Open PunkFleet opened 3 days ago
Hi @PunkFleet , can you please share the route tables of centos-s-2vcpu-2gb-nyc3-02
host?
Hi @PunkFleet , can you please share the route tables of
centos-s-2vcpu-2gb-nyc3-02
host?
Hi @nser77 , the host route tables in here
[root@centos-s-2vcpu-2gb-nyc3-02 ~]# ip route show
default via 138.197.96.1 dev eth0 proto static metric 100
10.17.0.0/16 dev eth0 proto kernel scope link src 10.17.0.7 metric 100
10.99.99.0/24 dev eth1 proto kernel scope link src 10.99.99.4 metric 101
138.197.96.0/20 dev eth0 proto kernel scope link src 138.197.99.141 metric 100
[root@centos-s-2vcpu-2gb-nyc3-02 ~]# ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
valid_lft forever preferred_lft forever
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000
link/ether 6a:23:c4:67:3e:ed brd ff:ff:ff:ff:ff:ff
altname enp0s3
altname ens3
inet 138.197.99.141/20 brd 138.197.111.255 scope global noprefixroute eth0
valid_lft forever preferred_lft forever
inet 10.17.0.7/16 brd 10.17.255.255 scope global noprefixroute eth0
valid_lft forever preferred_lft forever
inet6 fe80::6823:c4ff:fe67:3eed/64 scope link
valid_lft forever preferred_lft forever
3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000
link/ether 22:da:6e:1a:8f:81 brd ff:ff:ff:ff:ff:ff
altname enp0s4
altname ens4
inet 10.99.99.4/24 brd 10.99.99.255 scope global noprefixroute eth1
valid_lft forever preferred_lft forever
inet6 fe80::20da:6eff:fe1a:8f81/64 scope link
valid_lft forever preferred_lft forever
and there are my iptables information
[root@centos-s-2vcpu-2gb-nyc3-02 ~]# curl http://169.254.169.254/metadata/v1/interfaces/public/0/anchor_ipv4/address
10.17.0.7
[root@centos-s-2vcpu-2gb-nyc3-02 ~]# iptables -L -v -n
Chain INPUT (policy ACCEPT 19M packets, 3237M bytes)
pkts bytes target prot opt in out source destination
19M 3237M KUBE-FIREWALL 0 -- * * 0.0.0.0/0 0.0.0.0/0
Chain FORWARD (policy ACCEPT 0 packets, 0 bytes)
pkts bytes target prot opt in out source destination
Chain OUTPUT (policy ACCEPT 23M packets, 3165M bytes)
pkts bytes target prot opt in out source destination
23M 3165M KUBE-FIREWALL 0 -- * * 0.0.0.0/0 0.0.0.0/0
Chain KUBE-FIREWALL (2 references)
pkts bytes target prot opt in out source destination
0 0 DROP 0 -- * * !127.0.0.0/8 127.0.0.0/8 /* block incoming localnet connections */ ! ctstate RELATED,ESTABLISHED,DNAT
Chain KUBE-KUBELET-CANARY (0 references)
pkts bytes target prot opt in out source destination
There doesn't appear to be anything wrong with your keepalived configuration; your haproxy configuration is not a matter for this list.
You have shown above that keepalived is working correctly, since the IP address 10.99.99.99 is configured on eth1, so you are really going to have to diagnose the problem at your end.
Since 10.99.99.99 is on the same subnet as 10.99.99.4, the "No route to host" error message means that 10.99.99.4 cannot obtain a MAC address for 10.99.99.99 (if 10.99.99.4 could communicate with 10.99.99.99 but there were simply no service listening in port 5000, then you would get a "Connection refused" error message from nc.
I suggest you try using wireshark or tcpdump to trace what is happening on on both 10.99.99.4 and 10.99.99.2 to see if you can identify where things are going wrong.
the "No route to host" error message means that 10.99.99.4 cannot obtain a MAC address for 10.99.99.99
10.99.99.99 is VIP not real machine. then i how to config there MAC address? Im newbie in here, Is there any way I can fix this error?
From above, it shows:
3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000
link/ether 8e:99:21:72:51:83 brd ff:ff:ff:ff:ff:ff
altname enp0s4
altname ens4
inet 10.99.99.2/24 brd 10.99.99.255 scope global eth1
valid_lft forever preferred_lft forever
inet 10.99.99.99/32 scope global eth1
valid_lft forever preferred_lft forever
inet6 fe80::8c99:21ff:fe72:5183/64 scope link
valid_lft forever preferred_lft forever
The IP address 10.99.99.99 is configured on eth1, and therefore the MAC address associated with 10.99.99.99 is 8e:99:21:72:51:83.
Can you please try your nc command on 10.99.99.4 while keepalived is running and the address 10.99.99.99 is configured, and immediately afterwards execute arp -a
on 10.99.99.4 and post the results in this log.
Can you please try your nc command on 10.99.99.4 while keepalived is running and the address 10.99.99.99 is configured, and immediately afterwards execute
arp -a
on 10.99.99.4 and post the results in this log.
keepalived VIP is 10.99.99.99 and keepalived is running on LB1 machine
root@lb1:~# systemctl status keepalived
● keepalived.service - Keepalive Daemon (LVS and VRRP)
Loaded: loaded (/usr/lib/systemd/system/keepalived.service; enabled; preset: enabled)
Active: active (running) since Tue 2024-10-01 10:43:13 UTC; 16h ago
Docs: man:keepalived(8)
man:keepalived.conf(5)
man:genhash(1)
https://keepalived.org
Main PID: 47012 (keepalived)
Tasks: 3 (limit: 1112)
Memory: 2.2M (peak: 2.5M)
CPU: 14.558s
CGroup: /system.slice/keepalived.service
├─47012 /usr/sbin/keepalived --dont-fork
├─47013 /usr/sbin/keepalived --dont-fork
└─47014 /usr/sbin/keepalived --dont-fork
Oct 01 10:43:13 lb1 Keepalived_vrrp[47014]: (VI_1) received lower priority (100) advert from 10.99.99.3 - discarding
Oct 01 10:43:14 lb1 Keepalived_vrrp[47014]: (VI_1) received lower priority (100) advert from 10.99.99.3 - discarding
Oct 01 10:43:15 lb1 Keepalived_vrrp[47014]: (VI_1) received lower priority (100) advert from 10.99.99.3 - discarding
Oct 01 10:43:16 lb1 Keepalived_vrrp[47014]: (VI_1) received lower priority (100) advert from 10.99.99.3 - discarding
Oct 01 10:43:16 lb1 Keepalived_vrrp[47014]: (VI_1) Entering MASTER STATE
Oct 01 10:43:18 lb1 Keepalived_healthcheckers[47013]: TCP_CHECK on service [10.99.99.3]:tcp:8443 failed.
Oct 01 10:43:18 lb1 Keepalived_healthcheckers[47013]: Removing service [10.99.99.3]:tcp:8443 from VS [10.99.99.99]:tcp:8443
Oct 01 10:43:20 lb1 Keepalived_healthcheckers[47013]: TCP_CHECK on service [10.99.99.2]:tcp:8443 failed.
Oct 01 10:43:20 lb1 Keepalived_healthcheckers[47013]: Removing service [10.99.99.2]:tcp:8443 from VS [10.99.99.99]:tcp:8443
Oct 01 10:43:20 lb1 Keepalived_healthcheckers[47013]: Lost quorum 1-0=1 > 0 for VS [10.99.99.99]:tcp:8443
vrrp_instance VI_1 {
state MASTER
interface eth1
virtual_router_id 51
priority 101
advert_int 1
garp_master_delay 1
authentication {
auth_type PASS
auth_pass 1111
}
unicast_src_ip 10.99.99.2
unicast_peer {
10.99.99.3
}
virtual_ipaddress {
10.99.99.99
}
track_script {
check_apiserver
}
}
and on Master1 machine:
[root@centos-s-2vcpu-2gb-nyc3-02 ~]# nc -v 10.99.99.99 8443
Ncat: Version 7.92 ( https://nmap.org/ncat )
Ncat: No route to host.
[root@centos-s-2vcpu-2gb-nyc3-02 ~]# nc -v 10.99.99.99
Ncat: Version 7.92 ( https://nmap.org/ncat )
Ncat: No route to host.
[root@centos-s-2vcpu-2gb-nyc3-02 ~]# arp -a
? (10.99.99.7) at ce:52:d2:e7:8a:51 [ether] on eth1
? (10.17.0.6) at <incomplete> on eth0
? (10.99.99.3) at 4e:79:e6:8b:41:6c [ether] on eth1
? (10.99.99.6) at b6:05:98:06:51:15 [ether] on eth1
? (10.17.0.99) at <incomplete> on eth0
? (10.99.99.99) at <incomplete> on eth1
_gateway (138.197.96.1) at fe:00:00:00:01:01 [ether] on eth0
? (10.17.0.5) at <incomplete> on eth0
? (10.17.0.2) at <incomplete> on eth0
? (10.99.99.2) at 8e:99:21:72:51:83 [ether] on eth1
? (10.99.99.5) at 9e:f9:4e:cd:8e:eb [ether] on eth1
The arp -a output show that 10.99.99.4 can get an ARP response for 10.99.99.2 but not for 10.99.99.99. I am not sure of the relevance of the information before that arp -a output.
You need to investigate what is happening on your network to identify the cause of your problem; there is nothing we can do since we do not have access to your network.
The arp -a output show that 10.99.99.4 can get an ARP response for 10.99.99.2 but not for 10.99.99.99. I am not sure of the relevance of the information before that arp -a output.
You need to investigate what is happening on your network to identify the cause of your problem; there is nothing we can do since we do not have access to your network.
10.99.99.2 is LB1 machine private ip, 10.99.99.99 is KeepAlived VIP address on LB1 and LB2 Here's the problem, I can link the real private IP of the load balancer machine, but I can't link the VIP anyway
As I said above:
You need to investigate what is happening on your network to identify the cause of your problem; there is nothing we can do since we do not have access to your network.
I fllowoing this setting a LB for k8s cluster : https://github.com/kubernetes/kubeadm/blob/main/docs/ha-considerations.md#options-for-software-load-balancing but the cluster nodes can not connection LB VIP. i post my all config in here , i think my keepalived + haproxy config is correct, but erorr still exist. i doubt it's ISP limited machine VIP? https://stackoverflow.com/questions/79037704/how-to-build-keepalived-haproxy-load-balancer-on-digitalocean
haproxy.cfg
keepalived.conf
ip a is wokring