Trying to wrap my head around coredump files.

[DarrenPenner]

So, for reasons I need to redirect our coredumps to a shared LUN.

I am trying to wrap my head around it though as I am confused.

I originally had them set up as this http://pubs.vmware.com/vsphere-60/index.jsp#com.vmware.vsphere.storage.doc/GUID-FD3877CD-DB36-4F75-B...

but then when talking to VMware support they told me I needed to use a partition and not to file, thanks to language barrier and him being relayed info from someone else I never was able to find out why.

When we tried to download the coredumps I would constantly get WinSCP Error Code 4.

So at this point I am reading through documentation trying to get this sorted out.

Should I have it dumping to a partition or to a file?

We would like to have all five hosts dump to one location, and VMware support told me we can dump to one partition but http://pubs.vmware.com/vsphere-60/index.jsp#com.vmware.vsphere.storage.doc/GUID-2F6CBA06-5BCE-42BB-B... says

"If a host that uses a shared diagnostic partition fails, reboot the host and extract log files immediately after the failure. Otherwise, the second host that fails before you collect the diagnostic data of the first host might not be able to save the core dump. ".

So others who have it pointed to a not local directory or partition, how do you have yours set up and how are you recovering them?

And does any one know how to get Ireland support instead of Indian support when calling in?

[continuum · Accepted solution]

Have you seen
Generating a VMkernel zdump manually from a dump file in ESXi host (2081902) | VMware KB

[DarrenPenner]

No I haven't, thank you.

In all my searching I was only able to find KBs on how to retrieve from core dump partitions.

[DarrenPenner]

So I have tried testing this and I haven't been able to find a KB that answers this, when I run it I am getting the following message.

DiagnosticPartition: Unable to copy the dump partition: WARNING: Start offset less than header size!

I am not using a parition, I am running

esxcfg-dumppart --file --copy --devname active --zdumpname /vmfs/volumes/Datastore/HostName_Dump

Something to note, when I run "esxcli system coredump file get" I am given the paths of the core dumps I have for my hosts, if I run

esxcli system coredump partition get

I also get an output of a configured and active dump partition, is it normal to have both active at once?

Should I completely disable dump partitions since this is 6.0 anyways and it is recommended to run file dumps?