voyager529
Dabbler
- Joined
- Jul 27, 2011
- Messages
- 36
Hello everyone!
One of these days, I'll stop messing around with my TrueNAS build. Today is not that day. After my last attempt at upgrades, my TrueNAS seems very unstable, and I'm not quite sure where to look.
Let's start with some Current Specs:
--TrueNAS-12.0-U1.1
--Ryzen 5 2600
--MSI Tomahawk B450 Mobo
--24GB DDR4 RAM (Non-ECC; ECC RAM is coming by the end of the month...)
--Corsair 750W PSU
--Intel PCIe NIC (To replace the Realtek chipset with known BSD issues)
--...some random ancient GPU enough to show a CLI because the CPU doesn't do graphics
--2xSyba PCIe x1 HBAs
--Dell SAS controller (forget which; it's in HBA mode though)
--pool1: 4x3TB Toshiba X300 + 1 HGST disk (replaced failed)
--pool2: 4x3TB WD Black + 1 HGST disk (replaced failed)
--pool3: 3x4TB HGST SAS disks
Here's the saga...
I had a failed disk in pool1. "no problem; I'll replace it!" Got the HGST disk, but I figured I'd try to do a few more upgrades while I was there. I got two more of the HGST disks with the attempt to do a simple mirror array; not much, but enough to take the edge off a pool2 that's 80% full. Well, I'm out of PCIe slots. I found an ASUS ROG STRIX B450-F mobo at Microcenter, most notably it had more PCIe slots than my current one, so I could add an additional HBA or two.
So, over the weekend, I tried putting that board in. I nearly had a heart attack when pool1 wasn't visible; turns out that TrueNAS doesn't seem to read the SATA ports on that board when it's in AHCI mode - maybe it would have in RAID...but I thought that was a bad idea and could mess up my pool even if I didn't create an array? I didn't try. I'm half thinking of getting a few more of those super stable Dell controllers, but I'd be more-or-less back to square one since I'd be using a PCIe slot to replace the onboard SATA ports.
So, I put the old board back in, replaced the failed drive in pool1 and let it rebuild, and everything is copasetic...except that now the unit has rebooted on its own, twice in as many days. I expect that sort of behavior from Windows
. On a slightly ironic note, this thing seems to hang on a reboot command that I issue - if I try to shut it down manually, it hangs after all the disks are synced...but it doesn't actually shut down or restart unless I hold the button.
I'm happy to provide logs if desired, I just don't know which logs to provide. Thanks in advance for the help.
One of these days, I'll stop messing around with my TrueNAS build. Today is not that day. After my last attempt at upgrades, my TrueNAS seems very unstable, and I'm not quite sure where to look.
Let's start with some Current Specs:
--TrueNAS-12.0-U1.1
--Ryzen 5 2600
--MSI Tomahawk B450 Mobo
--24GB DDR4 RAM (Non-ECC; ECC RAM is coming by the end of the month...)
--Corsair 750W PSU
--Intel PCIe NIC (To replace the Realtek chipset with known BSD issues)
--...some random ancient GPU enough to show a CLI because the CPU doesn't do graphics
--2xSyba PCIe x1 HBAs
--Dell SAS controller (forget which; it's in HBA mode though)
--pool1: 4x3TB Toshiba X300 + 1 HGST disk (replaced failed)
--pool2: 4x3TB WD Black + 1 HGST disk (replaced failed)
--pool3: 3x4TB HGST SAS disks
Here's the saga...
I had a failed disk in pool1. "no problem; I'll replace it!" Got the HGST disk, but I figured I'd try to do a few more upgrades while I was there. I got two more of the HGST disks with the attempt to do a simple mirror array; not much, but enough to take the edge off a pool2 that's 80% full. Well, I'm out of PCIe slots. I found an ASUS ROG STRIX B450-F mobo at Microcenter, most notably it had more PCIe slots than my current one, so I could add an additional HBA or two.
So, over the weekend, I tried putting that board in. I nearly had a heart attack when pool1 wasn't visible; turns out that TrueNAS doesn't seem to read the SATA ports on that board when it's in AHCI mode - maybe it would have in RAID...but I thought that was a bad idea and could mess up my pool even if I didn't create an array? I didn't try. I'm half thinking of getting a few more of those super stable Dell controllers, but I'd be more-or-less back to square one since I'd be using a PCIe slot to replace the onboard SATA ports.
So, I put the old board back in, replaced the failed drive in pool1 and let it rebuild, and everything is copasetic...except that now the unit has rebooted on its own, twice in as many days. I expect that sort of behavior from Windows
I'm happy to provide logs if desired, I just don't know which logs to provide. Thanks in advance for the help.