Re: PKGBase and Embedded Systems
- Reply: Bob Bishop : "Re: PKGBase and Embedded Systems"
- In reply to: Bob Bishop : "Re: PKGBase and Embedded Systems"
- Go to: [ bottom of page ] [ top of archives ] [ this month ]
Date: Tue, 12 Aug 2025 13:55:16 UTC
On 8/12/2025 08:51, Bob Bishop wrote: > Hi, > >> On 12 Aug 2025, at 12:40, Karl Denninger<karl@denninger.net> wrote: >> >> Well, ok, "sort-of" embedded systems. Think firewalls. >> Right now I build a USB stick-based setup for these on NanoBSD and, for some other hardware in somewhat-similar applications (e.g. home control, etc.) for the PI series using Crochet. >> /var is volatile on both where /usr/local/etc has a "save" mechanism (along with /etc) in both environments; that is, its volatile while running, but can be instructed to sync with the saved copy thus on a reboot/reset/powerloss the last-saved is retained. >> A couple of times I've concluded the "best" way to deal with things that dump state they'd like to keep in /var somewhere (usually in /var/db), where the "thing" doesn't have a command-line switch to change that, is to move that directory to /usr/local/etc/db and then symlink it during the setup, thus it becomes "volatile but subject to save" as with anything else in /usr/local/etc. > We used to do that kind of thing. Now that storage, RAM and 64bit boxes are cheap we just use a full install on ZFS and make everything except the volatile bits read-only… Its not so much a "how cheap is the resource" problem (yes, that's gotten a lot cheaper over time) its a "the box MUST come back online after an unsolicited power event." That in turn means the physical volume cannot be open for write, particularly in the instance of a volume that co-mingles various metadata that is internal to the device itself (e.g. an SSD which does its own internal wear leveling and such.) 95% of the time IMHO isn't good enough. -- Karl Denninger karl@denninger.net /The Market Ticker/ /[S/MIME encrypted email preferred]/