I decided to create this topic following a discussion in another forum section, more specifically here.
First I'll talk briefly about the issue that I am facing and what I perceive to be the origin of the problem.
I recently had OMSA (v18.104.22.168) alert me to a degraded array (RAID-5) due to a failed disk on a PowerEdge T310 server (the RAID controller is a PERC S100). About an hour before powering down and replacing the disk I decided to upgrade OMSA to the latest version (v8.5), which I did by upgrading the old version. At that point I don't remember clearly if the issue that I would notice later on was already present, but it became obvious after swapping the failed disk. The issue that I'm talking about is an incorrect disk capacity presented by OMSA v8.5 - OMSA alone shows this, not the OS and not the controller's BIOS config menu. Apart from this there is no indication by OMSA of anything remotely wrong with the array.
To my surprise, upon booting and starting the rebuild process, OMSA was showing me a capacity of 8,187.84GB instead of the correct 930.48GB (3x 500GB HDDs). This incorrect value persisted across reboots, uninstalls, reinstalls of OMSA. After experimenting with several OMSA versions, with each one giving it's own highly exaggerated disk capacity; in some cases the same OMSA version would even report multiple capacities by just restarting the Data Manager service, for example:56,098,816.00GB55,574,528.00GB39,059,456.00GB57,671,680.00GB57,147,392.00GB52,166,656.00GB18,874,368.00GB
The most amusing reading would come from v8.4: 7,234,142,328.78GB
Anyway, as I was explaining, I found that the disk capacity was only reported incorrectly in OMSA v8.2 and later. The last accurate reading of this setting came from v8.1. After further testing I isolated this correct/incorrect reporting to a DLL from the Dell Storage Management (SysMgt\sm\dellvl\dsm_sm_swrvil.dll) - a different DLL would probably be used for different RAID controllers' series.
In conclusion, OMSA v8.5 can indeed report the correct disk capacity in this particular system but only if it uses Storage Management v22.214.171.124 (dsm_sm_swrvil.dll, shipped with OMSA v8.1) or an earlier version of this DLL. Later versions of this DLL report the wrong VD capacity on my system.
It would be great if someone from the programming team responsible for that particular library could comment on this and try to shed some light on this oddity, which I'm not sure if anyone has ever experienced before.