I just installed ESXi 6 and I'm preparing a Centos 7 VM but it's having some strange problems.
It randomly freezes any I/O operation that it's doing and keeps showing the following message in dmesg:
[ 656.996856] sd 0:0:0:0: [sda] Failed to abort cmd ffff883d6ccf79c0
[ 656.996861] sd 0:0:0:0: [sda] task abort on host 0, ffff883d6ccf5a40
[ 656.996862] sd 0:0:0:0: [sda] Failed to abort cmd ffff883d6ccf5a40
[ 656.996865] sd 0:0:0:0: [sda] task abort on host 0, ffff883d6ccf4fc0
[ 656.996866] sd 0:0:0:0: [sda] Failed to abort cmd ffff883d6ccf4fc0
[ 656.996868] sd 0:0:0:0: [sda] task abort on host 0, ffff883d6ccf5340
[ 656.996870] sd 0:0:0:0: [sda] Failed to abort cmd ffff883d6ccf5340
[ 656.996872] sd 0:0:0:0: [sda] task abort on host 0, ffff883d6ccf7100
[ 656.996873] sd 0:0:0:0: [sda] Failed to abort cmd ffff883d6ccf7100
After a few seconds it finally resumes I/O operations and shows:
[ 656.996960] sd 0:0:0:0: [sda] SCSI device reset on scsi0:0
Some important points:
It's very unlikely that this is caused by a hardware problem. This server is pretty new and I was running Centos 7 (bare-metal) before without problems.
The RAID card ( LSI MegaRAID 9361-8i ) doesn't show any issues. All SSDs are online and without SMART errors.
In case it matters I'm using the Paravirtual SCSI controller. It was automatically selected by ESXi.
Did anyone else experienced this problem?