jsr

Disk#8 Below: root@TowerBackup:~# xfs_info /mnt/disk8 meta-data=/dev/md8p1 isize=512 agcount=4, agsize=244188659 blks = sectsz=512 attr=2, projid32bit=1 = crc=1 finobt=1, sparse=1, rmapbt=1 = reflink=1 bigtime=1 inobtcount=1 nrext64=1 = exchange=1 metadir=0 data = bsize=4096 blocks=976754633, imaxpct=5 = sunit=0 swidth=0 blks naming =version 2 bsize=4096 ascii-ci=0, ftype=1, parent=1 log =internal log bsize=4096 blocks=476930, version=2 = sectsz=512 sunit=0 blks, lazy-count=1 realtime =none extsz=4096 blocks=0, rtextents=0 = rgcount=0 rgsize=0 extents = zoned=0 start=0 reserved=0 Disk#13 Below: root@TowerBackup:~# xfs_info /mnt/disk13 meta-data=/dev/md13p1 isize=512 agcount=4, agsize=122094660 blks = sectsz=512 attr=2, projid32bit=1 = crc=1 finobt=1, sparse=1, rmapbt=1 = reflink=1 bigtime=1 inobtcount=1 nrext64=1 = exchange=1 metadir=0 data = bsize=4096 blocks=488378638, imaxpct=5 = sunit=0 swidth=0 blks naming =version 2 bsize=4096 ascii-ci=0, ftype=1, parent=1 log =internal log bsize=4096 blocks=238466, version=2 = sectsz=512 sunit=0 blks, lazy-count=1 realtime =none extsz=4096 blocks=0, rtextents=0 = rgcount=0 rgsize=0 extents = zoned=0 start=0 reserved=0 The only thing that seemed to be different to me was exchange =1 I promise, I only had done a preclear and format under unraid...

OK. Follow Up. I have removed the ASM1061 - assuming it was a source of errors and replaced with a LSI 9211-8i (yes, maybe overkill, but gives future options) purchased from the venerable ArtOfServer. After erasing and performing multiple preclear passes on Drive 8 and Drive 13 - both passed (although concernedly they do have smart raw read errors) Now I have within the Log Files the following weird EXPERIMENTAL XFS drive flags: Oct 2 15:51:57 TowerBackup emhttpd: mounting /mnt/disk8 Oct 2 15:51:57 TowerBackup emhttpd: shcmd (659): mkdir -m 0666 -p /mnt/disk8 Oct 2 15:51:57 TowerBackup emhttpd: shcmd (660): mount -t xfs -o noatime,nodiscard,nouuid /dev/md8p1 /mnt/disk8 Oct 2 15:51:57 TowerBackup kernel: XFS (md8p1): EXPERIMENTAL exchange-range feature enabled. Use at your own risk! Oct 2 15:51:57 TowerBackup kernel: XFS (md8p1): EXPERIMENTAL parent pointer feature enabled. Use at your own risk! Oct 2 15:51:57 TowerBackup kernel: XFS (md8p1): Mounting V5 Filesystem e2890e67-66ef-4d56-b285-3c60b8732b99 Oct 2 15:51:57 TowerBackup kernel: XFS (md8p1): Ending clean mount Does anyone know why I am getting these? I have already: used the "new configuration" tool within unraid, kept drive assignments, selected the Drives 8 and 13 clicked "erase" under each disk's setting formatted after bringing array online I have searched the forums and can't seem to find any firm clues. One possible clue may be that in the log file, when the drive is mounted the following flag is different than other disks: exchange=1 Oct 2 15:51:57 TowerBackup emhttpd: shcmd (661): xfs_growfs /mnt/disk8 Oct 2 15:51:57 TowerBackup root: meta-data=/dev/md8p1 isize=512 agcount=4, agsize=244188659 blks Oct 2 15:51:57 TowerBackup root: = sectsz=512 attr=2, projid32bit=1 Oct 2 15:51:57 TowerBackup root: = crc=1 finobt=1, sparse=1, rmapbt=1 Oct 2 15:51:57 TowerBackup root: = reflink=1 bigtime=1 inobtcount=1 nrext64=1 Oct 2 15:51:57 TowerBackup root: = exchange=1 metadir=0 Oct 2 15:51:57 TowerBackup root: data = bsize=4096 blocks=976754633, imaxpct=5 Oct 2 15:51:57 TowerBackup root: = sunit=0 swidth=0 blks Oct 2 15:51:57 TowerBackup root: naming =version 2 bsize=4096 ascii-ci=0, ftype=1, parent=1 Oct 2 15:51:57 TowerBackup root: log =internal log bsize=4096 blocks=476930, version=2 Oct 2 15:51:57 TowerBackup root: = sectsz=512 sunit=0 blks, lazy-count=1 Oct 2 15:51:57 TowerBackup root: realtime =none extsz=4096 blocks=0, rtextents=0 Oct 2 15:51:57 TowerBackup root: = rgcount=0 rgsize=0 extents Oct 2 15:51:57 TowerBackup root: = zoned=0 start=0 reserved=0 Oct 2 15:51:58 TowerBackup emhttpd: creating volume: disk13 (xfs) Attached is a full system diagnostics. towerbackup-diagnostics-20261002-1600.zip

Well... I switched Disk 5 and Disk 10 on the ASM1061. Specifically switched the cables at the ports on the card. At the same time, I had also moved the card from a 16x PCIe slot to a dedicated 1x PCIe slot just in case that was an issue. I also had run a Memtest86+ last week (8 cycles) - no apparent memory problems The ATA bus errors / link resets followed from drive 10 to drive 5. The ASM1061 seems to possibly going bad? Or at least one port? Have you guys encountered one of these cards going bad? I also seem to have another drive (Disk 13) that is now showing pending and un-correctable sectors. Probably time to replace this one. What is the current recommended 2 or 4 port SATA expansion card? ASM1061/2 still seems to be recommended a bunch. Diagnostics Attached. Cheers, jsr towerbackup-diagnostics-20260905-0000.zip

Long time lurker now reaching out for help. A few weeks ago I had one of my drives (drive 9) throw an error that took it offline. Unraid put it in emulated state. After running 3 preclear cycles on the drive, it passed. I then rebuilt the array back onto the flagged for failure drive. Since that time, I started digging into the logs curious for why I had the failure (something else caused the error?) Attached is system log of when I first had the failure (dated 8/25). I do have a full diagnostic on 8/25 but it is not "anonimized" Here (in the earlier system log) you will see that Disk 9 failed and disabled Disk 10 had interface problems (ATA bus errors - maybe bad cable?) I have had various CRC and Raw Read errors on numerous drives - although please note some CRC errors had occurred long in the past. Since then I have Run xfs file checks on all drives while in maintenance mode Run full extended smart checks on all disks (they all pass) Reseated SATA cables Replaced the Cables connected to the ASM1061 (Disk 5 and Disk 10) Replaced the Power Supply ( I had a 3-Rail 12Volt Power Supply, replaced with Single Rail - EVGA 550G3) Placed a fan on the LSI 9207-8i (I discovered it was running hot at 78 degrees C) [Disks 6,7,8,9,11,12,13,14 connected here] ***TODAY*** If I run a correcting parity sync - Unraid still informs me of a few dozen sync errors If I then run another non-correcting parity sync immediately after, Unraid is still complaining of sync errors At this point I am now trying to determine if drives may be failing. I have multiple drives with Raw Read errors, but NONE have uncorrectable, reallocated, or pending sectors. Hard drives are kinda expensive right now........ but Drive 8 and 9 seem to be increasing Raw Read errors (but nothing else) I started using Unraid Mover to clear Drive 8 to possibly remove from the array, maybe replace. During this operation, the log AGAIN complains of another "interface fatal error" I believe this was a link error on Drive 10 again if I am reading the log correctly. Main Question - what is causing the ongoing sync errors? Drive 8 or/and Drive 9 Raw Read Errors Drive 10 IO Error (cable has been replaced) Something Else? Would someone please help me look into my woes with fresh (and more experience) eyes. I seem to be running in circles and may possibly be having simultaneous problems. I have also attached a current diagnostics file. I appreciate your time.... syslog - 2026-08-25.txt towerbackup-diagnostics-20260901-2158.zip

@Hoopster When I use both the Dynamix plugin along with the IPMI Tools to display temperatures and fan speeds in the footer, the two sets of numbers keep flipping for me. In other words, Dynamix sensor information first, then IPMI Tools sensor information. A couple seconds later... IPMI Tools sensor information first, then Dynamix. Does that happen for you? My assumption is it has to do with the sensor refresh, but the flipping of the two sets of numbers is very distracting.

@Hoopster If at all possible, would you please identify which Samsung RAM modules came with the bundle you got off Ebay. Since the bundle is currently not available on ebay, I'm also looking at other options. I was wondering if the Samsung 32GB stick is the exact one mentioned in AsRock's qualified list, or if other versions work. Server memory starts to make my head spin looking not only at speeds, but Rank x Org.