Hi Paul,
On 03/13/2017 09:15 PM, Paul Cammarata wrote:
We have gluster 3.8.9 setup with 2 nodes, and a number of replicated
volumes. We are using Ganesha for HA. This weekend we had to reboot both
nodes and seem to have run into an issue. We first attempted to bring
them up one at a time but when only one was up the gluster shared
storage wouldn’t start and NFS-Ganesha was running but not setting up
the ha (no virtual IPs were being added). Once we had both of them up we
were able to get everything back up by doing the following.
1. gluster volume set all cluster.enable-shared-storage disable
2. gluster volume set all cluster.enable-shared-storage enable
3. gluster nfs-ganesha disable
4. gluster nfs-ganesha enable
These operations shouldn't be needed.
One concern is, what happens if we have a failure and are only able to
bring up one of the nodes? All our clients are pointed at the virtual
IP’s so they wont be able to connect without us manually adding them to
the interface. Is that really the way it should work?
That shouldn't be the case. In case of 2-node Ganesha/Pacemaker cluster,
if one node fails, other node should be hosting both the virtual-IPs.
Could you try setting below options and re-trigger your tests -
pcs property set stonith-enabled=false
pcs property set no-quorum-policy=ignore
Since you are using gluster-3.8.9, post reboot, only nfs-ganesha and HA
services need to be started (unless they are enabled to be started by
default via init/systemd)
#systemctl start pcsd
#systemctl start pacemaker
I suggest to enable above services to be started by default post reboot
#systemctl enable pcsd
#systemctl enable pacemaker
& start nfs-ganesha service
#systemctl start nfs-ganesha
The second thing is just doing a normal reboot. Should we have to
disable and enable those services when it comes back up? Do we need to
stop certain services before rebooting? Is there something in the setup
we may have done wrong?
Nothing in particular need to stopped prior to reboot. But post reboot,
please check if above mentioned services are up and running.
Thanks,
Soumya
*Paul Cammarata***
SIEM System Administrator
Secur*IT*360
530 Beacon Pkwy W, Suite 901 | Birmingham, AL 35209
*O:*205.419.9066 x1022 | *P:* 205.532.9646 | *F*: 205.449.1425
www.securit360.com <http://www.securit360.com/>| paul@xxxxxxxxxxxxxx
<mailto:paul@xxxxxxxxxxxxxx>
CONFIDENTIALITY: This email (including any attachments) may contain
confidential, proprietary and privileged information, and unauthorized
disclosure or use is prohibited. If you received this email in error,
please notify the sender and delete this email from your system. Thank you.
_______________________________________________
Gluster-users mailing list
Gluster-users@xxxxxxxxxxx
http://lists.gluster.org/mailman/listinfo/gluster-users
_______________________________________________
Gluster-users mailing list
Gluster-users@xxxxxxxxxxx
http://lists.gluster.org/mailman/listinfo/gluster-users
