 Xen must have worked around it because I can confirm that FreeBSD ISOs  
 boot just fine on Xen 4.2 on Intel hardware.
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: Need for XenServer tools for XENHVM Freebsd9

[Mark Felder]

On Thu, 23 Aug 2012 16:34:58 -0500, arvind viswanathan  
 wrote:


Do we have install XenServer tools to Freebsd XENHVM kernel. What  
advantage

does it offer. Even without the tools the xn interfaces come up.


The XenServer tools and additional scripts are needed if you are using  
XenServer or XCP as your hypervisor and wish to be able to use certain  
features such as memory ballooning, live migration, etc.

___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: kern/154428: [xen] xn0 network interface and PF - Massive performance drop

[Mark Felder]

The following reply was made to PR kern/154428; it has been noted by GNATS.

From: Mark Felder 
To: bug-follo...@freebsd.org
Cc: a...@ahhyes.net
Subject: Re: kern/154428: [xen] xn0 network interface and PF - Massive
 performance drop
Date: Thu, 23 Aug 2012 16:14:18 -0500

 I've hit this on 9.0-RELEASE as well using XCP 1.5beta as the hypervisor.
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

question about xen-tools versions

[Mark Felder]

Hi guys,

Right now sysutils/xen-tools is on version 4.0.1. The latest version  
available in the 4.0.x branch is 4.0.4 and there's also 4.1.3 available.  
Is anyone here close enough to the Xen project that they might know what  
the difference is or if it's safe to run 4.1.3 tools everywhere? If so,  
I'll get the port updated to 4.1.3.



Thanks!
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: Citrix Xenserver and FreeBSD migration/suspend scripts

[Mark Felder]

Hi guys, I've done some recent updates:

We're down to a resonable amount of scripts now:

%%LOCALBASE%%/etc/rc.d/xe-daemon
%%LOCALBASE%%/sbin/xe-daemon
%%LOCALBASE%%/sbin/xe-update-guest-attrs
%%LOCALBASE%%/sbin/xe_wrapper


Other changes:

- Cleaner memory reporting
- IPs for your network interfaces are reported correctly
- New version and build to match what the Linux tools report on XenServer  
6.0.2 which should make the "out of date" errors go away



Feel free to leave some more feedback. I expect this will be joining the  
xen-tools port very soon now.

___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: Citrix Xenserver and FreeBSD migration/suspend scripts

[Mark Felder]

On Tue, 14 Aug 2012 02:00:25 -0500, Egoitz Aurrekoetxea Aurre  
 wrote:


I suppose should be different ports... Xenstore is a Xen system  
capability but the Citrix rules for working with the xapi and xencenter  
are just for Citrix's Xen systems... You don't need them if you're  
virtualizing in a Gentoo dom0 for example... I'll create a different  
port for it in September/October... Which of course install by  
dependency xen-tools's port... But should not be the same port.


The xen-tools should include this work. It has nothing to do with what  
dom0 you're using -- if you don't have a use for this just don't enable it  
in rc.conf. We could even make it an optional KNOB in the port if  
necessary. There's just no sound reason why this should be a completely  
separate port.

___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: Citrix Xenserver and FreeBSD migration/suspend scripts

[Mark Felder]

Egoitz, I've included a link to my github repository several times now. I've 
taken your scripts and made many changes:

- no more BASH
- some scripts have been completely removed (integrated elsewhere)
- no more need for /etc/freebsd-version or temp files in /procfalse
- rc script is completely rewritten and compliant with freebsd standards

It is much more simplified and I expect this week I will get time to finish 
merging everything (except the daemon wrapper) into a single shell script. 
There's no need to have so many different scripts to accomplish this.


 https://github.com/felderado/freebsd-xen-tools-scripts


Please look at what I've done and feel free to contribute additional changes.
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: Citrix Xenserver and FreeBSD migration/suspend scripts

[Mark Felder]

On Sun, 12 Aug 2012 07:50:16 -0500, Jay West  wrote:


Questions:
1) on the stock 64bit 9.0 & 9.1-beta XENHVM kernels, we still had the  
boot
problem where after installing the kernel and rebooting, the boot failed  
as
being unable to find the boot device. We had to pass a boot string of  
ad0p2

instead of the default ada0p2, and then edit fstab accordingly. I assume
others here had to do that as well?



Yes, but I think that's because with the XENHVM kernel the disks are using  
a different driver. There's a difference between the ad and ada drivers.


2) I'm sure this will show a lack of understanding on my part,  but I'm  
just
curious about the xen-tools port that has been in the ports collection  
for a
little while. As per the pdf on the migration/suspend scripts, this  
needs to

be installed first. I would think the current port - being called
"xen-tools", would/should include these migration/suspend scripts. It's  
not
xen-tools (in Citrix parlance) without that capability. So, is this a  
case

of the new scripts just "finish off" the work that went into the current
xen-tools script and the xen-tools port in the ports collection is
functionally incomplete? I guess I'm unclear as to the
history/relationship/status of the xen-tools port and these
migration/suspend scripts.


This is likely where it will show up. I've been talking to the ports@ team  
and it seems to make sense that we include it in that port.




3) In some of my googling on 64 bit XENHVM "how-to's", one of them (can't
find it at the moment) mentioned 64 bit Freebsd not working right under  
Xen
with regards to SMP, and the how-to offered several patches to disable  
SMP
amongst other related items. To run the new migration/suspend scripts,  
did

other people here need to use those patches?



I have a feeling this is an old/resolved issue as I've not seen it yet.
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: Citrix Xenserver and FreeBSD migration/suspend scripts

[Mark Felder]

On Fri, 10 Aug 2012 03:31:12 -0500, Egoitz Aurrekoetxea  
 wrote:


Aug 10 10:31:28 xenserver xe_daemon[10567]:  
/usr/local/sbin/xe-update-guest-attrs: cannot open eth0 | 192.168.227.33
Aug 10 10:31:28 xenserver xe_daemon[10567]: eth1 | 192.168.227.161: No  
such file or directory
Aug 10 10:32:15 xenserver xe_daemon[126]:  
/usr/local/sbin/xe-update-guest-attrs: cannot open eth0 | 192.168.227.33
Aug 10 10:32:15 xenserver xe_daemon[126]: eth1 | 192.168.227.161: No  
such file or directory
Aug 10 10:32:28 xenserver xe_daemon[10567]:  
/usr/local/sbin/xe-update-guest-attrs: cannot open eth0 | 192.168.227.33
Aug 10 10:32:28 xenserver xe_daemon[10567]: eth1 | 192.168.227.161: No  
such file or directory
Aug 10 10:33:15 xenserver xe_daemon[126]:  
/usr/local/sbin/xe-update-guest-attrs: cannot open eth0 | 192.168.227.33
Aug 10 10:33:15 xenserver xe_daemon[126]: eth1 | 192.168.227.161: No  
such file or directory


Let me check this please but surely it would be something easy to solve  
or something that is not complete on the installation


He's using my modified version of the xen tools scripts and I just caught  
this bug this morning. It appears to be fixed here:


https://github.com/felderado/freebsd-xen-tools-scripts/commit/9250a10ae8a1ab1045f15964bd1cfce1124423bd#src/usr/local/sbin/xe-update-guest-attrs
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: Citrix Xenserver and FreeBSD migration/suspend scripts

[Mark Felder]

On Fri, 10 Aug 2012 08:56:30 -0500, Egoitz Aurrekoetxea  
 wrote:



was Xenserver 6.2.0 wasn't it??


No, it was 6.0.2. I haven't had a chance to figure this out as I don't  
have a 6.0.2 XenServer available right now.

___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: Citrix Xenserver and FreeBSD migration/suspend scripts

[Mark Felder]

On Thu, 09 Aug 2012 13:42:03 -0500, moto kawasaki   
wrote:



1) XenCenter shows warning as;
 "XenServer Tools out of date (version 6.0 installed)"
  Changing MicroVersion from 0 to 2 in xe-update-guest-attr file
   doesn't have effect on this issue.
xenstore_write_cached "attr/PVAddons/MicroVersion" "2"
^^^
   0 to 2


On XenServer 6.0.0 the current settings work fine. I imagine we need to  
investigate what version the tools are on other platforms and emulate that  
version accordingly.



2) "Shutdown" button on XenCenter works but doesn't complete.
   I guess XenServer/XenCenter expect something like "shutdown -p now"
   on the FreeBSD VM, but actually it runs like "shutdown -h now".
   Thus, FreeBSD VM has shut down, but doesn't cut its power, thus
   XenCenter remains on the way shutting down (VM icon stays in
   green, not goes to red.).


I haven't looked to see how this is handled at all, but I bet it's  
something we can fix.



3) Moving VM from one member host to another in the pool works.
   wow!!


Thanks for giving us another confirmation!
___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: Citrix Xenserver and FreeBSD migration/suspend scripts

[Mark Felder]

On Wed, 08 Aug 2012 09:41:26 -0500, Gót András   
wrote:



What would you need to host the tarballs?


If someone or some company has a vested interest in pushing this and  
developing it then by all means go ahead and take the current codebase and  
run with it. Otherwise I think what I'll end up doing is tagging releases  
in github and using that as the primary mirror and getting a copy of the  
tarballs onto the FreeBSD mirrors.

___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Re: Citrix Xenserver and FreeBSD migration/suspend scripts

[Mark Felder]

I've had a few positive responses off-list about this and the desire to  
get this into the ports tree somehow. Someone also had the original  
author's email address (he subscribes to this list), so I'm going to CC  
him on this email as well.


We could probably get this into the ports tree relatively quickly. I've  
been working closely with the ports@ team recently, so it wouldn't take  
much effort in that respect. Some questions were brought up about what it  
should be named or if it should be a separate port and I don't know what  
the right answer here is, really. Perhaps emulators/xe-tools or  
emulators/xenserver-tools would be sufficient.


I guess the first step would be to find some place to officially host the  
tarball and/or tag a release on github. As it stands things are fully  
functional, so getting this into the ports tree shouldn't be too big of a  
problem. Long term I'd like to see this cleaned up a bit more... it's  
quite the mess. Honestly, everything could (and should) be rolled into one  
shell script because there's really no need to have xe-daemon,  
xe-ip-if.sh, and xe-update-guest-attrs. For the record,  
xe-update-guest-attrs is the file with the Citrix/GPL header. It has the  
examples of how we should be calling the other xen-tools to report data to  
XenServer. Whether we concentrate on this script or a rewrite happens  
doesn't really matter to me; it's not a lot of code I'd just like to see  
all of these really awful hacks removed :-)


Someone was asking about whether or not this was production-worthy. I  
really don't see why not. I suppose if this fails you'll just end up with  
a VM running without recognized tools again. I don't believe it's possible  
for the OS to crash from things these scripts run, and the ability to  
migrate/suspend Xen VMs is not reliant on these tools in a fully open  
source Xen environment; this is just a requirement Citrix imposes.



Feel free to speak up if you have any thoughts or suggestions :-)


Egoitz, can you confirm what license you originally released everything  
under? I'd like to properly honor that license if you had one in mind.

___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"

Citrix Xenserver and FreeBSD migration/suspend scripts

[Mark Felder]

So about a month ago someone made some scripts public that seem to pass  
XenServer the data it needs to believe you're running fully supported Xen  
Tools on FreeBSD. I'm surprised this didn't make headlines. Perhaps it was  
just posted to the wrong audience. Anyway, with these scripts in place you  
can then do suspend and migrate which is something I've personally been  
dying to have access to. Without migration we've been stuck with VMWare so  
this gives us a glimmer of hope as far as a reasonable alternative goes.


Original post here:

http://lists.xen.org/archives/html/xen-api/2012-06/msg00137.html

"I can suspend, move, use ballooning, display ip, disk bandwitdth and  
so... from XenCenter. "


The wiki here:

http://wiki.xen.org/wiki/FreeBSD_9.0_64-bit_HVM_on_XCP_1.1

seems to indicate that the original author intended for this to be using  
an OpenBSD license. I can't actually see that listed anywhere publicly,  
and I wasn't on that list and haven't found his real email address yet to  
thank him for these scripts.


I took this, put it on github, and spent a little bit of time tonight  
trying to make sense of it and clean it up a bit. It could definitely use  
an overhaul in a few places and have some more cruft cleaned up, but it's  
better than nothing. I've successfully removed any dependency on BASH as  
well as made the rc script a proper rc script. Finally, we're at a point  
where you need no strange cache files (in /procfalse? what was that  
about?) or an /etc/freebsd-version file. It's fairly standalone now.


I've been testing this on a 9.0-RELEASE with a XEN-HVM kernel. I've tested  
suspend, but haven't had a chance to do a migrate yet. Memory usage is  
reporting correctly. I can't seem to see anything functionally wrong with  
this, but I'm sure it could use some love. Ideally I'd like to see this at  
a point where it's clean enough that we can add it into the  
sysutils/xen-tools port so everyone running Citrix XenServer has easy  
access to this capability.


So here you go:

https://github.com/felderado/freebsd-xen-tools-scripts


Please -- consume, fork, tinker, etc. I don't have any particularly  
outstanding scripting skills and have no reason to "own" this project.  
Perhaps we'll end up with a much cleaner and final version floating out  
there soon enough.

___
freebsd-xen@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-xen
To unsubscribe, send any mail to "freebsd-xen-unsubscr...@freebsd.org"