Home » Mailing lists » Users » OpenVZ on Fedora Core 5 (networking question)
OpenVZ on Fedora Core 5 (networking question) [message #2890] |
Tue, 25 April 2006 12:49 ![Go to next message Go to next message](/theme/ovz3/images/down.png) |
Aleksandar
Messages: 4 Registered: April 2006
|
Junior Member |
|
|
Hello,
after installing and successfully running one OpenVZ VPS I wasn't able
to ping or ssh to the VPS. After analyzing a little bit I came to the
following:
1) host venet0 isn't configured even after vzctl start 101
2) venet0 in the guest (vps) is set to 127.0.0.1
Is this normal ? I've assigned 10.0.0.1 to the VPS and if I manually
configure host's venet0 to 10.0.0.2 and guest's venet0 to 10.0.0.1
(through vzctl exec) I'm being able to ping and ssh to the vps.
Also, could someone point me to some detailed information about the
OpenVZ networking concept ? Is it like vmware (separate interfaces) ? Is
it like jail (ethernet aliases) ? The only thing I found about
networking was "vzctl set XYZ --ipadd A.B.C.D" :)
thanks,
Aleksandar
|
|
|
|
Re: OpenVZ on Fedora Core 5 (networking question) [message #2899 is a reply to message #2896] |
Tue, 25 April 2006 17:25 ![Go to previous message Go to previous message](/theme/ovz3/images/up.png) ![Go to next message Go to next message](/theme/ovz3/images/down.png) |
Aleksandar
Messages: 4 Registered: April 2006
|
Junior Member |
|
|
Hello Kiril,
just to mention that I've installed the vz*-3.0.0-7 packages and created
the vps with them. I couldn't start the yum update process so I tried to
get some help on irc. Someone pointed me to the wrong limits which were
for x86_64, I've corrected them and since then I was able to start
pretty much everything. Since the *-3.0.0-8 packages were released I've
updated to them and I used them to write this email.
regards,
Aleksandar
P.S. Don't forget to scroll down :-)
Kirill Korotaev wrote:
> Aleksandr,
>
> Can you please, provide the following information:
> 1. kernel version you use:
> # uname -a
>
Linux somehost.somedomain.net 2.6.16-1.2080_FC5.026test007 #1 Mon Apr 3
18:44:34 MSD 2006 i686 i686 i386 GNU/Linux
> 2. have you started vz service before VPS start?
> # /etc/init.d/vz restart
>
# service vz status
OpenVZ is running...
#service vz restart
Stopping OpenVZ: [OK]
Starting OpenVZ: [OK]
Bringing up interface venet0: [OK]
Configuring interface venet0: net.ipv4.conf.venet0.send_redirects = 0 [OK]
I've restarted it a lot of times
> 3. have you enabled ip forwarding?
> # cat /proc/sys/net/ipv4/ip_forward
>
# cat /proc/sys/net/ipv4/ip_forward
1
I actually don't need forwarding for the moment, because I want to be
able to ssh to the vps from the hardware node.
> 4. which template have you used for your VPS?
> # cat /etc/sysconfig/vz-scripts/101.conf
>
The fedora 5 template and I've updated it with yum through internet, I
didn't download the tarball.
=== BEGIN 101.conf ===
VERSION="2"
ONBOOT="no"
CLASSID="2"
# UBC parameters (in form of barrier:limit)
# Primary parameters
AVNUMPROC="40:40"
NUMPROC="65:65"
NUMTCPSOCK="80:80"
NUMOTHERSOCK="80:80"
VMGUARPAGES="6144:2147483647"
# Secondary parameters
KMEMSIZE="3578265:3816816"
TCPSNDBUF="319488:524288"
TCPRCVBUF="319488:524288"
OTHERSOCKBUF="132096:336896"
DGRAMRCVBUF="132096:132096"
OOMGUARPAGES="6144:2147483647"
# Auxiliary parameters
LOCKEDPAGES="32:32"
SHMPAGES="8192:8192"
PRIVVMPAGES="22528:24576"
NUMFILE="1280:1280"
NUMFLOCK="100:110"
NUMPTY="16:16"
NUMSIGINFO="256:256"
DCACHESIZE="1048576:1097728"
PHYSPAGES="0:2147483647"
NUMIPTENT="128:128"
# Disk quota parameters (in form of softlimit:hardlimit)
DISKSPACE="1048576:1153434"
DISKINODES="200000:220000"
QUOTATIME="0"
# CPU fair sheduler parameter
CPUUNITS="1000"
OFFLINE_MANAGEMENT="yes"
VE_ROOT="/vz/root/$VEID"
VE_PRIVATE="/vz/private/$VEID"
OSTEMPLATE="fedora-core-5-i386-default"
ORIGIN_SAMPLE="vps.basic"
HOSTNAME="www.rousse.org"
IP_ADDRESS="10.0.0.1"
NAMESERVER="XX.YY.154.19"
=== END 101.conf ===
> 5. from host and VPS:
> # ip a l
> # ip r l
>
Host:
# ip a l
2: lo: <LOOPBACK,UP> mtu 16436 qdisc noqueue
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
4: eth0: <BROADCAST,MULTICAST,UP> mtu 1500 qdisc pfifo_fast qlen 1000
link/ether 00:50:bf:d1:f5:84 brd ff:ff:ff:ff:ff:ff
inet XX.YY.154.19/27 brd XX.YY.154.31 scope global eth0
inet XX.YY.154.28/27 brd XX.YY.154.31 scope global secondary eth0:0
inet6 fe80::250:bfff:fed1:f584/64 scope link
valid_lft forever preferred_lft forever
6: sit0: <NOARP> mtu 1480 qdisc noop
link/sit 0.0.0.0 brd 0.0.0.0
5: venet0: <BROADCAST,POINTOPOINT,NOARP,UP> mtu 1500 qdisc noqueue
link/void
# ip r l
XX.YY.154.0/27 dev eth0 proto kernel scope link src XX.YY.154.19
169.254.0.0/16 dev eth0 scope link
default via XX.YY.154.1 dev eth0
Guest:
# vzctl start 101
Starting VPS ...
VPS is mounted
Adding IP address(es): 10.0.0.1
/bin/cp: preserving times for `/etc/sysconfig/network.2701': Function
not implemented
ERROR: Can't copy file /etc/sysconfig/network
Setting CPU units: 1000
Set hostname: www.rousse.org
/bin/cp: preserving times for `/etc/hosts.2710': Function not implemented
ERROR: Can't copy file /etc/hosts
VPS start in progress...
# vzctl exec 101 ip a l
1: lo: <LOOPBACK,UP> mtu 16436 qdisc noqueue
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
3: venet0: <BROADCAST,POINTOPOINT,NOARP,UP> mtu 1500 qdisc noqueue
link/void
inet 127.0.0.1/32 scope host venet0
# vzctl exec 101 ip r l
191.255.255.0/24 dev venet0 scope host
169.254.0.0/16 dev venet0 scope link
default via 191.255.255.1 dev venet0
> Thanks,
> Kirill
>
>> Hello,
>>
>> after installing and successfully running one OpenVZ VPS I wasn't
>> able to ping or ssh to the VPS. After analyzing a little bit I came
>> to the following:
>> 1) host venet0 isn't configured even after vzctl start 101
>> 2) venet0 in the guest (vps) is set to 127.0.0.1
>>
>> Is this normal ? I've assigned 10.0.0.1 to the VPS and if I manually
>> configure host's venet0 to 10.0.0.2 and guest's venet0 to 10.0.0.1
>> (through vzctl exec) I'm being able to ping and ssh to the vps.
>>
>> Also, could someone point me to some detailed information about the
>> OpenVZ networking concept ? Is it like vmware (separate interfaces) ?
>> Is it like jail (ethernet aliases) ? The only thing I found about
>> networking was "vzctl set XYZ --ipadd A.B.C.D" :)
>>
>> thanks,
>> Aleksandar
>>
>>
|
|
|
Re: OpenVZ on Fedora Core 5 (networking question) [message #2900 is a reply to message #2890] |
Tue, 25 April 2006 22:53 ![Go to previous message Go to previous message](/theme/ovz3/images/up.png) ![Go to next message Go to next message](/theme/ovz3/images/down.png) |
Santi
Messages: 7 Registered: April 2006
|
Junior Member |
|
|
El 25/04/2006, a las 14:49, Aleksandar escribió:
> Hello,
>
> after installing and successfully running one OpenVZ VPS I wasn't
> able to ping or ssh to the VPS. After analyzing a little bit I came
> to the following:
> 1) host venet0 isn't configured even after vzctl start 101
> 2) venet0 in the guest (vps) is set to 127.0.0.1
>
> Is this normal ? I've assigned 10.0.0.1 to the VPS and if I
> manually configure host's venet0 to 10.0.0.2 and guest's venet0 to
> 10.0.0.1 (through vzctl exec) I'm being able to ping and ssh to the
> vps.
>
> Also, could someone point me to some detailed information about the
> OpenVZ networking concept ? Is it like vmware (separate
> interfaces) ? Is it like jail (ethernet aliases) ? The only thing I
> found about networking was "vzctl set XYZ --ipadd A.B.C.D" :)
>
Hello,
I have similar problem with a freshly installed OpenVZ in a Debian
Sarge box, it's described at:
http://forum.openvz.org/index.php?t=tree&goto=2725
Addtional command outputs:
# uname -a
Linux midori 2.6.8-022stab076-smp #1 SMP Sun Apr 16 16:32:30 CEST
2006 i686 GNU/Linux
# cat /proc/sys/net/ipv4/ip_forward
1
# ip a l
2: lo: <LOOPBACK,UP> mtu 16436 qdisc noqueue
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
4: eth0: <BROADCAST,MULTICAST,UP> mtu 1500 qdisc pfifo_fast qlen 1000
link/ether 00:05:1c:19:67:f4 brd ff:ff:ff:ff:ff:ff
inet xxx/19 brd 255.255.255.255 scope global eth0
6: eth1: <BROADCAST,MULTICAST,UP> mtu 1500 qdisc pfifo_fast qlen 1000
link/ether 00:05:1c:03:26:36 brd ff:ff:ff:ff:ff:ff
inet 192.168.0.210/24 brd 192.168.0.255 scope global eth1
3: venet0: <BROADCAST,POINTOPOINT,NOARP,UP> mtu 1500 qdisc noqueue
link/void
# ip r l
192.168.0.5 dev venet0 scope link src 83.213.134.130
192.168.0.6 dev venet0 scope link src 83.213.134.130
192.168.0.0/24 dev eth1 proto kernel scope link src 192.168.0.210
83.213.128.0/19 dev eth0 proto kernel scope link src xxx
default via 83.213.128.1 dev eth0
Regards!
--
Santi Saez
|
|
|
Re: OpenVZ on Fedora Core 5 (networking question) [message #2905 is a reply to message #2899] |
Wed, 26 April 2006 07:35 ![Go to previous message Go to previous message](/theme/ovz3/images/up.png) ![Go to next message Go to next message](/theme/ovz3/images/down.png) |
dev
Messages: 1693 Registered: September 2005 Location: Moscow
|
Senior Member |
![5072133](/theme/ovz3/images/icq.png)
|
|
Aleksandr,
can you please apply workaround for FC5 cp problem as described here:
http://forum.openvz.org/index.php?t=tree&th=405&mid= 2355&&rev=&reveal=
and recheck the networking.
Thanks,
Kirill
> Hello Kiril,
>
> just to mention that I've installed the vz*-3.0.0-7 packages and created
> the vps with them. I couldn't start the yum update process so I tried to
> get some help on irc. Someone pointed me to the wrong limits which were
> for x86_64, I've corrected them and since then I was able to start
> pretty much everything. Since the *-3.0.0-8 packages were released I've
> updated to them and I used them to write this email.
>
> regards,
> Aleksandar
>
> P.S. Don't forget to scroll down :-)
>
> Kirill Korotaev wrote:
>
>> Aleksandr,
>>
>> Can you please, provide the following information:
>> 1. kernel version you use:
>> # uname -a
>>
> Linux somehost.somedomain.net 2.6.16-1.2080_FC5.026test007 #1 Mon Apr 3
> 18:44:34 MSD 2006 i686 i686 i386 GNU/Linux
>
>> 2. have you started vz service before VPS start?
>> # /etc/init.d/vz restart
>>
> # service vz status
> OpenVZ is running...
> #service vz restart
> Stopping OpenVZ: [OK]
> Starting OpenVZ: [OK]
> Bringing up interface venet0: [OK]
> Configuring interface venet0: net.ipv4.conf.venet0.send_redirects = 0 [OK]
>
> I've restarted it a lot of times
>
>> 3. have you enabled ip forwarding?
>> # cat /proc/sys/net/ipv4/ip_forward
>>
> # cat /proc/sys/net/ipv4/ip_forward
> 1
> I actually don't need forwarding for the moment, because I want to be
> able to ssh to the vps from the hardware node.
>
>> 4. which template have you used for your VPS?
>> # cat /etc/sysconfig/vz-scripts/101.conf
>>
> The fedora 5 template and I've updated it with yum through internet, I
> didn't download the tarball.
> === BEGIN 101.conf ===
> VERSION="2"
>
> ONBOOT="no"
> CLASSID="2"
>
> # UBC parameters (in form of barrier:limit)
> # Primary parameters
> AVNUMPROC="40:40"
> NUMPROC="65:65"
> NUMTCPSOCK="80:80"
> NUMOTHERSOCK="80:80"
> VMGUARPAGES="6144:2147483647"
> # Secondary parameters
> KMEMSIZE="3578265:3816816"
> TCPSNDBUF="319488:524288"
> TCPRCVBUF="319488:524288"
> OTHERSOCKBUF="132096:336896"
> DGRAMRCVBUF="132096:132096"
> OOMGUARPAGES="6144:2147483647"
> # Auxiliary parameters
> LOCKEDPAGES="32:32"
> SHMPAGES="8192:8192"
> PRIVVMPAGES="22528:24576"
> NUMFILE="1280:1280"
> NUMFLOCK="100:110"
> NUMPTY="16:16"
> NUMSIGINFO="256:256"
> DCACHESIZE="1048576:1097728"
>
> PHYSPAGES="0:2147483647"
> NUMIPTENT="128:128"
>
> # Disk quota parameters (in form of softlimit:hardlimit)
> DISKSPACE="1048576:1153434"
> DISKINODES="200000:220000"
> QUOTATIME="0"
>
> # CPU fair sheduler parameter
> CPUUNITS="1000"
> OFFLINE_MANAGEMENT="yes"
> VE_ROOT="/vz/root/$VEID"
> VE_PRIVATE="/vz/private/$VEID"
> OSTEMPLATE="fedora-core-5-i386-default"
> ORIGIN_SAMPLE="vps.basic"
> HOSTNAME="www.rousse.org"
> IP_ADDRESS="10.0.0.1"
> NAMESERVER="XX.YY.154.19"
> === END 101.conf ===
>
>> 5. from host and VPS:
>> # ip a l
>> # ip r l
>>
> Host:
> # ip a l
> 2: lo: <LOOPBACK,UP> mtu 16436 qdisc noqueue
> link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
> inet 127.0.0.1/8 scope host lo
> inet6 ::1/128 scope host
> valid_lft forever preferred_lft forever
> 4: eth0: <BROADCAST,MULTICAST,UP> mtu 1500 qdisc pfifo_fast qlen 1000
> link/ether 00:50:bf:d1:f5:84 brd ff:ff:ff:ff:ff:ff
> inet XX.YY.154.19/27 brd XX.YY.154.31 scope global eth0
> inet XX.YY.154.28/27 brd XX.YY.154.31 scope global secondary eth0:0
> inet6 fe80::250:bfff:fed1:f584/64 scope link
> valid_lft forever preferred_lft forever
> 6: sit0: <NOARP> mtu 1480 qdisc noop
> link/sit 0.0.0.0 brd 0.0.0.0
> 5: venet0: <BROADCAST,POINTOPOINT,NOARP,UP> mtu 1500 qdisc noqueue
> link/void
> # ip r l
> XX.YY.154.0/27 dev eth0 proto kernel scope link src XX.YY.154.19
> 169.254.0.0/16 dev eth0 scope link
> default via XX.YY.154.1 dev eth0
>
> Guest:
> # vzctl start 101
> Starting VPS ...
> VPS is mounted
> Adding IP address(es): 10.0.0.1
> /bin/cp: preserving times for `/etc/sysconfig/network.2701': Function
> not implemented
> ERROR: Can't copy file /etc/sysconfig/network
> Setting CPU units: 1000
> Set hostname: www.rousse.org
> /bin/cp: preserving times for `/etc/hosts.2710': Function not implemented
> ERROR: Can't copy file /etc/hosts
> VPS start in progress...
> # vzctl exec 101 ip a l
> 1: lo: <LOOPBACK,UP> mtu 16436 qdisc noqueue
> link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
> inet 127.0.0.1/8 scope host lo
> 3: venet0: <BROADCAST,POINTOPOINT,NOARP,UP> mtu 1500 qdisc noqueue
> link/void
> inet 127.0.0.1/32 scope host venet0
> # vzctl exec 101 ip r l
> 191.255.255.0/24 dev venet0 scope host
> 169.254.0.0/16 dev venet0 scope link
> default via 191.255.255.1 dev venet0
>
>> Thanks,
>> Kirill
>>
>>> Hello,
>>>
>>> after installing and successfully running one OpenVZ VPS I wasn't
>>> able to ping or ssh to the VPS. After analyzing a little bit I came
>>> to the following:
>>> 1) host venet0 isn't configured even after vzctl start 101
>>> 2) venet0 in the guest (vps) is set to 127.0.0.1
>>>
>>> Is this normal ? I've assigned 10.0.0.1 to the VPS and if I manually
>>> configure host's venet0 to 10.0.0.2 and guest's venet0 to 10.0.0.1
>>> (through vzctl exec) I'm being able to ping and ssh to the vps.
>>>
>>> Also, could someone point me to some detailed information about the
>>> OpenVZ networking concept ? Is it like vmware (separate interfaces) ?
>>> Is it like jail (ethernet aliases) ? The only thing I found about
>>> networking was "vzctl set XYZ --ipadd A.B.C.D" :)
>>>
>>> thanks,
>>> Aleksandar
>>>
>>>
|
|
|
Re: OpenVZ on Fedora Core 5 (networking question) [message #2907 is a reply to message #2900] |
Wed, 26 April 2006 16:26 ![Go to previous message Go to previous message](/theme/ovz3/images/up.png) ![Go to next message Go to next message](/theme/ovz3/images/down.png) |
dev
Messages: 1693 Registered: September 2005 Location: Moscow
|
Senior Member |
![5072133](/theme/ovz3/images/icq.png)
|
|
Santi,
As replied in private email to you, this helps:
midori:~# echo 1 > /proc/sys/net/ipv4/conf/eth1/proxy_arp
midori:~# echo 1 > /proc/sys/net/ipv4/conf/venet0/proxy_arp
midori:~# echo 1 > /proc/sys/net/ipv4/conf/all/proxy_arp
I also wonder what host system you have? I didn't find lots of vz
scripts in /etc/sysconfig, so I suppose your installation of OVZ is
somehow broken or not full?
The same proxy_arp settings should have been automatically setup by
vzctl... Strange...
I also replaced you MASQUARADE rules with:
# iptables -t nat -A POSTROUTING -s 192.168.0.0/24 -d ! 192.168.0.0/24
-j MASQUERADE
i.e. packets coming to 192.168.0.0/24 should not be masquaraded.
Thanks,
Kirill
> El 25/04/2006, a las 14:49, Aleksandar escribió:
>
>> Hello,
>>
>> after installing and successfully running one OpenVZ VPS I wasn't
>> able to ping or ssh to the VPS. After analyzing a little bit I came
>> to the following:
>> 1) host venet0 isn't configured even after vzctl start 101
>> 2) venet0 in the guest (vps) is set to 127.0.0.1
>>
>> Is this normal ? I've assigned 10.0.0.1 to the VPS and if I manually
>> configure host's venet0 to 10.0.0.2 and guest's venet0 to 10.0.0.1
>> (through vzctl exec) I'm being able to ping and ssh to the vps.
>>
>> Also, could someone point me to some detailed information about the
>> OpenVZ networking concept ? Is it like vmware (separate interfaces) ?
>> Is it like jail (ethernet aliases) ? The only thing I found about
>> networking was "vzctl set XYZ --ipadd A.B.C.D" :)
>>
>
> Hello,
>
> I have similar problem with a freshly installed OpenVZ in a Debian
> Sarge box, it's described at:
>
> http://forum.openvz.org/index.php?t=tree&goto=2725
>
> Addtional command outputs:
>
> # uname -a
> Linux midori 2.6.8-022stab076-smp #1 SMP Sun Apr 16 16:32:30 CEST 2006
> i686 GNU/Linux
>
> # cat /proc/sys/net/ipv4/ip_forward
> 1
>
> # ip a l
> 2: lo: <LOOPBACK,UP> mtu 16436 qdisc noqueue
> link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
> inet 127.0.0.1/8 scope host lo
> 4: eth0: <BROADCAST,MULTICAST,UP> mtu 1500 qdisc pfifo_fast qlen 1000
> link/ether 00:05:1c:19:67:f4 brd ff:ff:ff:ff:ff:ff
> inet xxx/19 brd 255.255.255.255 scope global eth0
> 6: eth1: <BROADCAST,MULTICAST,UP> mtu 1500 qdisc pfifo_fast qlen 1000
> link/ether 00:05:1c:03:26:36 brd ff:ff:ff:ff:ff:ff
> inet 192.168.0.210/24 brd 192.168.0.255 scope global eth1
> 3: venet0: <BROADCAST,POINTOPOINT,NOARP,UP> mtu 1500 qdisc noqueue
> link/void
>
> # ip r l
> 192.168.0.5 dev venet0 scope link src 83.213.134.130
> 192.168.0.6 dev venet0 scope link src 83.213.134.130
> 192.168.0.0/24 dev eth1 proto kernel scope link src 192.168.0.210
> 83.213.128.0/19 dev eth0 proto kernel scope link src xxx
> default via 83.213.128.1 dev eth0
>
> Regards!
>
> --
> Santi Saez
>
>
>
>
>
>
|
|
|
Re: OpenVZ on Fedora Core 5 (networking question) [message #2928 is a reply to message #2905] |
Fri, 28 April 2006 14:09 ![Go to previous message Go to previous message](/theme/ovz3/images/up.png) ![Go to next message Go to next message](/theme/ovz3/images/down.png) |
Aleksandar
Messages: 4 Registered: April 2006
|
Junior Member |
|
|
Hello Kiril,
it worked !!!
thanks,
Aleksandar
Kirill Korotaev wrote:
> Aleksandr,
>
> can you please apply workaround for FC5 cp problem as described here:
> http://forum.openvz.org/index.php?t=tree&th=405&mid= 2355&&rev=&reveal=
> and recheck the networking.
>
> Thanks,
> Kirill
>
>> Hello Kiril,
>>
>> just to mention that I've installed the vz*-3.0.0-7 packages and
>> created the vps with them. I couldn't start the yum update process so
>> I tried to get some help on irc. Someone pointed me to the wrong
>> limits which were for x86_64, I've corrected them and since then I
>> was able to start pretty much everything. Since the *-3.0.0-8
>> packages were released I've updated to them and I used them to write
>> this email.
>>
>> regards,
>> Aleksandar
>>
>> P.S. Don't forget to scroll down :-)
>>
>> Kirill Korotaev wrote:
>>
>>> Aleksandr,
>>>
>>> Can you please, provide the following information:
>>> 1. kernel version you use:
>>> # uname -a
>>>
>> Linux somehost.somedomain.net 2.6.16-1.2080_FC5.026test007 #1 Mon Apr
>> 3 18:44:34 MSD 2006 i686 i686 i386 GNU/Linux
>>
>>> 2. have you started vz service before VPS start?
>>> # /etc/init.d/vz restart
>>>
>> # service vz status
>> OpenVZ is running...
>> #service vz restart
>> Stopping OpenVZ: [OK]
>> Starting OpenVZ: [OK]
>> Bringing up interface venet0: [OK]
>> Configuring interface venet0: net.ipv4.conf.venet0.send_redirects = 0
>> [OK]
>>
>> I've restarted it a lot of times
>>
>>> 3. have you enabled ip forwarding?
>>> # cat /proc/sys/net/ipv4/ip_forward
>>>
>> # cat /proc/sys/net/ipv4/ip_forward
>> 1
>> I actually don't need forwarding for the moment, because I want to be
>> able to ssh to the vps from the hardware node.
>>
>>> 4. which template have you used for your VPS?
>>> # cat /etc/sysconfig/vz-scripts/101.conf
>>>
>> The fedora 5 template and I've updated it with yum through internet,
>> I didn't download the tarball.
>> === BEGIN 101.conf ===
>> VERSION="2"
>>
>> ONBOOT="no"
>> CLASSID="2"
>>
>> # UBC parameters (in form of barrier:limit)
>> # Primary parameters
>> AVNUMPROC="40:40"
>> NUMPROC="65:65"
>> NUMTCPSOCK="80:80"
>> NUMOTHERSOCK="80:80"
>> VMGUARPAGES="6144:2147483647"
>> # Secondary parameters
>> KMEMSIZE="3578265:3816816"
>> TCPSNDBUF="319488:524288"
>> TCPRCVBUF="319488:524288"
>> OTHERSOCKBUF="132096:336896"
>> DGRAMRCVBUF="132096:132096"
>> OOMGUARPAGES="6144:2147483647"
>> # Auxiliary parameters
>> LOCKEDPAGES="32:32"
>> SHMPAGES="8192:8192"
>> PRIVVMPAGES="22528:24576"
>> NUMFILE="1280:1280"
>> NUMFLOCK="100:110"
>> NUMPTY="16:16"
>> NUMSIGINFO="256:256"
>> DCACHESIZE="1048576:1097728"
>>
>> PHYSPAGES="0:2147483647"
>> NUMIPTENT="128:128"
>>
>> # Disk quota parameters (in form of softlimit:hardlimit)
>> DISKSPACE="1048576:1153434"
>> DISKINODES="200000:220000"
>> QUOTATIME="0"
>>
>> # CPU fair sheduler parameter
>> CPUUNITS="1000"
>> OFFLINE_MANAGEMENT="yes"
>> VE_ROOT="/vz/root/$VEID"
>> VE_PRIVATE="/vz/private/$VEID"
>> OSTEMPLATE="fedora-core-5-i386-default"
>> ORIGIN_SAMPLE="vps.basic"
>> HOSTNAME="www.rousse.org"
>> IP_ADDRESS="10.0.0.1"
>> NAMESERVER="XX.YY.154.19"
>> === END 101.conf ===
>>
>>> 5. from host and VPS:
>>> # ip a l
>>> # ip r l
>>>
>> Host:
>> # ip a l
>> 2: lo: <LOOPBACK,UP> mtu 16436 qdisc noqueue
>> link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
>> inet 127.0.0.1/8 scope host lo
>> inet6 ::1/128 scope host
>> valid_lft forever preferred_lft forever
>> 4: eth0: <BROADCAST,MULTICAST,UP> mtu 1500 qdisc pfifo_fast qlen 1000
>> link/ether 00:50:bf:d1:f5:84 brd ff:ff:ff:ff:ff:ff
>> inet XX.YY.154.19/27 brd XX.YY.154.31 scope global eth0
>> inet XX.YY.154.28/27 brd XX.YY.154.31 scope global secondary eth0:0
>> inet6 fe80::250:bfff:fed1:f584/64 scope link
>> valid_lft forever preferred_lft forever
>> 6: sit0: <NOARP> mtu 1480 qdisc noop
>> link/sit 0.0.0.0 brd 0.0.0.0
>> 5: venet0: <BROADCAST,POINTOPOINT,NOARP,UP> mtu 1500 qdisc noqueue
>> link/void
>> # ip r l
>> XX.YY.154.0/27 dev eth0 proto kernel scope link src XX.YY.154.19
>> 169.254.0.0/16 dev eth0 scope link
>> default via XX.YY.154.1 dev eth0
>>
>> Guest:
>> # vzctl start 101
>> Starting VPS ...
>> VPS is mounted
>> Adding IP address(es): 10.0.0.1
>> /bin/cp: preserving times for `/etc/sysconfig/network.2701': Function
>> not implemented
>> ERROR: Can't copy file /etc/sysconfig/network
>> Setting CPU units: 1000
>> Set hostname: www.rousse.org
>> /bin/cp: preserving times for `/etc/hosts.2710': Function not
>> implemented
>> ERROR: Can't copy file /etc/hosts
>> VPS start in progress...
>> # vzctl exec 101 ip a l
>> 1: lo: <LOOPBACK,UP> mtu 16436 qdisc noqueue
>> link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
>> inet 127.0.0.1/8 scope host lo
>> 3: venet0: <BROADCAST,POINTOPOINT,NOARP,UP> mtu 1500 qdisc noqueue
>> link/void
>> inet 127.0.0.1/32 scope host venet0
>> # vzctl exec 101 ip r l
>> 191.255.255.0/24 dev venet0 scope host
>> 169.254.0.0/16 dev venet0 scope link
>> default via 191.255.255.1 dev venet0
>>
>>> Thanks,
>>> Kirill
>>>
>>>> Hello,
>>>>
>>>> after installing and successfully running one OpenVZ VPS I wasn't
>>>> able to ping or ssh to the VPS. After analyzing a little bit I came
>>>> to the following:
>>>> 1) host venet0 isn't configured even after vzctl start 101
>>>> 2) venet0 in the guest (vps) is set to 127.0.0.1
>>>>
>>>> Is this normal ? I've assigned 10.0.0.1 to the VPS and if I
>>>> manually configure host's venet0 to 10.0.0.2 and guest's venet0 to
>>>> 10.0.0.1 (through vzctl exec) I'm being able to ping and ssh to the
>>>> vps.
>>>>
>>>> Also, could someone point me to some detailed information about the
>>>> OpenVZ networking concept ? Is it like vmware (separate interfaces)
>>>> ? Is it like jail (ethernet aliases) ? The only thing I found about
>>>> networking was "vzctl set XYZ --ipadd A.B.C.D" :)
>>>>
>>>> thanks,
>>>> Aleksandar
>>>>
>>>>
|
|
|
Re: OpenVZ on Fedora Core 5 (networking question) [message #2929 is a reply to message #2907] |
Fri, 28 April 2006 14:39 ![Go to previous message Go to previous message](/theme/ovz3/images/up.png) |
Santi
Messages: 7 Registered: April 2006
|
Junior Member |
|
|
El 26/04/2006, a las 18:33, Kirill Korotaev escribió:
Hello Kirill,
Sorry for the delay in the email reply.. :-(
Thanks, with those changes now I can ping VPS from the local network
and viceversa!
>
> I also wonder what host system you have? I didn't find lots of vz
> scripts in /etc/sysconfig, so I suppose your installation of OVZ is
> somehow broken or not full?
> The same proxy_arp settings should have been automatically setup by
> vzctl... Strange...
Hardware node it's a Debian Sarge "out of the box", and OpenVZ has
been installed following the steps of this howto:
http://www.howtoforge.com/openvz_debian_vps
I don't remember any error at installation time..
>
> I also replaced you MASQUARADE rules with:
> # iptables -t nat -A POSTROUTING -s 192.168.0.0/24 -d !
> 192.168.0.0/24 -j MASQUERADE
>
> i.e. packets coming to 192.168.0.0/24 should not be masquaraded.
Thanks again ;-)
Regards,
--
Santi Saez
|
|
|
Goto Forum:
Current Time: Wed Sep 11 22:15:48 GMT 2024
Total time taken to generate the page: 0.05554 seconds
|