Hi there,
We have a very strange issues on our new Xen platform, i cant seem to get everything up and running without performance issues. So far all our other xen servers work perfectly but these new ones act strange.
Server Configuration:
HP Proliant DL360 G2
2x Quad-core Xeon
32 GB 667Mhz Memory
2x 72GB SAS 10krpm disks
2x 300GB SAS 10rpm disks
2x Gbit Nic Onboard
cu1:~# uname -a
Linux cu1.wipa.nl 2.6.26-2-xen-amd64 #1 SMP Thu Nov 25 06:39:26 UTC 2010 x86_64 GNU/Linux
cu1:~#
eth0 Link encap:Ethernet HWaddr 00:26:55:7b:9a:88
inet addr:83.96.235.37 Bcast:83.96.235.127 Mask:255.255.255.128
inet6 addr: fe80::226:55ff:fe7b:9a88/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:558266 errors:0 dropped:0 overruns:0 frame:0
TX packets:47770 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:29222362 (27.8 MiB) TX bytes:5541863 (5.2 MiB)
eth1 Link encap:Ethernet HWaddr 00:26:55:7b:aa:ec
inet addr:172.50.20.150 Bcast:172.50.20.255 Mask:255.255.255.0
inet6 addr: fe80::226:55ff:fe7b:aaec/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:16 errors:0 dropped:0 overruns:0 frame:0
TX packets:6 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:1024 (1024.0 B) TX bytes:492 (492.0 B)
Interrupt:19 Memory:fa000000-fa012100
peth0 Link encap:Ethernet HWaddr 00:26:55:7b:9a:88
inet6 addr: fe80::226:55ff:fe7b:9a88/64 Scope:Link
UP BROADCAST RUNNING PROMISC MULTICAST MTU:1500 Metric:1
RX packets:1174819 errors:0 dropped:0 overruns:0 frame:0
TX packets:403716 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:360329375 (343.6 MiB) TX bytes:53501555 (51.0 MiB)
Interrupt:18 Memory:f8000000-f8012100
vif12.0 Link encap:Ethernet HWaddr fe:ff:ff:ff:ff:ff
inet6 addr: fe80::fcff:ffff:feff:ffff/64 Scope:Link
UP BROADCAST RUNNING PROMISC MULTICAST MTU:1500 Metric:1
RX packets:33426 errors:0 dropped:0 overruns:0 frame:0
TX packets:459074 errors:0 dropped:225 overruns:0 carrier:0
collisions:0 txqueuelen:32
RX bytes:1907670 (1.8 MiB) TX bytes:111555928 (106.3 MiB)
vif6.0 Link encap:Ethernet HWaddr fe:ff:ff:ff:ff:ff
inet6 addr: fe80::fcff:ffff:feff:ffff/64 Scope:Link
UP BROADCAST RUNNING PROMISC MULTICAST MTU:1500 Metric:1
RX packets:161967 errors:0 dropped:0 overruns:0 frame:0
TX packets:647936 errors:0 dropped:207 overruns:0 carrier:0
collisions:0 txqueuelen:32
RX bytes:18441838 (17.5 MiB) TX bytes:149516297 (142.5 MiB)
vif7.0 Link encap:Ethernet HWaddr fe:ff:ff:ff:ff:ff
inet6 addr: fe80::fcff:ffff:feff:ffff/64 Scope:Link
UP BROADCAST RUNNING PROMISC MULTICAST MTU:1500 Metric:1
RX packets:81477 errors:0 dropped:0 overruns:0 frame:0
TX packets:514458 errors:0 dropped:155 overruns:0 carrier:0
collisions:0 txqueuelen:32
RX bytes:8176067 (7.7 MiB) TX bytes:83209185 (79.3 MiB)
Chain FORWARD (policy ACCEPT)
target prot opt source destination
ACCEPT all -- ns1.itrends.eu anywhere PHYSDEV match --physdev-in vif6.0
ACCEPT udp -- anywhere anywhere PHYSDEV match --physdev-in vif6.0 udp spt:bootpc dpt:bootps
ACCEPT all -- ip-space.colo.itrends.eu anywhere PHYSDEV match --physdev-in vif7.0
ACCEPT udp -- anywhere anywhere PHYSDEV match --physdev-in vif7.0 udp spt:bootpc dpt:bootps
ACCEPT all -- web01.touch.itrends.eu anywhere PHYSDEV match --physdev-in vif12.0
ACCEPT udp -- anywhere anywhere PHYSDEV match --physdev-in vif12.0 udp spt:bootpc dpt:bootps
For now not all Iptables rules are here, i flushed it and rebooted the DomU. But the ones i rebooted you can see up here in the iptables, it sets the forwards.
cu1:/proc/sys/net/ipv4/conf/all# cat proxy_arp
1
cu1:/proc/sys/net/ipv4/conf/all#
cu1:/proc/sys/net/ipv4/conf/all# cat forwarding
1
cu1:/proc/sys/net/ipv4/conf/all#
Example host file:
#
# Kernel + memory size
#
kernel = '/boot/vmlinuz-2.6.26-2-xen-amd64'
ramdisk = '/boot/initrd.img-2.6.26-2-xen-amd64'
memory = '512'
extra = "console=tty xencons=tty"
#
# Disk device(s).
#
root = '/dev/sda2 ro'
disk = [
'phy:/dev/vg/ns1.xxxxx.eu-swap,sda1,w',
'phy:/dev/vg/ns1.xxxxx.eu-disk,sda2,w',
]
#
# Hostname
#
name = 'ns1. xxxxx.eu'
#
# Networking
#
vif = [ 'ip=XX.XX.XX.XX.9,mac=00:16:3E:F5:16:21' ]
#
# Behaviour
#
on_reboot = 'restart'
on_crash = 'restart'
--------
So far all well, it looks like everything has been setup correctly.
When i ping from the dom0 to the domU:
--- XX.XX.XX.9 ping statistics ---
19 packets transmitted, 12 received, 36% packet loss, time 18021ms
rtt min/avg/max/mdev = 0.032/0.181/1.794/0.486 ms
This means that it drops a lot of connections and this domU seriously can't function like this. I've never seen this on any of our other Xen servers so this one is really bugging me.
Anybody else had this problem before? The log is not showing anything (no errors or faults)
------------------------------------------------------------------------
Met vriendelijk groet, / Kind regards,
WIPA Nederland B.V.
Wieger Bontekoe
Telefoon: 036 - 522 3633 (Zeewolde)
Telefoon: 070 - 711 0260 (Den Haag)
------------------------------------------------------------------------
