> Howdy,
> 
> I was wondering if anyone has experience or anything to say on the
> following topics:
> 
> 1. Linux 2.4.18 with rtl-3.1 on two disks as RAID-1 with _hardware_ RAID
> controller.

running 2.4.18 with DAC960 on a PI-200 without any problems - did not
notice any specific differences to 2.4.18-rtl3.1 on other non-raid systems.

> 
> 2. Ditto with software RAID.
> 
> 3. XFS-1.1 filesystem on linux-2.4.18-rtl-3.1.
> (Patched with rt-2.4.18 patch, and then with the xfs patch from Silicon
> graphics: http://oss.sgi.com/projects/xfs/)
> 
> Basically - as you might guess from the questions - I am trying to get
> a very fail-proof rtlinux system working (it is a automated
> astronomical telescope control). I try to avoid long fscks on ext2 or
> not completely automatic boot in due to a rescue shell (if something
> happens, although there is UPS, frequent backups, etc, but still).
> There is also hardware watchdog card in the PC, which resets it if does
> not respond in time (and this also has to be avoided during long fsck).
>
you can reduce fsck problems a lot if you mount the system read-only (excluding
any data storage partitions) and I would claim that moving any large 
data-storage to a remote system should be the better solution than trying
to make a single node fail safe. Have you concidered using network 
block-devices + a read-only rtlinux box ?

hofrat
_______________________________________________
Rtl mailing list
[EMAIL PROTECTED]
http://www2.fsmlabs.com/mailman/listinfo.cgi/rtl

Reply via email to