OpenVZ Forum: Support » Compatibility: FC9

Home » General » Support » Compatibility: FC9

Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend

Compatibility: FC9 [message #30661]

Mon, 02 June 2008 12:10

sjdean
Messages: 30
Registered: May 2008

Member

Hi,

I really want to use OpenVZ. I've built a new server (big PC) using software RAID on a Logical Volume formatted to ext3.

AMD Athlon x64 3500, 2 Gig RAM.

I've installed Fedora Core 9 (FC9) x64.

Is OpenVZ compatible with this, or do I need to go down FC6? This post is here because I read of an issue with 64 bit OSs and not seeing anywhere where FC9 is reported as being OK.

Can 32 bit OS's be installed on a 64 bit host?

Does the Host kernel version need to match OpenVZ?

Additionally, in complaint of FC9, I tried making Xen work, but apparently support for dom0 doesn't work. So what's the point of this release?

Cya
Simon

Report message to a moderator

Re: Compatibility: FC9 [message #30679 is a reply to message #30661]

Mon, 02 June 2008 21:44

sjdean
Messages: 30
Registered: May 2008

Member

So FC9 doesn't allow me to use Xen due it not being able to hand dom0. KVM doesn't seem to like specifying an ISO file and reports "Could not find an installable distribution at XXXX".

By the time I managed to recompile kernel 2.6.18 after applying the OpenVZ patches (sumversion.c does not have #include <limits.h>, I restarted the box and now getting a Kernel Panic. Seems to get past the bit where it detects the onboard network card, then gives me:

"not syncing: attempted to kill init"

in relation to something along the lines of.... "netlink_update_listeners"

Went back to install FC6 i386, but that doesn't like my nForce 560 SATA Chipset and doesn't like the CD Rom!

OpenVZ just won't boot for me. What version of Fedora Core HAS it been tested with?

Report message to a moderator

Re: Compatibility: FC9 [message #30685 is a reply to message #30679]

Tue, 03 June 2008 06:49

sjdean
Messages: 30
Registered: May 2008

Member

I've added the development kernel 2.6.24 to the system, and it seems to work. I've downloaded the tar.gz for fed core 9 default.

But on the console, Im now getting various errors:

eg when I try to add an ip address:
"/etc/sysconfig/network-scripts/bak/ifcfg-venet0:0: no such file or directory"

Or when I try to check sshd status:

"/etc/init.d/sshd: cannot make pipe for command substitution: Cannot allocate memory"

This is quite driving me mad.

Never mind.

Report message to a moderator

Re: Compatibility: FC9 [message #31059 is a reply to message #30685]

Mon, 16 June 2008 22:41

TuPari
Messages: 2
Registered: June 2008

Junior Member

cat /proc/user_beancounters to see which parameters you have to increase to avoid running out of memory.

Report message to a moderator

Re: Compatibility: FC9 [message #31289 is a reply to message #30661]

Tue, 24 June 2008 05:44

dowdle
Messages: 261
Registered: December 2005
Location: Bozeman, Montana

Senior Member

I *REALLY* like Fedora as a desktop OS. I have been using Red Hat since 1996... and use Fedora on my personal desktops... BUT given the rapid nature of their release cycle (every 6 months), their lack of a supported/recommended upgrade path from one release to the next, and their short support cycle (2N +1)... I really CAN NOT recommend using Fedora for a host node OS for OpenVZ. I *REALLY* recommend that if you want to use a Red Hat-based distro for your OpenVZ host node you use RHEL or any RHEL-based distro like CentOS. Two of the three OpenVZ "stable" kernel trees are based on RHEL kernels (RHEL4 2.6.9 and RHEL5 2.6.18) so the stable OpenVZ kernels are actually designed for RHEL/CentOS.

If you want to try using Xen and OpenVZ together in the same kernel, there is also a combined kernel that you can download from the OpenVZ project website... although I don't recall if it has been tagged as "stable" or not... but it too is RHEL kernel based if I remember correctly.

If you want to try KVM and OpenVZ in the same kernel, I'd recommend you give Proxmox VE a try. It is based on a 2.6.24 kernel and has some issues with checkpointing/migration but it has been stable in my limited experience with it.

--
TYL, Scott Dowdle
Belgrade, Montana, USA

Report message to a moderator

Re: Compatibility: FC9 [message #32594 is a reply to message #31289]

Mon, 18 August 2008 17:13

neutrino38
Messages: 3
Registered: August 2008
Location: Grenoble, France

Junior Member

Well I went one step further:

Using cat /proc/user_beancounters I was able to determine that there is not enough dcachemem available.

I edited /etc/vz/conf/101.conf and increased the parameter.

Then I was able to start the VPS but the network was not started and when I start it using

/usr/sbin/vzctl exec 101 /etc/init.d/network start

I get some error messages:

[root@erp ~]# /etc/init.d/network restart
Shutting down interface venet0:                            [  OK  ]
Shutting down loopback interface:                          [  OK  ]
Cannot find device "sit0"
Bringing up loopback interface:  Error, some other host already uses address 127.0.0.1.
                                                           [FAILED]
Bringing up interface venet0:  SIOCADDRT: No such process
                                                           [  OK  ]
RTNETLINK answers: File exists
RTNETLINK answers: File exists
RTNETLINK answers: File exists
RTNETLINK answers: File exists
RTNETLINK answers: File exists
RTNETLINK answers: File exists
RTNETLINK answers: File exists
RTNETLINK answers: File exists
RTNETLINK answers: File exists

I am able to launch sshd then and connnect using SSH. But the internet access does not work although it works on the node. Here are a few more information:

[root@erp ~]# netstat -r
Kernel IP routing table
Destination     Gateway         Genmask         Flags   MSS Window  irtt Iface
192.0.2.0       *               255.255.255.0   U         0 0          0 venet0
link-local      *               255.255.0.0     U         0 0          0 venet0
172.21.0.0      *               255.255.0.0     U         0 0          0 venet0
default         192.0.2.1       0.0.0.0         UG        0 0          0 venet0


[root@erp ~]# /sbin/ifconfig -a
lo        Link encap:Local Loopback
          LOOPBACK  MTU:16436  Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:0 (0.0 b)  TX bytes:0 (0.0 b)

venet0    Link encap:UNSPEC  HWaddr 00-00-00-00-00-00-00-00-00-00-00-00-00-00-00-00
          inet addr:127.0.0.3  P-t-P:127.0.0.3  Bcast:0.0.0.0  Mask:255.255.255.255
          UP BROADCAST POINTOPOINT RUNNING NOARP  MTU:1500  Metric:1
          RX packets:884 errors:0 dropped:0 overruns:0 frame:0
          TX packets:754 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:68416 (66.8 KiB)  TX bytes:102968 (100.5 KiB)

venet0:0  Link encap:UNSPEC  HWaddr 00-00-00-00-00-00-00-00-00-00-00-00-00-00-00-00
          inet addr:172.21.100.7  P-t-P:172.21.100.7  Bcast:172.21.100.7  Mask:255.255.255.255
          UP BROADCAST POINTOPOINT RUNNING NOARP  MTU:1500  Metric:1

I do not understand why the default gateway and the routing tables are setup like this. Anyway, I fail to access the Internet from the VPS.

--
Emmanuel

Report message to a moderator

Re: Compatibility: FC9 - problem partially solved [message #32600 is a reply to message #32594]

Mon, 18 August 2008 20:46

neutrino38
Messages: 3
Registered: August 2008
Location: Grenoble, France

Junior Member

Ok,

After stopping IP tables on the VPS, I was able to access the outside network and all the rest despite all the ugly error message s associated with network startup.

But then something strange happens:

1/ I had to launch ssh manually in the VPS (using vzctl exec)

1/ yum hangs. When we interrupt it with CTRL-C, here is the backtrace generated

[root@erp ~]# yum install vim
Traceback (most recent call last):
  File "/usr/bin/yum", line 4, in <module>
    import yum
  File "/usr/lib/python2.5/site-packages/yum/__init__.py", line 40, in <module>
    import rpmsack
  File "/usr/lib/python2.5/site-packages/yum/rpmsack.py", line 22, in <module>
    import misc
  File "/usr/lib/python2.5/site-packages/yum/misc.py", line 13, in <module>
    import tempfile
  File "/usr/lib64/python2.5/tempfile.py", line 33, in <module>
    from random import Random as _Random
  File "/usr/lib64/python2.5/random.py", line 838, in <module>
    _inst = Random()
  File "/usr/lib64/python2.5/random.py", line 94, in __init__
    self.seed(x)
  File "/usr/lib64/python2.5/random.py", line 108, in seed
    a = long(_hexlify(_urandom(16)), 16)
  File "/usr/lib64/python2.5/os.py", line 736, in urandom
    bytes += read(_urandomfd, n - len(bytes))
KeyboardInterrupt

2/ rpm basic commands do not work. For instance rpm -q -a reports NOTHING !!! Trying install a RPM downloaded manually basically does NOTHING.

I am pretty confused.

--
Emmanuel

Report message to a moderator

Re: Compatibility: FC9 - solved [message #32601 is a reply to message #32600]

Mon, 18 August 2008 20:54

neutrino38
Messages: 3
Registered: August 2008
Location: Grenoble, France

Junior Member

Issue #2 resolved:

/dev/urandom was a regular file. Had to rerecate it:

# rm /dev/urandom
rm: remove regular empty file `/dev/urandom'? o
# mknod -m 0644 /dev/urandom c 1 9

It fixed the rpm issue too !!!!!!

But it it ... its working isn't it?

--
Emmanuel

Report message to a moderator

Previous Topic:	Any tool to monitor the intercommunication between VPS and HN?
Next Topic:	problem compiling Kernel RHEL5 028stab057.2

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Mon Jul 15 21:06:47 GMT 2024

Total time taken to generate the page: 0.02545 seconds