1753797 Members
7182 Online
108799 Solutions
New Discussion

VA7410 Array

 
Faizer
Advisor

VA7410 Array

Hi Gurus;
I have a va7410 with 1TB capacity. This array is configured as follows.

Redundancy G1: Logical drives: 205gb Unallocated: 151gb redundancy:110gb active spare: 66.75 total: 534gb
Redundancy G2: Logical drives: 335gb Unallocated: 7gb redundancy:91gb active spare: 33 total: 467gb


Based on the above is the undermentioned assumptions correct.
1. The array has 2 groups of disk namely RG1 and RG2.
2. The active spare and the redundancy works on an algorithm that is factory set for the Array.
3. In the event of a disk failures then sequence of events is as follows
First the active spare disk in the respective Redundancy group is used till it becomes zero.
Then the Redundancy disk space is used up till it becomes zero before the respective redundancy group crashes/ is unavailable

Also let me know the high availability of this array.
Thanks is advance
Faizer
1 REPLY 1
Tim D Fulford
Honored Contributor

Re: VA7410 Array

Hi

I'm not sure I understand your qustions...

1. The array has 2 groups of disk namely RG1 and RG2
> Yep, that is the way it was designed


2. The active spare and the redundancy works on an algorithm that is factory set for the Array.
> Of course how else would it work? You either choose to have hot spares or not, so there is no choice about what algorythem is used. The only choices you have are
o RAID1+0
o RAID5DP (re-marketed RAID6 !!! ;-o )
o AutoRAID (a combination of the above)
The hot spare activity will vary depending on yopur choice

3. In the event of a disk failures then sequence of events is as follows
First the active spare disk in the respective Redundancy group is used till it becomes zero.
Then the Redundancy disk space is used up till it becomes zero before the respective redundancy group crashes/ is unavailable
> errr nope. The hot spare is actually a mis-anoma (or however you spell it) there is no single physical disk taht is awaiting activation. Wht it is is space RESEVEED on each disk such that one (or however many you choose) disk is/are free. You may also remember is that it is allocated as "LargestTwoDisks", so if you have a mixed set of disks it will reserve space for the larets two disks.

The array has a 3 tools it uses.
o Firstly it tries to find data that is one failure away from data loss & it will replicate this first, so any RAID1 data is replicated onto the spare space first at a high replication priority.
o Secondly it replicates the remaining data. so RAID5DP stuff is done last at a lower replication priority.
o Latsly IF there is not enough space to replicate the data the VA will try to convert any RAID1+0 into RAID5DP to free up the space required to protect data. Hence it is a bad idea to use solely RAID1+0 without any hot spares.

Also let me know the high availability of this array.
> It is highly available. It DOES have a single point of failure (so I'm told) with the communication between the array controllers (I've not seen this fail). I've seen figure of 99.95%, from personal experience I'd say more like 95-97%. If you want 99.95% (or more) use two arrays mirrored using say LVM of Veritas.

Most of my experience has been with its older brother VA7400, and the firmware & SDM version need to be AT LEAST HP18 & 1.06 respectively as when things go wrong the first thing you will need to do is fix these & hope data is not destroyed i the process. the VA7410 needs A100 & 1.06.

My MAJOUR HA concern; which brings down the HA figure in my opionion; is that it is billed taht you can do the firmware upgrades "on-line", i.e. take out one controller upgrade it, let it re-join & upgrade the other controller. This option has NEVER been available and the firmware upgrades are always ('till now at least) Off-line activity.

Regards

Tim

-