4 ms·
Why? I supported 40 of them (rpi 3) as remote serial hosts and custom reboot/repower for a QA farm. Not a single one failed and don't think I had a single issue
by stjohnswarts 6y ago
Why? I supported 40 of them (rpi 3) as remote serial hosts and custom reboot/repower for a QA farm. Not a single one failed and don't think I had a single issue during that whole time. As far as I know it's still going just fine. It was a cheap and dependable solution. 99% of the SD card issues go away when you don't cheap out and don't go cheap. get a large card that is 32GB plus from a respectable vendor. That's not the rpi's fault on the failures, it's the cheap ass SD cards that people buy.
- pishpash 6y agoIt is the fault of Pi hardware when it still refuses to provide a hardware shutdown button for proper file system unmounts, after all these years.
- Gibbon1 6y agoDependence on SD Cards and lack of orderly shutdown was what caused my company to not go with the Raspberry Pi. Even then we have back up power to allow the system time to shutdown.
- qubex 6y agoConsidering it was designed for IT teaching and not for embedded use, I cannot fathom why your company and many others contemplate using it in such a capacity.
- gorgoiler 6y agoMachines no longer degrade when the power is yanked. Any corruption caused by a mid-write devoltage event is handled gracefully by whatever modern filesystem of choice you are using, is it not? Is there some more fundamental hardware damage that occurs when USB devices are unplugged without prior warning to the device itself?
- throwaway8941 6y agoIt does, when your crappy hardware doesn't lie to the kernel. Go take a look at the amount of unrecoverable FS corruption discussions at LKML. It still happens relatively frequently, and is almost always caused by shitty hardware ignoring fsync and lying to the filesystem. SD cards and cheap SSDs are especially terrible in this regard (that's why I refuse to buy anything except from the most respected vendors, and they still have their issues). Many devices don't even report checksum mismatches on read requests and simply return garbage data, and most filesystems don't notice it (because they don't implement a data integrity checking mechanism of their own)
- Dylan16807 6y agoBecause of large erase blocks, it's unavoidable that a flash device needs to involve some kind of very careful bookkeeping and sector-shifting to safely handle power loss during writes. Otherwise it could lose a big chunk of sectors unrelated to what you were actually writing, and that's not something a filesystem can paper over. I barely trust a proper SSD controller to do this, let alone whatever cheap thing is in an SD card or flash drive.
- michaelt 6y ago1. A modern filesystem will stop the filesystem getting corrupted by power loss - but a file that was mid-write, or an install process that had copied half the files it needed to, will still leave that thing incomplete. 2. SD cards can wear out (i.e. too many write-erase cycles) regardless of the filesystem. 3. SD cards have controllers running firmware, doing things like write levelling and bad block remapping. Some cards have bugs, and if the card is lying to the OS about a write completing, or if the remapping table is less resilient than the filesystem's journaling, you get problems even if the OS and everything on it is perfect. 4. Problems like pins with poor connections and power supplies not providing enough current are exacerbated, as microsds have no space for power capacitors and few SD card holders are rated for hundreds of insertion cycles. 5. SD cards start out a lot cheaper than SSDs (easier to find a sub-$10 SD card than a sub-$50 SSD) and the market is awash in fakes All these problems look very similar - "My RPi stopped booting, I replaced the SD card with a freshly written one and it started booting again" - making forum anecdotes and user bug reports hard to rely on.
- sangnoir 6y agoWith the number of GPIO pins available - you could always wire-up one yourself if a physical button is important to you. That said, I don't think the Pi was ever designed for "serious" industrial usage; I don't think the production runs (or the foundation) could support that, without undoing the core mission.