Computing Blogs

(Particularly on 1Gps links)

Ensure at least 2 paths to each device, and set each path to Round Robin.

Set disk max IO to 512 from 32k

https://kb.vmware.com/kb/1003469

Set Round Robin max IOPS to 1 from 1000:

https://kb.vmware.com/kb/2069356

The max IOPS made the biggest difference in my lab. With default IOPS set to 1000:

With IOPS set to 1:

All the info I was finding was a grepping of the VMX files (always locked/busy on my ESX6.5 environment) or needed a Windows vCenter to run a Powershell command, not available on the current VCSA.

I found this little gem at https://github.com/lamw/vghetto-scripts/blob/master/shell/vswitchInfo.sh and duplicated here (Sept 2017) in case the original goes missing…

# Author: William Lam
# Website: www.virtuallyghetto.com
# Product: VMware ESXi
# Description: Query MACs on internal vSwitch
# Reference: http://www.virtuallyghetto.com/2011/05/how-to-query-for-macs-on-internal.html

if [[ $# -ne 1 ]] && [[ $# -ne 4 ]]; then
echo -e “Usage: $0 -l -v [vSWITCH] -p [PORT]\n”
echo ” -l List all ports of vSwitch(s)”
echo ” -v vSwitch to query”
echo ” -p Port to query on vSwitch”
echo -e “\n\t$0 -l”
echo -e “\t$0 -v vSwitch0 -p 1234\n”
exit 0
fi

if [ ! -e /sbin/vsish ]; then
echo “Script is only supported running on an ESXi host as vsish is not available by default on ESX”
exit 1
fi

VSISH_VSWITCH_PATH=/net/portsets

if [ $# -eq 1 ]; then
for vSwitch in $(vsish -e ls ${VSISH_VSWITCH_PATH});
do
VSWITCH=$(echo ${vSwitch} | sed ‘s/\///g’)
for port in $(vsish -e ls ${VSISH_VSWITCH_PATH}/${vSwitch}ports);
do
PORT=$(echo ${port} | sed ‘s/\///g’)
PORTINFO=$(vsish -e get ${VSISH_VSWITCH_PATH}/${vSwitch}ports/${port}status | sed ‘s/^[ \t]*//;s/[ \t]*$//’);
CLIENT=$(echo ${PORTINFO} | sed ‘s/ /\n/g’ | grep “clientName:” | awk -F “:” ‘{print $2}’)
MACADDRESS=$(echo ${PORTINFO} | sed ‘s/ /\n/g’ | grep “unicastAddr:” | uniq | sed ‘s/unicastAddr://;s/\(.*\)./\1/’)
echo -e “${VSWITCH}\t${PORT}\t${MACADDRESS%%::*}\t${CLIENT}”
done
done
fi

if [ $# -eq 4 ]; then
QUERY_PATH=”${VSISH_VSWITCH_PATH}/${2}/ports/${4}/status”
echo “Querying port path: ${QUERY_PATH}”
PNICS=$(vsish -e ls /net/portsets/${2}/uplinks/ | sed ‘$!N;s/\n/ /;s/\///g’)
echo -e “pNICS for vSwitch: ${PNICS}\n”
vsish -e get “${QUERY_PATH}”
fi

Using Tape drive as SCSI passthru, rather than whole controller.

Often causes glitches such as tape drive going offline after backup etc.

Possible solution, from https://communities.vmware.com/thread/334987 is:

Verify if the tape drive is using VMW_SATP_AULA (which is bad)

SSH into esxi 5 console

esxcli storage nmp device list

naa.500110a0014b774a

Device Display Name: HP Serial Attached SCSI Tape (naa.500110a0014b774a)

Storage Array Type: VMW_SATP_ALUA

Storage Array Type Device Config: {implicit_support=on;explicit_support=off; explicit_allow=on;alua_followover=on;{TPG_id=0,TPG_state=AO}}

Path Selection Policy: VMW_PSP_MRU

Path Selection Policy Device Config: Current Path=vmhba2:C0:T0:L0

Path Selection Policy Device Custom Config:

Working Paths: vmhba2:C0:T0:L0

Note: View the Storage Array Type

To change this to VMW_SATP_LOCAL

esxcli storage nmp satp rule add --satp=VMW_SATP_LOCAL --vendor="HP" --model="Ultrium 5-SCSI"

Next you need to remove any claims to that device

esxcli storage core claiming unclaim -t location -A vmhba1 -C 2 -T 0 -L 0

(your vmbha must match!)

esxcfg-rescan vmhba1
esxcli storage nmp device list

Reboot esxi host

Then add your Tape via SCSI device in the VM settings