HPE HC250 / Apollo 2000 Smart Storage Battery Problem

[rm3na_ceo]

I've recently had an issue with a HC250 in which the smart storage battery failed, and as the component failed, all of the 4 nodes went on a degraded state. 

Changed the battery and everything came back fine. But the question came up: This is a SPoF for all 4 nodes.. is there a way to put in another battery pack? As far as i can tell, the blank supports another lithium battery, and even has a socket to plug it into the board:

There is no documentation about it. Ideas?

Thanks in advance 

[AleksandraIsa]

Hi @rm3na_ceo

there is only one battery in the Apollo 2000. 

http://h20628.www2.hp.com/km-ext/kmcsdirect/emr_na-c04604645-13.pdf page 75.

The 4 nodes communicate to it over the HDD backplane, so it's normal and expected that all of the 4 show the same status regarding the battery. 

[rm3na_ceo]

Sure @Aleksandralsa, that's what i meant about a single point of failure. In my environment, this failure meant that none of the nodes would work as expected. If a customer only has this 4 nodes, that would be pretty bad. At least there should be a second one, one for every two nodes.

[AleksandraIsa]

@rm3na_ceo, it's not truely single point of failure, because when the battery has failed, it does not lead to reboot or any other issues, other then the cache is temporary not working. There is only one battery by design:

https://h20195.www2.hpe.com/v2/Getdocument.aspx?docname=c04622598 page 3 

Storage Controller Cache 4 GB battery-backed write cache per node (one shared battery per system)

[rm3na_ceo]

Yep, but this made a customer upset, as most of their vms started behaving bad (slow or not responsive at all). As i stated earlier, after the bbatt exchange was made, everything went back to normal. (Fw updated first)