Hi guys.
I get constant troubles given by guest's storage, both pvscsi & vnvme. The issues are critical to the filesystem & depending services - they usually happen under heavy load, eg.
[ 9339.405856] systemd-journald[3258]: File /var/log/journal/d2121f7ebdbd47af8e221dacf6f6e33b/system.jour
nal corrupted or uncleanly shut down, renaming and replacing.
[ 9339.459043] systemd[1]: Started Journal Service.
[ 9369.467268] nvme nvme0: I/O 31 (Write) QID 2 timeout, aborting
[ 9372.795833] nvme nvme0: Abort status: 0x0
[ 9403.610547] nvme nvme0: I/O 31 (Write) QID 2 timeout, aborting
[ 9406.938814] nvme nvme0: Abort status: 0x0
[ 9437.733881] nvme nvme0: I/O 31 (Write) QID 2 timeout, aborting
[ 9441.061753] nvme0n1: Write(0x1) @ LBA 17499779664, 2048 blocks, Command Abort Requested (sct 0x0 / sc
0x7)
[ 9441.061789] nvme nvme0: Abort status: 0x0
[ 9441.062113] I/O error, dev nvme0n1, sector 17499779664 op 0x1:(WRITE) flags 0x104000 phys_seg 63 prio
class 2
[ 9471.867188] nvme nvme0: I/O 31 (Write) QID 2 timeout, aborting
[ 9475.201777] nvme nvme0: Abort status: 0x0
[ 9475.201818] nvme0n1: Write(0x1) @ LBA 17499785808, 32 blocks, Command Abort Requested (sct 0x0 / sc 0x
7)
[ 9475.201833] nvme0n1: Write(0x1) @ LBA 17499777616, 2048 blocks, Command Abort Requested (sct 0x0 / sc
0x7)
[ 9475.202195] I/O error, dev nvme0n1, sector 17499785808 op 0x1:(WRITE) flags 0x100000 phys_seg 1 prio c
lass 2
[ 9475.202812] I/O error, dev nvme0n1, sector 17499777616 op 0x1:(WRITE) flags 0x104000 phys_seg 64 prio
class 2
[ 9506.000548] nvme nvme0: I/O 235 (Write) QID 2 timeout, aborting
then filesystem follows:
[ 9509.334793] XFS (nvme0n1p1): log I/O error -5
[ 9509.336133] nvme0n1p1: writeback error on inode 22927950636, offset 563740672, sector 17499744784
[ 9509.336348] XFS (nvme0n1p1): Filesystem has been shut down due to log error (0x2).
[ 9509.338189] nvme0n1p1: writeback error on inode 22927950636, offset 590581760, sector 17499795160
[ 9509.338481] XFS (nvme0n1p1): Please unmount the filesystem and rectify the problem(s).
[ 9509.342753] XFS (nvme0n1p1): log I/O error -5
We have some control over the ESXi 8 but not over lower level of underlying hardware. Archlinux is the guest.
I'll be grateful for any ideas |& suggestions on how & what to tweak/troubleshoot.
many thanks, L.
