Error in fedora quick start for Xen
by John Que
Hello,
I am following the instructions on Fedora Xen Quickstart page:
http://www.fedoraproject.org/wiki/FedoraXenQuickstart?highlight=%28xen%29
First thing, I want to say that the size for the fedora.img is
not enough now (I assume it was enough in the past)
because , after running (according to the instructions):
dd if=/dev/zero of=fedora.img bs=1M count=1 seek=1024
...
...
then running :
yum -c http://people.redhat.com/~katzj/yum-xen.conf --installroot=/mnt
-y groupinstall Base
gave me an error because of not enough space.
so I started the process whole over again from the beginning (and this took
more than an hour), this time allocating 2 giga:
dd if=/dev/zero of=fedora.img bs=1M count=1 seek=2048
So maybe this should be corrected in that web page.
I have a question:
Can the size of a file which was created thus
(dd if=/dev/zero of=fedora.img bs=1M count=1 seek=1024)
be changed ***without*** causing a damage to its contents?
In case there is such a way, how can it be done ?
Regards,
John
18 years
Puzzled by Xen
by Benjy Grogan
Hello:
I've read that Xen will allow servers that typically operate at 15% to
25% CPU capacity to operate at up to 80% with virtualization. Does
that mean that CPU usage will reach 80% instead of a lowly 15% to 25%?
Does CPU capacity equate CPU usage? Instead of putting more work on
different host OSes within Xen why not put all of that work in one
single OS to achieve that 80% CPU usage?
I've been reading up on Xen, and still this basic question puzzles me.
Is there something that I'm not getting about virtualization?
Benji
PS: To me CPU usage is when you look at the system monitor and see the load.
18 years
SIGSEGV on xenguest-install.py
by Chris Hall
Anybody seen this yet? I tried googling and got nothing. I run through the
xenguest-install.py while booted up to the xen0 kernel provided with FC5.
It asks all the normal questions, and begins the text based installation. I
get somewhere past network configuration (doesn't seem to matter what I put
here) and while it's trying to determine host and domain information, after
a few seconds to up to a minute or two later, it seg faults out with an
error "Loader Received SIGSEGV!", and shuts down the virtual machine in the
window. /var/log/xend-debug.log reveals the following:
File "/usr/sbin/xend", line 93, in main
return daemon.start()
File "/usr/lib/python2.4/site-packages/xen/xend/server/SrvDaemon.py", line
199, in start
self.run(w and os.fdopen(w, 'w') or None)
File "/usr/lib/python2.4/site-packages/xen/xend/server/SrvDaemon.py", line
301, in run
status.write('1')
ValueError: I/O operation on closed file
Any ideas what could be causing this? I've thus far tried removing and
reinstalling the xen kernel and the xen package. Didn't seem to work.
-Chris
18 years
xend (still) hanging on thinkpad T30 under fc5
by Brad Smith
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
I posted earlier about xend hanging when I tried to boot the xen0 kernel under FC5 on my Thinkpad
T30. I've tried again with a fresh install (last time it was an upgrade) and selinux disabled (last
time was set to permissive) and I'm still having the same problem: Booting into runlevel 3 with rhgb
off, the boot process hangs at "Starting xend" after a message about the bridging firewall. The
keyboard is still responsive so it's not the OS hanging, just xend. Attached is the xend.log
generated by attempting to boot (xend-debug.log is empty).
If anyone can offer assistance I'd really appreciate it.
- --Brad
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org
iD8DBQFELqRoDvp49DvQ8kcRAq7kAJ9yp59M+pUJbpKTwPx2Aoo2Z/nQOwCcCsK9
h90iuYpU1XSb0BxaZqjMCrk=
=btkn
-----END PGP SIGNATURE-----
18 years
Disk problems on FC5
by Brad Smith
Hi folks,
I set up Xen on a freshly-installed FC5 system, configured the guest using xenguest-install.py and
then began an installation via VNC. However, as soon as I got to the partitioning phase I received
an error that the partition table for my disk couldn't be read. I opted to ignore this as I've seen
that error even on real systems and have proceeded without incident. However, the installer
continued to produce errors about not being able to find the disk, being unable to read the
partition table, etc and eventually just died. Looking at the disk file I told Xen to create (I gave
it 4096 as the size) it appears to have a size of 0. Is that the way it should be? Any further
advice for troubleshooting?
Thanks,
--Brad
18 years
RE: [Fedora-xen] Problem booting jailtime images
by You, Yongkang
>-----Original Message-----
>From: Murat Işık [mailto:muratisik@bilgiguvenlik.com]
>Sent: 2006年4月1日 4:04
>To: fedora-xen(a)redhat.com
>Cc: You, Yongkang
>Subject: RE: [Fedora-xen] Problem booting jailtime images
>
>Hello,
>
>
>[root@muratisik fedora]# cat fedora.fc4.xen.cfg
>kernel = "/boot/vmlinuz-2.6.16-1.2080_FC5xen0"
I didn't try to use a xen0 kernel to boot xenU before. I am not sure if it is okay. But I think you can try vmlinuz-2.6.16-1.2080_FC5xenU which comes from another Xen RPMs.
>memory = 128
>name = "fedora.fc4"
>nics = 1
>dhcp = "dhcp"
>disk = ['file:/xen/fedora/fedora.fc4.img,sda1,w',
>'file:/xen/fedora/fedora.swap,sda2,w']
Although I think this config won't cause the error you meet, I suggest firstly let us try only use 1 disk for "/" and remove the swap disk config in here and fstab. If it can boot, then try to add swap back. :)
>root = "/dev/sda1 ro"
>
>[root@muratisik fedora]# mount -o loop fedora.fc4.img /mnt/
>
>[root@muratisik fedora]# cat /mnt/etc/fstab
># This file is edited by fstab-sync - see 'man fstab-sync' for details
>/dev/sda1 / ext3 defaults 1 1
>/dev/sda2 none swap sw 0 0
>none /dev/pts devpts
>gid=5,mode=620 0 0
>none /dev/shm tmpfs defaults 0 0
>none /proc proc defaults 0 0
>none /sys sysfs defaults 0 0
>
>[root@muratisik fedora]# umount /mnt/
>
>[root@muratisik fedora]# xm create fedora.fc4.xen.cfg -c
>
>and the error appears:
>
>...
>md: ... autorun DONE.
>VFS: Cannot open root device "sda1" or unknown-block(0,0)
>Please append a correct "root=" boot option
>Kernel panic - not syncing: VFS: Unable to mount root fs on
>unknown-block(0,0)
> [<c011c42c>] panic+0x3c/0x190
> [<c03c6a5c>] mount_block_root+0x19c/0x1aa
> [<c03c6ab3>] mount_root+0x49/0x4e
> [<c03c6b78>] prepare_namespace+0xc0/0xf1
> [<c0102267>] init+0x23e/0x2fe
> [<c0102029>] init+0x0/0x2fe
> [<c01029ad>] kernel_thread_helper+0x5/0xb
>
>
>Then I changed both image's fstab to hda1 and hda2 , also editted cfg to
>match hdas. and I get the same error. any ideas? is there any more
>information (log, config, debug vs.) I can supply?
Maybe you can try another thing. Is your image file's dev/ folder empty? If so, please mount the xenU image file, and copy your local /dev/* to image file's dev/ . My xenU dev/ has all device file prepared. ;)
>
>have a nice day.
>
Best Regards,
Yongkang (Kangkang) 永康
18 years
Problems with e1000 drivers with kernel 2.6.15...
by Thomas Uhl
Hi!
I am testing xen on a preproduction Intel box. The standard kernel of
FC5 (2.6.15-1.2054) does not detect the embedded e1000 NICs in my
setup. So I tried kernel 2.6.16-1.2102 from rawhide. It worked. When
will we see a "xenified" 2.6.16 kernel in rawhide?
Yours
Tom
Here the dmesg output with kernel 2.6.16...
Intel(R) PRO/1000 Network Driver - version 7.0.33-k2-NAPI
Copyright (c) 1999-2005 Intel Corporation.
ACPI: PCI Interrupt 0000:05:00.0[A] -> GSI 18 (level, low) -> IRQ 19
PCI: Setting latency timer of device 0000:05:00.0 to 64
e1000: 0000:05:00.0: e1000_probe: (PCI Express:2.5Gb/s:Width x4)
18 years
