@andrew_cb recordsize is an upper limit, ashift is the lower limit (should be 4k on modern storage devices, which is a value of 12, set on pool creation time and is viewable on the pool variables).

The record will be from one ashift allocation up to record size in multiples of ashift. The bigger the difference between ashift and recordsize, the higher potential for compression gains, as compression can only work on multiples of ashift inside the recordsize.

So if e.g. a write is 7KB, then it will be a 8KB write not 128KB.

The one change should just be boost zfs txg timeout to something like 60-120 seconds as a default.

Thank you for all of your input. Just to close my personl thread, I contacted NetGate support directly, they were very helpful.

My original purchase was this, it was ONLY 10Gb/s:
10Gtek ASF-10G-T80

Then I got this (as recommended by NetGate support), and it also worked at 1Gb/s as well as 10Gb/s:
10Gtek ASF-10G2-T

So far it's working great.

I'm nearly positive its an issue with the bearing in the fan. Fortunately, I found a replacement to order.

For those interested its this Delta Fan. Model number is KFB0612HAFDB.

It is available at Mouser and Digikey.

The fact it's in HA makes it more difficult. You would probably have to swap it on both nodes.

In the 1541 the ix NICs are in the SoC so I would not expect overheating to be an issue if the CPU is still within limits. But possibly some other component like a PHY. Certainly check the fans and sensor readings between the two nodes.

@pinkfluffyunicorn said in Netgate 4100 regular extreme-slowdowns, have to reboot:

verbiage "Improve the system load impact from Dashboard widgets"

It does not. Many redmines have a patch. This one does not but there is one in https://forum.netgate.com/topic/190824/cpu-load-on-1100/23. It may be an early version I don't know. Or just close the dashboard. :) Also the web GUI logs each web request for every widget.

@andrew_cb said in NVME Wear Rate on 6100:

@ahxcjay The KingSpec NE 2242 is one of the few widely-available options:
https://www.amazon.com/KingSpec-256GB-Performance-Internal-Ultrabook/dp/B08TTDQ5WH?th=1

Installed yesterday. It's tiny.

=== START OF INFORMATION SECTION === Model Number: NE-256 2242 Serial Number: Firmware Version: SN25845 PCI Vendor/Subsystem ID: 0x1e4b IEEE OUI Identifier: 0x0c82d5 Total NVM Capacity: 256,060,514,304 [256 GB] Unallocated NVM Capacity: 0 Controller ID: 0 NVMe Version: 1.4 Number of Namespaces: 1 Namespace 1 Size/Capacity: 256,060,514,304 [256 GB] Namespace 1 Formatted LBA Size: 512 Namespace 1 IEEE EUI-64: 0c82d5 9581000413 Local Time is: Tue Apr 22 21:16:29 2025 EDT Firmware Updates (0x16): 3 Slots, no Reset required Optional Admin Commands (0x0017): Security Format Frmw_DL Self_Test Optional NVM Commands (0x001f): Comp Wr_Unc DS_Mngmt Wr_Zero Sav/Sel_Feat Log Page Attributes (0x06): Cmd_Eff_Lg Ext_Get_Lg Maximum Data Transfer Size: 128 Pages Warning Comp. Temp. Threshold: 90 Celsius Critical Comp. Temp. Threshold: 95 Celsius Supported Power States St Op Max Active Idle RL RT WL WT Ent_Lat Ex_Lat 0 + 6.50W - - 0 0 0 0 0 0 1 + 5.80W - - 1 1 1 1 0 0 2 + 3.60W - - 2 2 2 2 0 0 3 - 0.7460W - - 3 3 3 3 5000 10000 4 - 0.7260W - - 4 4 4 4 8000 45000 Supported LBA Sizes (NSID 0x1) Id Fmt Data Metadt Rel_Perf 0 + 512 0 0 === START OF SMART DATA SECTION === SMART overall-health self-assessment test result: PASSED SMART/Health Information (NVMe Log 0x02) Critical Warning: 0x00 Temperature: 41 Celsius Available Spare: 100% Available Spare Threshold: 1% Percentage Used: 0% Data Units Read: 2,707 [1.38 GB] Data Units Written: 9,386 [4.80 GB] Host Read Commands: 50,627 Host Write Commands: 189,770 Controller Busy Time: 0 Power Cycles: 4 Power On Hours: 28 Unsafe Shutdowns: 2 Media and Data Integrity Errors: 0 Error Information Log Entries: 0 Warning Comp. Temperature Time: 0 Critical Comp. Temperature Time: 0 Temperature Sensor 1: 41 Celsius Temperature Sensor 2: 49 Celsius

It's been a while! But as I understand it the power supply used in those early 1U units was capable of retaining sufficient power that it might remain in standby after a power outage. In that situation the button on the back can be used to power it up.

If you're just using a script you might try telling the device to shutdown and halt rather than power off (go to standby).

Hi All,

For anyone who comes across this thread, here's what I've tried.

Worked with SNMP to no avail. If I had an MIB, I'd reapproach this.

Attempted to set up SMTP. The relay host only accepts IPs and not FQDNs. We configured an internal server, but were unable to relay messages off of it successfully. Next steps here would be to find a log to review or a packet capture. For now, I've set this aside.

Installed IPMItool on a Debian host and ran this command:
ipmitool -I lanplus -H x.x.x.x -U user -P password sensor list
(Substitute your IPMI IP for x.x.x.x and your IPMI user/password.)
Success... a list of all BMC-supported sensors:

After successfully implementing item 3, we pointed Zabix at the IPMI IP and can now view all the sensors. I'm still testing triggers in Zabix, but it seems to have found a path forward.

I'd love to hear what others are doing.

Depending on what your specific use case is, it may last about the same time-span, maybe a little longer since its newer. Larger the capacity you go, the more bits there are to wear and the longer it will last without needing replaced again. I run a 4 disk 2TB raid-10 on my 5100, nearly three years into wear and currently are all at 95% life remaining. Also, always good to make sure to have a battery-backup unit in place connected to the box and with either nut or apcupsd configured for it on pfSense to make sure voltage fluctuations or surges aren't whats frying it early, having graceful shutdowns when these kinda of times come up can be quite critical for storage devices to avoid corruption.

@stephenw10
Appreciate it. I just logged in and checked out what happens when you tell it to be static. I see all of the settings. Notice I had to add the gateway and set the CIDR. Simple enough!! It should go smoothly ;-)

@bumperjeep The Netgate 1100 and 2100 does not do anywhere near Gigabit when doing NAT and having the Firewall enabled. They do not have enough CPU power to handle the interrupt/packet rate.
I have a lot of experience with those units, and depending on you setup the 1100 will do a maximum of up to 500mbps. The 2100 will go a little higher because of the two NIC’s compared to only one on the 1100. The 2100 tops out in the very best cases at about 640mbps, but generally will do about 540mbps. If you want anything higher you will need to go netgate 4200 or higher.

@Gertjan

You did put me in de right direction. Did retry mounting the NVMe and made sure de Internet connection was functional (pppoe) and re-install with NVMe succeeded.

Thank you!

@rcoleman612

Haven’t really tried yet. Had other things going on. I’ll let you know. Thanks!

@petesei nice - how well does it fit on the 4200, is it less bowed top than it seems in the pics?

edit: oh I still owed you a picture - I might still do it, planning on pulling the rack out here when my replacement dock for my work laptop gets here so I can run a wire to the dock. Got back from my trip and it had died. I got a cheap hub off amazon to hold me so could still use both my monitors vs just the laptop screen.

But in mean time - I noticed on the link I gave they show exactly how I have mine setup ;)

picture.jpg

Oh and didn't mention before - I have my pi sitting on top of the fan, so it does double duty.. Dropped my pi temp quite a bit as well

pi.jpg

The pi dropped a good 15 F with sitting on top of the fan ;) if not 20..