3PAR StoreServ Storage
cancel
Showing results for 
Search instead for 
Did you mean: 

HP 3PAR Degraded Mode

 
LElie
Occasional Advisor

HP 3PAR Degraded Mode

Hello,

we have 3PAR 7200 with 9 enclosures, including the controller. 

Controller is having two nodes, used to work Active passive before we had a failed MB for node1.

once replaced, the system cannot boot using two nodes, noting that the rescuemdoe has been done successfully for the new node1. now the system is working on one node, shownode from CLI shows only node0.

we did many reboot, but no success.

can we have any advices here? 

 

 

12 REPLIES 12
LElie
Occasional Advisor

Re: HP 3PAR Degraded Mode

Do we have to do any small configuration from the controller, to let it recognize the two nodes installed?

Torsten.
Acclaimed Contributor

Re: HP 3PAR Degraded Mode

Usually the noderescue should start automatically, but you can do

 

startnoderescue node 1

 

this will "re-install" node 1.


Hope this helps!
Regards
Torsten.

__________________________________________________
There are only 10 types of people in the world -
those who understand binary, and those who don't.

__________________________________________________
No support by private messages. Please ask the forum!

If you feel this was helpful please click the KUDOS! thumb below!   
LElie
Occasional Advisor

Re: HP 3PAR Degraded Mode

We did satrtrescuenode -node 1, but it didn't work also.

we did a lot of reboots, but the funny thing that both nodes are working perfectly if only one node is racked into the controller. they have totally the same information, same node number (node0). the challenge is whenever we boot the system with the two nodes, only the node 0 will be recognizable and working and no existing to node 1 (new node installed).

Torsten.
Acclaimed Contributor

Re: HP 3PAR Degraded Mode

I would connect a serial console to node 1 and watch it booting.


Hope this helps!
Regards
Torsten.

__________________________________________________
There are only 10 types of people in the world -
those who understand binary, and those who don't.

__________________________________________________
No support by private messages. Please ask the forum!

If you feel this was helpful please click the KUDOS! thumb below!   
LElie
Occasional Advisor

Re: HP 3PAR Degraded Mode

we did the serial console connection, we saw the node booting and it was 100% rescued.

the challenge is, that the rescued node is not joining the cluster, and it doesnt appear on the management conole or cli (shownode)

Torsten.
Acclaimed Contributor

Re: HP 3PAR Degraded Mode

Do you see the correct serialno. during boot? How does the prompt finally looks like?

Hope this helps!
Regards
Torsten.

__________________________________________________
There are only 10 types of people in the world -
those who understand binary, and those who don't.

__________________________________________________
No support by private messages. Please ask the forum!

If you feel this was helpful please click the KUDOS! thumb below!   
LElie
Occasional Advisor

Re: HP 3PAR Degraded Mode

it will look completely like the node 0.

as if copy paste from node 1 to node 0

Re: HP 3PAR Degraded Mode

Are you saying both nodes "think" they are node 0?


Note: While I am an HPE Employee, all of my comments (whether noted or not), are my own and are not any official representation of the company

Accept or Kudo

LElie
Occasional Advisor

Re: HP 3PAR Degraded Mode

yes, it will look completely like node0 if you put it in the controller.

system is not booting using two nodes, meanwhile, all the traffic is conducting by node 0.

 

Torsten.
Acclaimed Contributor

Re: HP 3PAR Degraded Mode

This part from the serial boot log would be interesting

(this one is taken from another model, but yours should look similar):

 

+-----------------------------------------------------------------------------+
HP SPI Image 05.02.54. Release version. 07:37:57 Mar 01 2017.
+-----------------------------------------------------------------------------+
| CPU 2 x 2.40 GHz Ivy Bridge deca core HT dual
| Control Cache Size 191.100 GB (cpu mem type 11) CL9
| Set0 CC DIMM0.0.0 (J180000): 32768 MB CL6/9
| Set0 CC DIMM0.1.0 (J190000): 32768 MB CL6/9
| Set0 CC DIMM0.2.0 (J200000): 32768 MB CL6/9
| Set1 CC DIMM1.0.0 (J180001): 32768 MB CL6/9
| Set1 CC DIMM1.1.0 (J190001): 32768 MB CL6/9
| Set1 CC DIMM1.2.0 (J200001): 32768 MB CL6/9
| Data Cache Size 256 GB CL6/9
| Set0 DC DIMM0.0.0 (J140005): 32768 MB CL6/9
| Set0 DC DIMM0.1.0 (J160005): 32768 MB CL6/9
| Set1 DC DIMM0.0.1 (J150005): 32768 MB CL6/9
| Set1 DC DIMM0.1.1 (J170005): 32768 MB CL6/9
| Set2 DC DIMM1.0.0 (J140006): 32768 MB CL6/9
| Set2 DC DIMM1.1.0 (J160006): 32768 MB CL6/9
| Set3 DC DIMM1.0.1 (J150006): 32768 MB CL6/9
| Set3 DC DIMM1.1.1 (J170006): 32768 MB CL6/9
| Slot ID 0 [8 Node HP 3PAR 20840X Centerplane]
| Power Supply 0 OK Delta
| Power Supply 1 OK Delta
| Battery Backup Unit OK Delta 800-200012.A1 00000232 (69)
| FPGA Chimera v1.0
| SATA0 Disk SanDisk 512 GB
| SATA1 Disk SanDisk 512 GB
| PCI Slot 0 LSI-SAS 9300-16e
| PCI Slot 1 LSI-SAS 9300-16e
| PCI Slot 2 LSI-SAS 9300-16e
| PCI Slot 3 HP 560SFP+ 5 2-port
| PCI Slot 4 Emulex LPe16004 4-port
| PCI Slot 5 Emulex LPe16004 4-port
| PCI Slot 6 Emulex LPe16004 4-port
| Board 920-200050.A2 FXN 2016/52
| Cluster serial CZ1234567         <<<======= IMPORTANT
| Board reset reason COLD_POWERON
| Current Time 2019-02-07 10:52:25 (UTC+1)
+-----------------------------------------------------------------------------+


Hope this helps!
Regards
Torsten.

__________________________________________________
There are only 10 types of people in the world -
those who understand binary, and those who don't.

__________________________________________________
No support by private messages. Please ask the forum!

If you feel this was helpful please click the KUDOS! thumb below!   
LElie
Occasional Advisor

Re: HP 3PAR Degraded Mode

Hello Torsten,

 

Do you mean that I should change the serial number for the new motherboard replaced so it can be detected by the cluster?

Please confirm and huge thanks for your replies.

Torsten.
Acclaimed Contributor

Re: HP 3PAR Degraded Mode

The cluster serial is the serial of the 3PAR itself and must be set on each node.
noderescue will do this for you, but you should check it.

not the same serial = not the same cluster

Hope this helps!
Regards
Torsten.

__________________________________________________
There are only 10 types of people in the world -
those who understand binary, and those who don't.

__________________________________________________
No support by private messages. Please ask the forum!

If you feel this was helpful please click the KUDOS! thumb below!