On 02/22/2011 02:11 PM, Jan Kundrát wrote: > On 01/20/11 18:30, Troy Dawson wrote: >> kernel-2.6.18-238.1.1.el5 > > Hi list, > looks like I'm the only person having this problem. I've installed a > SL5.5 machine today as a fully virtualized guest in a SL5.2 Xen host > (kernel 2.6.18-194.32.1.el5xen). The installation went fine and the auto > updates brought in the 2.6.18-238.1.1.el5 kernel, which is where my > troubles started. > > This is the xen configuration I have: > > [root@ha3 ~]# cat /etc/xen/callisto > kernel = "/usr/lib/xen/boot/hvmloader" > builder='hvm' > memory = 512 > name = "callisto" > vcpus=1 > #vif = [ 'type=ioemu, bridge=xenbr0' ] > vif = [ 'mac=00:16:3e:e7:19:ce, bridge=xenbr0' ] > disk = [ 'phy:/dev/mapper/callisto,ioemu:hda,w' ] > on_reboot = 'restart' > on_crash = 'restart' > device_model = '/usr/lib64/xen/bin/qemu-dm' > boot='nc' > sdl=0 > vnc=1 > vncviewer=0 > stdvga=0 > serial='pty' > ne2000=0 > usb=1 > pae=1 > > For testing, I've replaced it with a very similar configuration, it just > uses tap:aio: as the backing store for the disk image and the vif= line > is commented out. The symptoms are, as far as I can tell, the same on > both configurations. I've also tried two different dom0 hosts (even > though with the same configuration), and there's no difference between them. > > Whenever I boot any kernel newer than 2.6.18-194.32.1.el5 (tried > 2.6.18-238.1.1.el5 from SL5.5 and 2.6.18-236, 2.6.18-245 from > http://people.redhat.com/jwilson/el5/), it hangs in one of these few places: > > Serial: 8250/16550 driver $Revision: 1.90 $ 4 ports, IRQ sharing enabled > [here] > serial8250: ttyS0 at I/O 0x3f8 (irq = 4) is a 16450 > [...] > PCI: Setting latency timer of device 0000:00:01.1 to 64 > ide0: BM-DMA at 0xc000-0xc007, BIOS settings: hda:pio, hdb:pio > ide1: BM-DMA at 0xc008-0xc00f, BIOS settings: hdc:pio, hdd:pio > [here] > Probing IDE interface ide0.. > [here] > hda: QEMU HARDDISK, ATA DISK drive > ide0 at 0x1f0-0x1f7,0x3f6 on irq 14 > > If it makes it past the last [here], it will always boot properly. > > I have absolutely no idea what's wrong here. I've tried commenting out > the "serial='pty'" line, that will make it not get stuck during the > serial initialization, but it still hangs before/after the "Probing IDE > interface ide0..". The behavior is same with all kernels newer than > 2.6.18-194.32.1.el5 that I tried. > > The only *working* kernel is the 2.6.18-238.1.1.el5xen, ie. a > Xen-specific version. This means running Xen dom0 inside HVM domU inside > dom0 :). > > According to the changelog message from 238, something changed in the > serial drivers realm: > > - [usb] serial: new driver MosChip MCS7840 (Stefan Assmann) [574507] > > ...but given that it's USB, it might not be related to my issue, This > one looks much more interesting, though: > > * Tue Oct 20 2009 Don Zickus<[log in to unmask]> [2.6.18-170.el5] > - [xen] allow booting with broken serial hardware (Chris Lalancette ) > [518338] > > Of course I can't access that report... > > SO I wonder if someone is running the same combination of kernels as I > do, ie. 2.6.18-194.32.1.el5 on dom0 and 2.6.18-238.1.1.el5 in HVM domU. > I'd be interested in hearing if that combination works for you. > > I'll try upgrading the dom0 to recent kernel tomorrow and keep you posted. > > With kind regards, > Jan > Hi Jan, Just because the client kernel is running 2.6.18-238.1.1.el5xen does not mean it is trying to be a xen host. It means that it is running paravirtualized. If your xen machine was setup to be a paravirtualized client, then it *has* to continue to run the xen kernel. You can't just switch from the one to the other (as far as I know). When you are running it on Gentoo, you probably set it up to not be paravirtualized, so it happily ran the regular kernel. If you are wondering, I did test the scenario you have. I currently have a xen host running 2.6.18-238.1.1.el5. Some of it's clients are/were running the older kernel, some 2.6.18-238.1.1.el5. All of them are working fine. Troy -- __________________________________________________ Troy Dawson [log in to unmask] (630)840-6468 Fermilab ComputingDivision/SCF/FEF/SLSMS Group __________________________________________________