Netdata package is now available in XCP-ng
-
-
Hmm, I did not plan to update netdata before the next release of XCP-ng, because I don't trust their QA enough for that. However I can provide updated packages in the testing repo.
-
Testing would be fine when @r1 modification is done With enough community to test it, we can promote it into main then
-
@olivierlambert
After running the command "yum install netdata-ui" on 2 hosts. Only one started working.I can reach the netdata page on the working server at: url:19999. I've rebooted the newer of the 2 servers and no matter what I've tried the netdata page doesn't appear.
If I try the command again I receive an error indicating that the package is already installed and there is nothing to do.
Is there an easy way to fix this or, what's the command to rollback this install?
Thanks to everyone in the community (especially @olivierlambert) for your hard work on this software.
-
Double check your firewall on the "non working host". Also check if Netdata service is up
-
Imagine that I don't know what to do. I've tried reviewing the Netdata documentation but it isn't specific to XCP-NG.
I ran the below code and received a response. Does that indicate where the issue could be? Also...what code do you suggest I run to check to see if Netdata is running? Why can't it bind to the host address?
[09:49 xcp-ng3 ~]# /usr/sbin/netdata 2020-06-16 09:49:55: netdata INFO : MAIN : SIGNAL: Not enabling reaper 2020-06-16 09:49:55: netdata ERROR : MAIN : LISTENER: IPv4 bind() on ip '0.0.0.0' port 19999, socktype 1 failed. (errno 98, Address already in use) 2020-06-16 09:49:55: netdata ERROR : MAIN : LISTENER: Cannot bind to ip '0.0.0.0', port 19999 2020-06-16 09:49:55: netdata ERROR : MAIN : LISTENER: IPv6 bind() on ip '::' port 19999, socktype 1 failed. (errno 98, Address already in use) 2020-06-16 09:49:55: netdata ERROR : MAIN : LISTENER: Cannot bind to ip '::', port 19999 2020-06-16 09:49:55: netdata FATAL : MAIN :LISTENER: Cannot listen on any API socket. Exiting... # : Success 2020-06-16 09:49:55: netdata INFO : MAIN : EXIT: netdata prepares to exit with code 1... 2020-06-16 09:49:55: netdata INFO : MAIN : EXIT: cleaning up the database... 2020-06-16 09:49:55: netdata INFO : MAIN : Cleaning up database [0 hosts(s)]... 2020-06-16 09:49:55: netdata INFO : MAIN : EXIT: all done - netdata is now exiting - bye bye...
-
@PC_123 said in Netdata package is now available in XCP-ng:
Address already in use
It means there's already a service listening on this port. Try to stop it and see if it's still there on this port.
-
@PC_123 said in Netdata package is now available in XCP-ng:
Also...what code do you suggest I run to check to see if Netdata is running?
systemctl status netdata.service
-
@stormi
Green is a good.[10:06 xcp-ng3 ~]# systemctl status netdata.service ā netdata.service - Real time performance monitoring Loaded: loaded (/usr/lib/systemd/system/netdata.service; enabled; vendor pres et: disabled) Active: active (running) since Tue 2020-06-16 09:48:42 EDT; 18min ago Process: 1588 ExecStartPre=/usr/libexec/netdata/xcpng-iptables-restore.sh (cod e=exited, status=0/SUCCESS) Process: 1523 ExecStartPre=/bin/chown -R netdata:netdata /var/run/netdata (cod e=exited, status=0/SUCCESS) Process: 1470 ExecStartPre=/bin/mkdir -p /var/run/netdata (code=exited, status =0/SUCCESS) Process: 1404 ExecStartPre=/bin/chown -R netdata:netdata /var/cache/netdata (c ode=exited, status=0/SUCCESS) Process: 1359 ExecStartPre=/bin/mkdir -p /var/cache/netdata (code=exited, stat us=0/SUCCESS) Main PID: 1861 (netdata) CGroup: /system.slice/netdata.service āā1861 /usr/sbin/netdata -P /var/run/netdata/netdata.pid -D -W set... āā2035 /usr/libexec/netdata/plugins.d/apps.plugin 1 āā2058 /usr/libexec/netdata/plugins.d/go.d.plugin 1 āā2063 /usr/libexec/netdata/plugins.d/xenstat.plugin 1 āā2069 /usr/bin/python /usr/libexec/netdata/plugins.d/python.d.plu...
-
I ran identical code on two machines. One worked and the other didn't. The machine that didn't work is not the master of the pool. Could that be the reason? Do I need to setup the centralized reporting that was mentioned at the beginning of this thread?
-
If you installed
netdata-ui
, each instance of netdata is independant and has no knowledge of the pool setup. There's no obvious reason why it works on one and not on the other. Except if you installednetdata
only instead ofnetdata-ui
. -
What's the output of
iptables -L
on both hosts? -
@stormi
Not working host:[10:07 xcp-ng3 ~]# iptables -L Chain INPUT (policy ACCEPT) target prot opt source destination xapi_nbd_input_chain tcp -- anywhere anywhere tcp dpt:nbd ACCEPT gre -- anywhere anywhere RH-Firewall-1-INPUT all -- anywhere anywhere Chain FORWARD (policy ACCEPT) target prot opt source destination RH-Firewall-1-INPUT all -- anywhere anywhere Chain OUTPUT (policy ACCEPT) target prot opt source destination xapi_nbd_output_chain tcp -- anywhere anywhere tcp spt:nbd Chain RH-Firewall-1-INPUT (2 references) target prot opt source destination ACCEPT all -- anywhere anywhere ACCEPT icmp -- anywhere anywhere icmp any ACCEPT udp -- anywhere anywhere udp dpt:bootps ACCEPT all -- anywhere anywhere ctstate RELATED,ESTABLISHED ACCEPT udp -- anywhere anywhere ctstate NEW udp dpt:ha-cluster ACCEPT tcp -- anywhere anywhere ctstate NEW tcp dpt:ssh ACCEPT tcp -- anywhere anywhere ctstate NEW tcp dpt:http ACCEPT tcp -- anywhere anywhere ctstate NEW tcp dpt:https ACCEPT tcp -- anywhere anywhere tcp dpt:21064 ACCEPT udp -- anywhere anywhere multiport dports hpoms-dps-lstn,netsupport REJECT all -- anywhere anywhere reject-with icmp-host-prohibited Chain xapi_nbd_input_chain (1 references) target prot opt source destination REJECT all -- anywhere anywhere reject-with icmp-port-unreachable Chain xapi_nbd_output_chain (1 references) target prot opt source destination REJECT all -- anywhere anywhere reject-with icmp-port-unreachable
Working Host:
[10:28 xcp-ng2 ~]# iptables -L Chain INPUT (policy ACCEPT) target prot opt source destination xapi_nbd_input_chain tcp -- anywhere anywhere tcp dpt:nbd ACCEPT gre -- anywhere anywhere RH-Firewall-1-INPUT all -- anywhere anywhere Chain FORWARD (policy ACCEPT) target prot opt source destination RH-Firewall-1-INPUT all -- anywhere anywhere Chain OUTPUT (policy ACCEPT) target prot opt source destination xapi_nbd_output_chain tcp -- anywhere anywhere tcp spt:nbd Chain NETDATA (1 references) target prot opt source destination ACCEPT tcp -- anywhere anywhere ctstate NEW tcp dpt:dnp-sec Chain RH-Firewall-1-INPUT (2 references) target prot opt source destination ACCEPT all -- anywhere anywhere ACCEPT icmp -- anywhere anywhere icmp any ACCEPT udp -- anywhere anywhere udp dpt:bootps ACCEPT all -- anywhere anywhere ctstate RELATED,ESTABLISHED ACCEPT udp -- anywhere anywhere ctstate NEW udp dpt:ha-cluster ACCEPT tcp -- anywhere anywhere ctstate NEW tcp dpt:ssh ACCEPT tcp -- anywhere anywhere ctstate NEW tcp dpt:http ACCEPT tcp -- anywhere anywhere ctstate NEW tcp dpt:https ACCEPT tcp -- anywhere anywhere tcp dpt:21064 ACCEPT udp -- anywhere anywhere multiport dports hpoms-dps-lstn,netsupport NETDATA all -- anywhere anywhere REJECT all -- anywhere anywhere reject-with icmp-host-prohibited Chain xapi_nbd_input_chain (1 references) target prot opt source destination REJECT all -- anywhere anywhere reject-with icmp-port-unreachable Chain xapi_nbd_output_chain (1 references) target prot opt source destination REJECT all -- anywhere anywhere reject-with icmp-port-unreachable
-
The firewall rule for netdata is clearly missing here on the first host. I don't know why.
-
What's the output of:
rpm -qa | grep netdata
on both hosts? -
[10:28 xcp-ng3 ~]# rpm -qa | grep netdata netdata-1.19.0-3.xcpng8.1.x86_64 netdata-ui-1.19.0-3.xcpng8.1.x86_64
[10:29 xcp-ng2 ~]# rpm -qa | grep netdata netdata-1.19.0-3.xcpng8.1.x86_64 netdata-ui-1.19.0-3.xcpng8.1.x86_64
-
I'm glad you were able to spot a difference. My untrained eye still doesn't see the difference.
-
Can you run
/usr/libexec/netdata/xcpng-iptables-restore.sh
on the host where it's not working and share the output? -
That fixed it, thank you. Any idea why the firewall rule wasn't initially created?
[10:59 xcp-ng3 ~]# /usr/libexec/netdata/xcpng-iptables-restore.sh Applying firewall rules for netdata from /etc/sysconfig/iptables_netdata
-
No, I don't know. The iptables rules are rebuilt each time your host starts, and the additional rules for netdata are added when the netdata service starts.