VMware with no ECC memory

I had to bump RAM for ESXi 4.1 hosts.  The only thing it disabled ECC RAM.  I wonder if this very important for VMware?  Should I have it back to ECC enabled or no bother?
LVL 17
Tiras25Asked:
Who is Participating?
 
Andrew Hancock (VMware vExpert / EE MVE^2)VMware and Virtualization ConsultantCommented:
You should really be using ECC enabled.

ECC stands for "Error Correction Codes" and is a method used to detect and correct errors introduced during storage or transmission of data.

Otherwise you could get a Purple Screen of Death, it's not specific to VMware, any production server should be using ECC ram or enabled.
0
 
Andrew Hancock (VMware vExpert / EE MVE^2)VMware and Virtualization ConsultantCommented:
So are you now running a mixture of ECC and non-ECC in the same server?
0
 
Tiras25Author Commented:
Hmm I am reading some blogs I see a lot of crashes relate to the ECC memory crashes. Especially same Dell blades as I have.
 So I see if no ECC enabled so less troubles to the hardware? less crashes.  No?
0
Ultimate Tool Kit for Technology Solution Provider

Broken down into practical pointers and step-by-step instructions, the IT Service Excellence Tool Kit delivers expert advice for technology solution providers. Get your free copy now.

 
Andrew Hancock (VMware vExpert / EE MVE^2)VMware and Virtualization ConsultantCommented:
ECC memory - are an important feature of servers to correct memory errors.

If you don't have ECC enabled, or use non-ECC memory, if memory errros occur it's likely your server will crash. As servers are used 24x7 it's more likely that they will suffer more memory errors compared to a desktop computer.

VMware ESX Servers, that depend on high utilization of CPU and Memory, should have ECC enabled.
0
 
Andrew Hancock (VMware vExpert / EE MVE^2)VMware and Virtualization ConsultantCommented:
Why did you turn if off? did you have issues with ECC?
0
 
bgoeringCommented:
ECC is always good but not really required - but that is pretty much at a hardware level and thus transparent to VMware. It would be the same for VMware as it would if you had Linux or Windows for the server OS.

Worst case is memory errors without ECC will always be errors and likely take down the server, and with virtualization the impact can likely be greater because several virtual servers will go down.

If you can return the memory you bought and get some compatible ECC memory I would highly recommend it.
0
 
Andrew Hancock (VMware vExpert / EE MVE^2)VMware and Virtualization ConsultantCommented:
0
 
Andrew Hancock (VMware vExpert / EE MVE^2)VMware and Virtualization ConsultantCommented:
The conclusion we draw is that error correcting codes are crucial for reducing the large number of memory errors to a manageable number of uncorrectable errors.

Source: http://www.cs.toronto.edu/~bianca/papers/sigmetrics09.pdf
0
 
Tiras25Author Commented:
Yes the reason is mismatched DIMMs on the same machine..So they blade reports ECC disabled.
0
 
Andrew Hancock (VMware vExpert / EE MVE^2)VMware and Virtualization ConsultantCommented:
You really *MUST* ensure, that DIMMs are matched, this can cause issues and stability issues in iteself, even if it has disabled ECC, that will not be a supported config by Dell or VMware.
0
 
andyalderCommented:
> So I see if no ECC enabled so less troubles to the hardware? less crashes.  

In a way yes, if you get a dual bit error in a block of data (rather than in a block of code) the server won't crash if you use non-ECC RAM. Do you really want wrong data to be committed? personally I'd rather the box crashed.
0
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.