Proxmox HA fencing and Dell iDrac7

[philwild]

Hi all,

This is not really a proxmox issue but I am hoping someone may offer some advice.

I have a proxmox cluster built but am having issues with fencing. The servers were supplied with idrac7 cards. These appear to differ slightly to idrac6 and fencing does not work

When a fault is detected, proxmox issues a reboot to the drac on the second node, but receives a failure message back. It then retries, over and over again. The server continually resets.


In testing outside of proxmox, I receive the following (see at bottom) when testing fencing... Is there any way round this or am I screwed?

Many thanks

Phil

root@gt1:/etc/default# fence_drac5 --ip="10.3.0.92" --username="root" --password="xxxx" --ssh --verbose --debug-file="/tmp/foo" --command-prompt="admin1->" --action="off"
Traceback (most recent call last):
File "/usr/sbin/fence_drac5", line 147, in <module>
main()
File "/usr/sbin/fence_drac5", line 130, in main
result = fence_action(conn, options, set_power_status, get_power_status, get_list_devices)
File "/usr/share/fence/fencing.py", line 837, in fence_action
if wait_power_status(tn, options, get_power_fn):
File "/usr/share/fence/fencing.py", line 752, in wait_power_status
if get_power_fn(tn, options) != options["-o"]:
File "/usr/sbin/fence_drac5", line 38, in get_power_status
status = re.compile("(^|: )(ON|OFF|Powering ON|Powering OFF)\s*$", re.IGNORECASE | re.MULTILINE).search(conn.before).group(2)
AttributeError: 'NoneType' object has no attribute 'group'

 

[dietmar]

Maybe fence_ipmilan works better?

 

[philwild]

Many thanks,
This is exactly what I did and all is working fine

 

[RCK]

I have a proxmox cluster built but am having issues with fencing. The servers were supplied with idrac7 cards. These appear to differ slightly to idrac6 and fencing does not work

When a fault is detected, proxmox issues a reboot to the drac on the second node, but receives a failure message back. It then retries, over and over again. The server continually resets.


In testing outside of proxmox, I receive the following (see at bottom) when testing fencing... Is there any way round this or am I screwed?

Many thanks

Phil

root@gt1:/etc/default# fence_drac5 --ip="10.3.0.92" --username="root" --password="xxxx" --ssh --verbose --debug-file="/tmp/foo" --command-prompt="admin1->" --action="off"

Hello Phil, I'm using fence_drac5 without any problem on PowerEdge R620 and iDRAC7 enterprise.
But you have to know it won't work with SSH, you will have to turn on the DRAC telnet service first. (moreover TELNET fencing is much more faster than SSH one)

/usr/sbin/fence_drac5 -o status -a ip_drac -l username -p password -c 'admin1->' -v

 

[curlingcanteen]

ssh works fine on iDRAC 7:

fence_drac5 --ip="10.1.6.7" --username="prox-a-drac" --password="***********************" --ssh --verbose --debug-file="/tmp/foo" --command-prompt="admin1->" --action="off"

BIOS ver. 2.1.3
Firmware: 1.56.55 (Build 05)

 

[francisco.]

I am experiencing problems with iDRAC7 too.

iDRAC 6:
root@host:# fence_idrac -a IP -l USER -p PASSWORD -o status
Status: ON

root@host:# fence_drac5 --ip="IP" --username="USER" --password="PASSWORD" --ssh --verbose --debug-file="/tmp/foo" --command-prompt="admin1->" --action="status"
Status: ON

root@host:# fence_ipmilan -a IP -l USER -p PASSWORD -o status
Status: ON

iDRAC7:
root@host:# fence_idrac -a IP2 -l USER -p PASSWORD -o status
ERROR:root:Failed: Unable to obtain correct plug status or plug is not available

root@host:# fence_drac5 --ip="IP2" --username="USER" --password="PASSWORD" --ssh --verbose --debug-file="/tmp/foo" --command-prompt="admin1->" --action="status"
Status: ON

root@host:# fence_ipmilan -a IP2 -l USER -p PASSWORD -o status
ERROR:root:Failed: Unable to obtain correct plug status or plug is not available