Search results

@twin_savage I read up on it back then, and as I said I suspect it's more due to low temperature at the point of writing. Can't find the paper right now, but there's this slide in a(nother) JEDEC presentation: This is very old data, but I suspect the general trend of lower active temperature...

About 6 months from new or so for me in the post above until I started getting read errors that result in block relocations?

This is probably somewhat of an extreme case since the server can drop to a few °C above freezing during winter, affecting retention, but among my 3-way special-vdev drives the QLC 870 EVO is clearly worst: The 850 did 40kh or so in my workstation before ending up here, so total hours is 73k...

@mattaw Just to confirm this, would you be able to run iperf or similar? Since PCIe3 x4 would cap at 4 * ~1GB/s * 8bits/Byte ~= 32Gbit/s, and the transceiver is able to do 40Gbit/s, any number between 32 & 40 would definitively settle it.

The line mars 08 14:46:14 HOST kernel: txg_sync_thread+0x2e0/0x4a0 [zfs] suggests to my amateur reading that we're stuck waiting for a write to disk to complete, and since there's no active drive in the pool it can't complete. I managed to find some slightly older roadmap-powerpoints from one...

Are you sure they weren't talking about giving the VM the whole controller?

So then both active members of the pool is gone? That might be a state that recovery isn't tested for. Might have to file a bug for that. If you check the kernel log there should be a hung task-calltrace, something like: kernel: INFO: task kworker/u66:2:2121 blocked for more than 122 seconds...

Hi! Can you clarify the pool layout? A RAID-Z can't be done with 2 disks.

I mean you can't expect that much of a change if you're not changing the value - just inserting the default value? :) Another way to boost read speed can be to increase zfs_vdev_async_read_max_active. But doing that at the cost of write latency, which may give complex secondary effects.

Isn't that the default value? # uname -a FreeBSD 13.1-RELEASE-p7 # zfs -V zfs-2.1.11-1 zfs-kmod-v2023072100-zfs_0eb787a7e # sysctl -a|grep zfetch vfs.zfs.zfetch.max_idistance: 67108864 vfs.zfs.zfetch.max_distance: 67108864 $ uname -a Linux 6.7.6-gentoo-x86_64 # zfs -V zfs-2.2.3-r0-gentoo...

A datapoint at least: H12SSL-i + 7302P 16-Core 155W TDP with a loadavg of 6 currently 4x 64GB 3200MT/s RAM, Nvidia 3090 idling at 22 W, 2x LSI SAS2008 that are 10 W each iirc, Optane 900p @ around 4W or so 12x HDD:s & 7 consumer SSD:s - no clue 2 small 8 port gigabit switches powering 1...

@PoolsmanTeam Feature suggestion: enable usage of LUKS as an alternative underlying encryption too, while still pulling SMART-data from the underlying device. Maybe with an overridable check for two members of the pool ending up on the same underlying device. Between the developer funding for...

The description of the benchmark screenshots alternate between "Solidigm P44 Pro 1TB" and "Solidigm P44 Pro 2TB".

Is the most recent BMC firmware as recent for the dual socket version as it is for the single socket?

Think I might just order a new board while the RMA process goes through. Thanks for the help!

Who does that any more? Nah, I'm just saying it didn't pick it up at all. But I can see the drive in BIOS. Currently reformating the drive to FAT16 instead to see if that makes any difference. Since they left instructions on how to fit 16 DIMMS on a 8 DIMM board I thought Super.ROM might be some...

SUPER.ROM doesn't seem to be a thing on the H12 boards any longer? Just tried, it can find the USB stick, but it doesn't pick up a new BIOS. So, no success in booting an OS.

The H12-boards are listed as new BTW, CPUs as used. No way to verify it, but no signs of use, no wear around the holes for screws or similar.

Haha, unpacked, of course. I'd be impressed if it ran with a .zip as firmware.

Ah, further down. But yes, it's not blinking as it should. Aligns with BIOS saying it's not working.