General question about CF II vs. Microdrive
I have just bought an ALIX 2c1 board with a 4GB CF II card and setup pfSense version 1.2.
I have tried different routers before (OpenWRT, DD-WRT, etc…) but pfSense so far really seems to be the most matured BSD-based router platform... not only in functionality but also in stability
Enough asskissing ;-) I am wondering why setting up the full installation on a CF card posthaste destroys the card... Why doesnt this happen to an Linksys WRT54GL flash memory? ??? Anyone some links to a CF II interna?
Sorry if this is a dump question...
This only happens if you install the full version onto a CF.
This is due to repeated cycles to the same sector.
Read up on wikipedia why these sectors die.
The alternative is to use a microdrive (Harddisc in C-Format).
But usually you install the embedded version to a CF.
With the embedded version, the card will be mounted read only and wont die.
Thx for your quick reply. Well, I have red the German Wikipedia article about CF and if I have got it the right way the problem relys on how the firmware handles randomized access on the card. Just a theoretic question: if one would be able to change the way the firmware accesses the sectors one could use a CF card for full install or generally spoken as a harddisk?
Im Grunde gnommen ja.
Es gibt auch CF-Karten die schon von Haus aus die schreibzugriffe regelmässig über die ganze karte verteilen.
Die sind allerdings meist etwas teurer und werden Industrial-Flash oder "solid state disk" genannt.
There are also CF-cards that level the write-cycles over the whole card by default.
Usually they are called industrial-flash or solid state disk.
Hmm… I don't understand this. Industrial grade CF cards nowadays have wear-leveling built-in and endurances of 1,000,000-5,000,000 write cycles and MTBFs in the millions of hours as well. Many of the hard drives used in pfSense boxes nowadays, will never reach those MTBFs, just because of mechanical wear. Yet everyone is still told to favour microdrives etc over solid-state (or industrial grade) CompactFlash cards. Why is that?
It seems to me using those industrial grade cards should be recommended rather than discouraged. Surely it used to be the case that even the best CF cards failed regularly, but we have moved on since then, though obviously one shouldn't use cards for $2.
Thanks again to everyone working on this project, it is a blessing.
Probably because most people will not be buying the (relatively expensive) industrial grade cards but the consumer ones.