Hi Folks,
We have a couple of Dell R930 Servers (768 GB RAM, 72 Cores and 3 TB SSD in RAID 1+0) and we are running VmWare ESXi Version 6.0 Update 2.0.
We use these servers for scale testing of our software which is based out of Linux 2.6.32.24 Kernel. These tests involve high iops and from last couple of days I keep on seeing the below mentioned error messages on the Guest VMs and finally all VMs crash due to Kernel panic.
Jun 23 23:37:39 2016: %KERN-6-INFO: mptscsih: ioc0: attempting task abort! (sc=ffff8834466f59c0).
Jun 23 23:37:39 2016: %KERN-6-INFO: sd 2:0:0:0: [sda] CDB: cdb[0]=0x2a: 2a 00 18 ca 6e c6 00 04 00 00.
Jun 23 23:37:39 2016: %KERN-6-INFO: mptscsih: ioc0: task abort: SUCCESS (sc=ffff8834466f59c0).
Jun 23 23:37:39 2016: %KERN-6-INFO: mptscsih: ioc0: attempting task abort! (sc=ffff8834467c43c0).
Jun 23 23:37:39 2016: %KERN-6-INFO: sd 2:0:0:0: [sda] CDB: cdb[0]=0x2a: 2a 00 18 be 53 16 00 04 00 00.
Jun 23 23:37:39 2016: %KERN-6-INFO: mptscsih: ioc0: task abort: SUCCESS (sc=ffff8834467c43c0).
Jun 23 23:37:39 2016: %KERN-6-INFO: mptscsih: ioc0: attempting task abort! (sc=ffff8844570f58c0).
Jun 23 23:37:39 2016: %KERN-6-INFO: sd 2:0:0:0: [sda] CDB: cdb[0]=0x2a: 2a 00 18 be 57 16 00 04 00 00.
Jun 23 23:37:39 2016: %KERN-6-INFO: mptscsih: ioc0: task abort: SUCCESS (sc=ffff8844570f58c0).
I have tried reformatting the severs , rebuilding the RAID and re-installation of VmWare ESXi multiple times, still the same issue is seen when the VMs are up and running for couple of hours.
I am unable to proceed with my scale testing due to this issue ? Would really appreciate if someone can provide some pointers to fix this issue.
Thanks,
Saurabh