Logs relating to network

[sel57]

Does anyone know what would be the best log to investigate to tell if a vm has (intermittently) lost connectivity to the network? Someone is reporting an issue that I'm fairly certain it's not my issue (relating to the vm, host, standard switch, etc.), but I would still like to do my part before I pass it on to the network team. Thank you. :smileyhappy:

[varfuso]

You have the other VM in the same network that work, if YES the problem not is the VMware.

[sel57]

@varfuso - I am assuming the same thing, but unfortunately just saying that without performing any due diligence and providing more proof of effort won't fly from a support standpoint. There are other vm's in that same subnet on that same host, and all of them appear to be working fine right now. (the vm in question included) The problem as it was reported to me was that a particular vm loses the ability to ping another vm randomly for a couple minutes overnight on sporadic dates, which causes a soft reboot (that part by design). I don't know if this means the source vm is going offline, the destination vm is, or if there is an underlying issue on the subnet and this overnight issue on the other related subnet vm's is just going unreported.

My suspicions aside, I just want to be able to say that (among my other tests) I researched the vmware "xyz log" (if there is one I could use to investigate such an issue) and found nothing wrong with any of the infrastructure I support.

[vfk]

What about the logs inside the guest OS?  Do you have monitoring tools that you can use to correlate the ping drops?  The ping drop over-night, even though sporadic dates, does it happen more or less around the same time. Consider using affinity-rules to keep the VM together.

[mathanmailid]

You can check vmkernel.log to ensure you don`t have any issues :smileyhappy:

[sel57]

@mathanmailid - Thank you. I took a look in that log, but didn't see anything relating to network issues. (not that I understood much of what it said in there anyway) :smileywink:

[mathanmailid]

Note down the exact time frame of the issue when noticed. Covert the time frame to UTC because on ESXi records its logs in UTC time zone. After converting it to UTC check the following logs for any issues related to network logged at that time.

1. hostd.log - from the ESXi host

2. vmware.log - from the problem VM

3. vmkernel.log - from the same ESXi host

If you do not see anything relevant to networking.. you can conclude its not VMware network related issue.

[sel57]

Thanks @mathanmailid. That's exactly what I was looking for, and while I see the soft reboots, I don't see anything relating to network issues. Thanks again!

Thanks @vfk. Nothing in the guest os helped, and I should have made it clear that the other vm is not inside my vCenter environment (re: affinity rules).

As long as I can review the logs on my end and fairly confidently say this is not a problem originating in my environment, I can pass it on to the owners of the destination vm or the network team. Thanks all for the help. Much appreciated.