The SoS utility writes the component log files into an output directory structure within the file system of the SDDC Manager instance in which the command is initiated, for example:

vcf@sddc-manager [ ~ ]$ sudo /opt/vmware/sddc-support/sos
[sudo] password for vcf
Welcome to Supportability and Serviceability(SoS) utility!
Performing SoS operation for MGMT domain components
Logs : /var/log/vmware/vcf/sddc-support/sos-2019-09-03-20-55-41-10053
Log file : /var/log/vmware/vcf/sddc-support/sos-2019-09-03-20-55-41-10053/sos.log
NOTE : The Health check operation was invoked without --skip-known-host-check, and so will skip Connectivity Health, Password Health and Certificate Health Checks because of security reasons.

Log Collection completed successfully for : [HEALTH-CHECK, SDDC-MANAGER, NSX_MANAGER, API-LOGS, ESX, VMS_SCREENSHOT, VCENTER-SERVER, VCF-SUMMARY]                           

esx Directory Contents

In each rack-specific directory, the esx directory contains the following diagnostic files collected for each ESXi host in the rack:

File Description
esx-FQDN.tgz

Diagnostic information from running the vm-support command on the ESXi host.

An example file is esx-esxi-1.vrack.vsphere.local.tgz.

SmartInfo-FQDN.txt

S.M.A.R.T. status of the ESXi host's hard drive (Self-Monitoring, Analysis, and Reporting Technology).

An example file is SmartInfo-esxi-1.vrack.vsphere.local.txt.

vsan-health-FQDN.txt

VMware vSAN cluster health information from running the standard command python /usr/lib/vmware/vsan/bin/vsan-health-status.pyc on the ESXi host.

An example file is vsan-health-esxi-1.vrack.vsphere.local.txt.

nsx Directory Contents

In each rack-specific directory, the nsx directory contains the diagnostic information files collected for the NSX Managers and NSX Edge instances deployed in that rack.

The number of files in this directory depends on the number of NSX Manager and NSX Edge instances that are deployed in the rack. In a given rack, each management domain has a cluster of three NSX Managers. The first VI workload domain has an additional cluster of three NSX Managers. Subsequent VI workload domains can deploy their own NSX Manager cluster, or use the same cluster as an existing VI workload domain. NSX Edge instances are optional.

File Description
VMware-NSX-Manager-tech-support-nsxmanagerIPaddr.tar.gz Standard NSX Manager compressed support bundle, generated using the NSX API POST https://nsxmanagerIPaddr/api/1.0/appliance-management/techsupportlogs/NSX, where nsxmanagerIPaddr is the IP address of the NSX Manager instance.

An example is VMware-NSX-Manager-tech-support-10.0.0.8.tar.gz.

VMware-NSX-Edge-tech-support-nsxmanagerIPaddr-edgeId.tgz
Note: This information is only collected if NSX Edges are deployed.
Standard NSX Edge support bundle, generated using the NSX API to query the NSX Edge support logs: GET https://nsxmanagerIPaddr/api/4.0/edges/edgeId/techsupportlogs, where nsxmanagerIPaddr is the IP address of the NSX Manager instance and edgeID identifies the NSX Edge instance.

An example is VMware-NSX-Edge-tech-support-10.0.0.7-edge-1.log.gz.

vc Directory Contents

In each rack-specific directory, the vc directory contains the diagnostic information files collected for the vCenter Server instances deployed in that rack.

The number of files in this directory depends on the number of vCenter Server instances that are deployed in the rack. In a given rack, each management domain has one vCenter Server instance, and any VI workload domains in the rack each have one vCenter Server instance.

File Description
vc-vcsaFQDN-vm-support.tgz Standard vCenter Server support bundle downloaded from the vCenter Server Appliance instance having a fully qualified domain name vcsaFQDN. The support bundle is obtained from the instance using the standard vc-support.sh command.