Search results

Googling 3DJ103-899 seems to show those being white labelled, refurb Seagate Exos X20's.

Googling your error led me to a kernel bugzilla entry which isn't a direct hit for your problem but is playing around in the same area. It appears to have finally been patched in kernel 7.2 so that might be worth trying if you are able to.

I'd be quite seriously pissed off if updating my BMC firmware reset the passwords/usernames. So here I am 100 miles away from the machine and needing to update it for good $reasons and I do so and now my credentials don't work any more? I think not.

I had a look at the specs of an H700 and it is PCIe 2.0 so quite bandwidth limited. From another post it would appear the OP has an E5-26xx-v2 based system which would be PCIe 3.0 so they'd be better off looking at an H730 which are much more modern and much better cards and also come with up to...

Also, if you aim to use a hardware RAID card I think you can do a lot better than an H700 as those are ancient.

RS Components in the UK are listing an Okdo Rock 5B SBC with 8GB for £107.12 which appears to be a misprice. The 4GB version is £116, the 16GB version is £195 so realistically you'd expect the 8GB version to be somewhere in the middle at about £160 which is what other sites are selling it for...

So £40 each but needs £400 RAM to make them halfway useful.

You might try running /usr/bin/update-smart-drivedb to see if an updated data file might id those unknown attributes.

What is the item in those screenshots that says "Power Reporting Deviation (Accu..." that is at 199.1%?

In linux after it reboots, does dmesg -T | grep -i reason say anything useful?

If it reboots "almost straight away" then I think it's unlikely to be thermals. My first thought would be PSU. And, yes, anything that causes your machine to reboot unexpectedly needs fixing even if it only happens under certain circumstances as you never know when you'll hit that in real life.

OTOH, once you have spent the $2500 on Claude, it's gone and you have nothing to show for it. If you spend it on hardware then at least you still have hardware with some residual value left over at the end.

I've never heard of "VMD" before but this sounds similar 219220 – nvme nvme0: I/O timeout, completion polled

Generally the Dell H3xx series of boards are gutless and underpowered for hardware RAID and have no onboard memory. You need to look at the H7xx series to get a decent card.

I'd look at the logs and see if there was a problem that is now resolved. That's about the only reason I can think of for a "failure" to be given a green tick and appear good. Perhaps it just needs acknowledging?

Be aware that the better the thermal pad conductivity, the less compressible the material is likely to be. I would guess for this purpose you'd want to use a thermal pad with the best conductivity that you can get, perhaps something like Gelid Ultimate or Extreme (not as good). I recently...

Maybe add ventilation to the plans...

You can also PXE boot systemrescue.

I think DDR3 RAM will be just too slow to be useful for this. DDR3 is about half the speed of DDR4 which in turn is about 1/3rd the speed of DDR5.

I don't understand why these things are so cheap in the USA and so expensive over here in .uk. I checked Ebay UK and the cheapest there is a 640 for £360 (US$480) and there are barely any others for sale - which probably explains why they're so expensive.