Help needed with HA in v6

Support
,
1

NethServer Version: 6.8
Module: nethserver-ha

Hi,

I’m facing problems configuring the HA setup in Nethserver v6, I have found some inconsistencies regarding the commands and after (fixing them?) then my HA cluster cannot communicate with the members.

I’m following this guide:
http://docs.nethserver.org/en/v6.8/ha.html

In the steps of pcs commands, this one is stating error:
/usr/sbin/pcs -f /tmp/mycluster resource group add mysqlha drbdFS VirtualIP sym_var_lib_mysql sym_etc_my.pwd sym_root_.my.cnf var_lib_nethserver_secrets mysqld

After closer inspection I see some resources were not created in the guide, and other resources are not grouped in the mysqlha group:
Not created:
sym_var_lib_mysql and var_lib_nethserver_secrets
Created but not included in group
sym_var_lib_asterisk

After removing the resources not created from the command and including the sym_var_lib_asterisk, the group was created but members could not communicate with each other, were giving some “authentication refused” so I went and run:

passwd hacluster (to change the passwd of the hacluster of corosync)
then
pcs cluster auth ns1 ns2

after that corosync became online:
PCSD Status:
ns1.localdomain: Online
ns2.localdomain: Online

But ns2 could not sync with ns1 and resources were 0 in ns2, resources in ns1 were 8, those defined in the guide.

ns2 from ns1 status was UNCLEAN(OFFLINE)

I tried many different ways to manually sync the members but I couldn’t even find the corosync.conf file in /etc/corosync, maybe another file is being used for nethserver ha setup.

Any help at this point would be wonderful or at least any other guide for HA setup.

PD: at which point is the /dev/vdb1 partition referenced by DRBD?? the guide points to /dev/drbd/by-res/drbd00 but that doesn’t exist neither

2

I guess that only @giacomo or @Stll0 can help you out on this

3

It’s hard to get the problem, I’ll try to give you some guidelines and a starting point

If symlink failed, probably drbd wasn’t created, isn’t working properly or isn’t mounted.
check in /var/log/messages if ther’s any error in nethserver-ha-save event. You should also check pacemaker logs.

DRBD is created by nethserver-ha-drbd action https://github.com/NethServer/nethserver-ha/blob/master/root/etc/e-smith/events/actions/nethserver-ha-drbd
this one could be useful http://www.drbd.org/en/doc/users-guide-84

To edit cluster configuration, use pcs console. This is The Documentation https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Linux/6/html/Configuring_the_Red_Hat_High_Availability_Add-On_with_Pacemaker/index.html

1 Like