TOPIC:

Hi,
I had installed HA-ISCSI like is in Documentations but unfortunately it's not working as it should.

I have two identical servers with Citrix Xen 6.2, there are in a POOL.

HOST ROLE: MASTER
DRBD ROLE: iscsi1=Primary
DRBD CONNECTION: iscsi1 in SyncSource state
ISCSI TARGET: tgtd (pid 11904 11903) is running...
VIRTUAL IP: 10.10.10.3 is local

HOST ROLE: SLAVE
VIRTUAL IP: 10.10.10.3 is not local
ISCSI TARGET: tgtd is stopped
DRBD ROLE: iscsi1=Secondary
DRBD CONNECTION: iscsi1 in WFConnection state


ha-lizard is enabled

When I turn off the MASTER nothing happens.
I got infromations like:

INFO-/etc/iscsi-ha/iscsi-ha.sh: Pool Master xxx.xxx.xxx.xxx not responding - replug_pbd exiting

iSCSI target: /etc/init.d/tgtd status stopped. Expected Stopped . [tgtd is stopped]

I can't connect to POOL from XenCenter. Log shows in circle the same states, even after a while SLAVE not become a MASTER, tgtd is stopped etc




My ha-lizard.pool.conf

DISABLED_VAPPS=()
ENABLE_LOGGING=1
FENCE_ACTION=stop
FENCE_ENABLED=1
FENCE_FILE_LOC=/etc/ha-lizard/fence
FENCE_HA_ONFAIL=1
FENCE_HEURISTICS_IPS=8.8.8.8
FENCE_HOST_FORGET=0
FENCE_IPADDRESS=
FENCE_METHOD=POOL
FENCE_MIN_HOSTS=2
FENCE_PASSWD=
FENCE_QUORUM_REQUIRED=1
FENCE_REBOOT_LONE_HOST=0
FENCE_USE_IP_HEURISTICS=1
GLOBAL_VM_HA=1
MAIL_FROM="root@localhost"
MAIL_ON=1
MAIL_SUBJECT="SYSTEM_ALERT-FROM_HOST:$HOSTNAME"
MAIL_TO=addres@addres.com
MONITOR_DELAY=15
MONITOR_KILLALL=1
MONITOR_MAX_STARTS=20
MONITOR_SCANRATE=10
OP_MODE=2
PROMOTE_SLAVE=1
SLAVE_HA=1
SLAVE_VM_STAT=0
XAPI_COUNT=2
XAPI_DELAY=10
XC_FIELD_NAME='ha-lizard-enabled'
XE_TIMEOUT=10


my iscsi-ha.conf

DRBD_RESOURCES=iscsi1
ISCSI_TARGET_SERVICE=/etc/init.d/tgtd
DRBD_VIRTUAL_IP=10.10.10.3
DRBD_VIRTUAL_MASK=255.255.255.0
DRBD_INTERFACE=xenbr1
MONITOR_MAX_STARTS=5
MONITOR_DELAY=10
MONITOR_KILLALL=1
MONITOR_SCANRATE=5
ENABLE_LOGGING=1
MAIL_ON=0
MAIL_SUBJECT="HA-Lizard noSAN SYSTEM ALERT - FROM HOST: $HOSTNAME"
MAIL_FROM="root@localhost"
MAIL_TO="This email address is being protected from spambots. You need JavaScript enabled to view it."
#MAIL_SCREEN_TIME=30
#END CONFIG FILE

Please tell me where can be the bug.
I will be grateful

Please post a complete log from the slave when you turn off the master.

On the slave, type "ha-cfg log" before turning off the master. Make sure to send the entire log.
thank you

logs from slave during when master is off

The attached log is from the iscsi-ha component. We need to see the ha-lizard log. - please send the ha-lizard log. you can capture that with: "ha-cfg log"

Thanks

My mistake, sorry
I send You logs from slave, from ha-cfg log command output

Thanks

Your slave host has HA suspended. This is likely resultant from your testing and occurs to prevent a potential split brain situation. Meaning, if network connectivity to the Master is lost AND quorum could not be achieved (by using IP heuristics) then the slave will enter into a suspended HA mode until you fix the errors and then re-enable HA on the slave.

You can see this in the following lines of the log file:

This host has self fenced - HA forcefully disabled. Exiting...
HA can be reactiveted for this host by running /etc/ha-lizard/scripts/recover_fenced_host

As an FYI - you have left the heuristic IP as the default 8.8.8.8. You should ideally make this something local to the host so that HA is not dependant on public internet access.

Checking HA status of the slave would also indicate that the host has HA suspended (try "ha-cfg status"). Run "/etc/ha-lizard/scripts/recover_fenced_host" to remove the suspension.

Please post your testing results when done for the benefit of other users.