INTL 3:770/1 3:770/3   
   REPLYADDR tnp@invalid.invalid   
   REPLYTO 3:770/3.0 UUCP   
   MSGID:  1c45c022   
   REPLY:  abfcb205   
   PID: SoupGate-Win32 v1.05   
   On 02/08/2024 23:08, Björn Lundin wrote:   
   > On 2024-08-02 16:12, The Natural Philosopher wrote:   
   >> On 02/08/2024 15:03, Björn Lundin wrote:   
   >>> On 2024-08-02 14:51, The Natural Philosopher wrote:   
   >>>   
   >>>> Never mind sync.   
   >>>>   
   >>>> It's important to wait anyway on an SSD/nvm until it has finished   
   >>>> its internal business.   
   >>>>   
   >>>> For any copy process.   
   >>>   
   >>>   
   >>> dd'ing 500 Gb to an ssd disk, I've seen sync taking 30 s or more   
   >>> and sync is of course started AFTER dd is done   
   >>>   
   >>> So - how do you know it is done its internal business?   
   >>> Not all drives have blinking LEDs   
   >>>   
   >>> sync makes it easy to know   
   >>>   
   >>>   
   >> Even sync may not be enough.   
   >> SSDS/NVM have their own internal  caching.   
   >>   
   >   
   > Then the question remains - how do you know when its done?   
   >   
   > from man sync you get   
   > "sync - Synchronize cached writes to persistent storage"   
   >   
   > In other words, doing sync at least clear the cache,   
   > which must be better than nothing? or?   
   >   
   > Or are you saying that ripping the power to a newly sync()ed disk may   
   > mess it up?   
   >   
   Yes.   
      
   > as in   
   >   
   > dd ....   
   > sync   
   > rip out usb cord for a 2.5" ssd sata in a usb case or something similar ?   
   >   
   >   
   I am on the edge of my comfort patch here.   
   If I were building an SSD I would have a diode and a large capacitor   
   inside it to make sure all its caches were dumped to NVRAM before the   
   voltage collapsed completely.   
      
   But on a big unit this could take a bit of time.   
      
   What happens between a SATA/USB plug and the actual NVRAM is a bit of a   
   mystery.   
      
   We know its nothing like a 1:1 correlation between 'sector' and physical   
   RAM location.   
   We knows that physical RAM locations are regularly shuffled for 'wear   
   levelling'   
   When is all this done?   
   What happens if, during it, there is power failure?   
      
   I honestly do not know, hence the warning to leave the SSD for a few   
   seconds before yanking any power cords   
      
   It can do no harm   
      
      
   --   
   "The most difficult subjects can be explained to the most slow witted   
   man if he has not formed any idea of them already; but the simplest   
   thing cannot be made clear to the most intelligent man if he is firmly   
   persuaded that he knows already, without a shadow of doubt, what is laid   
   before him."   
      
       - Leo Tolstoy   
      
   --- SoupGate-Win32 v1.05   
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)   
   SEEN-BY: 90/1 105/81 106/201 129/305 153/757 7715 218/700 840 220/70   
   SEEN-BY: 221/1 6 242 360 226/17 30 100 227/114 229/110 111 200 206   
   SEEN-BY: 229/300 317 400 426 428 470 550 616 664 700 230/0 266/512   
   SEEN-BY: 267/800 280/5003 282/1038 291/111 292/854 2226 8125 301/1   
   SEEN-BY: 310/31 320/219 322/757 335/364 341/66 342/200 396/45 410/9   
   SEEN-BY: 423/81 460/58 633/280 712/848 770/1 3 100 330 340 772/210   
   SEEN-BY: 772/220 230 5020/400 5058/104 5075/35   
   PATH: 770/3 1 218/840 221/6 1 292/854 229/426