Please wait...
Ooops, text client screwup. Anyway, this device is supposedly fine at
the block layer (99% health remaining based on total bytes written) so
this is a PCIe PHY issue. not sure what to make of it.
Are any of the other SMART diagnostics available? Total bytes written is a good estimator of remaining service life under normal conditions, but I'd be more interested in the number of error recoveries and block relocations it had to do, things like that.
If it were a device in a data center grade storage array, anything past a certain number of recoveries and relocations, it would consider the module to be on its death bed, and proactively evacuate it and tell you to replace it.
I get it though, you're spending your own money like the rest of us, and want to try hard not to replace anything that isn't truly broken.
Could sort of go in several rooms... news clip i saw today below. I also thought i heard it was being done it in another state, in the Midwest, limited runs. ( not same company tho, i dont think ). Sounds like long haul truckers are about to become obsolete.. and unemployed.
Aurora Innovation has launched a commercial self-driving truck service in Texas, making it the first company to deploy driverless, heavy-duty trucks for commercial use on public roads in the US. These trucks are making deliveries between Dallas and Houston. The company plans to expand to other locations like El Paso and Phoenix, according to TechCrunch.
Perhaps im less paranoid than i should be, but considering that monitoring this stuff via cellar is normal in that tech, i bet this is just a case of mistaken intent, to fuel the dislike for china. -> https://industrialcyber.co/utilities-energy-power-water-waste/us-energy-sector-at-risk-as-chinese-inverters-are-under-investigation-for-suspicious-communication-gear/
I have a feeling that worst case it was a 'should have disclosed' but it was really SoP, that everyone does at that level, so no one thought about mentioning it.
If im right, next up: "that ESP32 based smart outlet checked in with china for firmware updates.." um, well.. of course it did.
2025-04-19 16:12 from IGnatius T FoobarOoops, text client screwup. Anyway, this device is supposedly fine
atthe block layer (99% health remaining based on total bytes written)
sothis is a PCIe PHY issue. not sure what to make of it.
Are any of the other SMART diagnostics available? Total bytes written
is a good estimator of remaining service life under normal conditions,
but I'd be more interested in the number of error recoveries and block
relocations it had to do, things like that.
If it were a device in a data center grade storage array, anything
past a certain number of recoveries and relocations, it would consider
the module to be on its death bed, and proactively evacuate it and tell
you to replace it.of us,
I get it though, you're spending your own money like the rest
and want to try hard not to replace anything that isn't truly broken.
Yesterday I replaced the damned drive. I'm not aware of any error recovery or block relocation stats from SMART, but I surmise that wear leveling should take care of all that, and supposedly this drive has seen very little wear. So from what I can tell, the problems were all with the controller and the PCIe physical interface layer. But that just seemed to get gradually worse and worse.
I replaced it with a Western Digital and based on all the forum comments about people RMAing their whole datacenter fleet of Samsung 990 Pros, I will probably not be buying Samsung again any time soon. Left a bad taste in my mouth.