vCenter v.7 causes PSOD

[jamesnb1]

Hello everyone,

I am pretty new to vCenter so thank you for your patience

I installed ESXi v7 on our test PowerEdge R440 Dell, using the custom DellEMC image. This was installed on a RAID 1, Dell-brand SSD so eliminate the known SD card issues which I read on this forum as well.

The ESXi version 7 was running ok for about 2-3 days, and I have 3 running VMs, all good. Then I decided to install the vCenter v7.xx as well (latest). I also install this version on RAID1, Dell-branded SAS. Also following the advice from another member here, I use the "small" option to deploy the vCenter (19GB of RAM), etc... the installation was OK. However, a day later, the Dell server experiences PSOD and restarted itself. This caused all VMs stop working...

Is something wrong with the vCenter with this version or had I done something wrong with the installation?

Thank you for any tips on this?

Kind regards

[fabio1975]

Ciao 

It would be useful to retrieve the code dump to analyze it. Did you take a screenshot (using the iDRAC console) or a photo of the PSOD?

Otherwise, you can see to recover dump file after rebooting ESXi server in PSOD.

Depending on the configuration, the core dump can have one of these forms:

- On the scratch partition --> https://kb.vmware.com/s/article/1002769

- As a .dump file to one of the host's datastores  -->  https://kb.vmware.com/s/article/2081902

- As a .dump file on the vCenter - via the netdump service -->  https://kb.vmware.com/s/article/1032051

[vxprthu]

Hey,

You can check the following log files on the esxi host to figure out what caused the reboot/psod:

/var/log/vobd.log

/var/run/log/hostd.log

/var/log/vmksummary.log

/var/run/log/shell.log

 Also, if the host is rebooted after the psod just change back the  Misc.BlueScreenTimeout parameter's value to 0 in the host's advanced settings.

In CLI you can check the vaule with the following command:

esxcfg-advcfg -g /Misc/BlueScreenTimeout

Regards,

[jamesnb1]

Hello fabio1975,

No, unfortunately, the timeout was somehow set to "0" so it was restarted immediately and as a result, I did not have any of the sceenshot of it. I knew it was restarted because the iDRAC lifecycle records the reboot in the journal. Furthermore, all of my VMs were turned off.

I just do a complete reinstallation of the ESXi v7 and the vCenter v7 again. But this time, I will definitely follow your instruction about the log files and all below, to get the records as soon as it happened. I doubt it would again as I did not do anything different than the last time with regard to installation and configuration

I will report back here in the next day or so if it restarts again.

Thank you for all of your helps