![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
Correctable Memory ECC @ DIMMA2(CPU1) - Asserted
- Thread starter rufik
- Start date
dasaint
[H]ard|Gawd
- Joined
- Jun 1, 2002
- Messages
- 1,715
Depends how many times do you see it?
If Many times == Yes DIMM starting to go bad
One Time Offense == Eh lets see if it does it again, if not could have been just a fluke!
If Many times == Yes DIMM starting to go bad
One Time Offense == Eh lets see if it does it again, if not could have been just a fluke!
dasaint
[H]ard|Gawd
- Joined
- Jun 1, 2002
- Messages
- 1,715
Most likely the cause, depends on what info is in the PSOD but most likely culprit is the DIMM. There are other reasons for PSODs but bad acting ram can exacerbate the underlying issue
dasaint
[H]ard|Gawd
- Joined
- Jun 1, 2002
- Messages
- 1,715
The Coredump it left behind would be far more useful 
dasaint
[H]ard|Gawd
- Joined
- Jun 1, 2002
- Messages
- 1,715
It showed int he PSOD that it was stored on the host, check your /var/core directory
run a vm-support, that will snag it
dasaint
[H]ard|Gawd
- Joined
- Jun 1, 2002
- Messages
- 1,715
looking for a file with the name
vmkernel-zdump
should be in /var/core of that folder.
vmkernel-zdump
should be in /var/core of that folder.
How Can I open that files ?
Please take a look into below screenshoot
http://s23.postimg.org/ny5sk5evf/dump_var_core.png
You can't - VMware can
Dasaint, you have time to look at it? I'm on a flight to Tampa soon.
peanuthead
Supreme [H]ardness
- Joined
- Feb 1, 2006
- Messages
- 4,699
I needed that laugh today. Thanks!
dasaint
[H]ard|Gawd
- Joined
- Jun 1, 2002
- Messages
- 1,715
I can look at it i honestly like having the entire file just to analyze deeper also i have to rebuild the core from the fragements... any chance u can get me that file through a dropbox or a file downloader and i can extract the log and analyze the files
dasaint
[H]ard|Gawd
- Joined
- Jun 1, 2002
- Messages
- 1,715
Yes i am able to decode the files and see what happened..
Still Im waiting for outputs from Dasaint
I was trying do some investigation by myself
so ...
I was looking for vmkernel-zdump files.
I decided to extract it
By WINSCP i downloaded to my computer logs vmkernel-log.1 and vmkernel-log.2
and i found
in vmkernel-log.1
What it means that BlueScreen report in world:32784 name:"netCoalesce2World" (S)
What is does mean ?
in vmkernel-log.2
What it means that BlueScreen report in world:35137 name:"vmx" (U)
What is does mean ?
What is a root cause ?
Regards
I was trying do some investigation by myself
so ...
I was looking for vmkernel-zdump files.
I decided to extract it
By WINSCP i downloaded to my computer logs vmkernel-log.1 and vmkernel-log.2
and i found
in vmkernel-log.1
What it means that BlueScreen report in world:32784 name:"netCoalesce2World" (S)
What is does mean ?
in vmkernel-log.2
What it means that BlueScreen report in world:35137 name:"vmx" (U)
What is does mean ?
What is a root cause ?
Regards
dasaint
[H]ard|Gawd
- Joined
- Jun 1, 2002
- Messages
- 1,715
Get me those two log files thats what i was really going to get...
The link u sent me didnt work.
Are your VM's (ANY of them) running E1000 or E1000E Nics? I have seen something like this before... way to fix use all VMXNet3's i saw u were using a Server 2012 so it has to be e1000e or vmxnet3. And based on what i see from your logs there, this looks like a match. (Affects all Versions of 5.x).. In this case your faulted DIMM might not be the cause of the PSODs... this one has popped up a lot internally should be fixed in next major update (update not Patch).
http://kb.vmware.com/selfservice/mi...nguage=en_US&cmd=displayKC&externalId=2059053
The link u sent me didnt work.
Are your VM's (ANY of them) running E1000 or E1000E Nics? I have seen something like this before... way to fix use all VMXNet3's i saw u were using a Server 2012 so it has to be e1000e or vmxnet3. And based on what i see from your logs there, this looks like a match. (Affects all Versions of 5.x).. In this case your faulted DIMM might not be the cause of the PSODs... this one has popped up a lot internally should be fixed in next major update (update not Patch).
http://kb.vmware.com/selfservice/mi...nguage=en_US&cmd=displayKC&externalId=2059053
Last edited:
Hi
Dont know how, but my wordpress deleted .rar file
Below kernel logs.
http://www.virtualolivia.com/logs/vmkernel-log.1
http://www.virtualolivia.com/logs/vmkernel-log.2
/var/core is uploading
Dont know how, but my wordpress deleted .rar file
Below kernel logs.
http://www.virtualolivia.com/logs/vmkernel-log.1
http://www.virtualolivia.com/logs/vmkernel-log.2
/var/core is uploading
dasaint
[H]ard|Gawd
- Joined
- Jun 1, 2002
- Messages
- 1,715
based on the logs it looks like you are hitting that issue, change out all your VMs to VMXNet3 see if your issue stops... if it continues i would be surprised to see the PSOD matching this issue...
To resolve this issue all vm's need to be using VMXNET3 (No e1000 or e1000e on the hosts at all) should get you all repaired up!
To resolve this issue all vm's need to be using VMXNET3 (No e1000 or e1000e on the hosts at all) should get you all repaired up!
dasaint
[H]ard|Gawd
- Joined
- Jun 1, 2002
- Messages
- 1,715
VMXNet3 is VMwares implementation of a Paravirtual NIC Card, it can go up to 10Gb speeds so VMs within the same host can cross communicate at 10Gb speeds as they flow within the same virtual switch backplane (Unless vlans and routing is involved and don't have a virtual router)
E1000 is a stock intel driver that has been in use since the dawn of whenever usually its a tried and true Nic... sadly for this case a big bad bug got it! E1000 is limited also on what it can do from within the VM unless you are using non-stock drivers to handle the advanced interfaces such as guest vlan tagging and other functions.
In the past it kind of went Flexible -> E1000 -> VMXNet3 where in VMwares world the VMXNet3 is the king of the drivers.. its continually updated via the VMware Tools so we put a high standard on delivering good drivers in most cases.
see this KB to explain it all
http://kb.vmware.com/selfservice/mi...nguage=en_US&cmd=displayKC&externalId=1001805
E1000 is a stock intel driver that has been in use since the dawn of whenever
usually its a tried and true Nic... sadly for this case a big bad bug got it! E1000 is limited also on what it can do from within the VM unless you are using non-stock drivers to handle the advanced interfaces such as guest vlan tagging and other functions.In the past it kind of went Flexible -> E1000 -> VMXNet3 where in VMwares world the VMXNet3 is the king of the drivers.. its continually updated via the VMware Tools so we put a high standard on delivering good drivers in most cases.
see this KB to explain it all
http://kb.vmware.com/selfservice/mi...nguage=en_US&cmd=displayKC&externalId=1001805
I just checked. I have ....
1 x Flexible" network adapter om Windows XP VM
9 x E1000 on vatta virtual router VM
1 x E1000E on Windows Server 2012 R2 VM
2 x VMXNET on CentOS VM
4 x E1000E on Windows Server 2012 R2 VM
1 x E1000 on Windows Server 2003 R2 VM
1 x Flexible" network adapter om Windows XP VM
9 x E1000 on vatta virtual router VM
1 x E1000E on Windows Server 2012 R2 VM
2 x VMXNET on CentOS VM
4 x E1000E on Windows Server 2012 R2 VM
1 x E1000 on Windows Server 2003 R2 VM
Last edited:
dasaint
[H]ard|Gawd
- Joined
- Jun 1, 2002
- Messages
- 1,715
From all i see there, the one i am concerned with is the Vyatta Router depending on the version... The old stuff use to have issues but i think they fixed it in Core 6.4 with the VMXNet3 support.
XP with Flexible (very old school) should be VMXNet anyways... Server 2k3 and XP (Might show up as a different speed dont worry about that as 10Gb Interfaces didnt really exist in the XP Days but still works great! [ http://kb.vmware.com/selfservice/mi..._1_1&dialogID=136506672&stateId=0 0 136516516 ]
If you were running 5.0 or 5.1 u would have hit the E1000E bug with Server 2012 (http://kb.vmware.com/selfservice/mi...nguage=en_US&cmd=displayKC&externalId=2058692)
VMXNet is awesome with native support in Centos
XP with Flexible (very old school) should be VMXNet anyways... Server 2k3 and XP (Might show up as a different speed dont worry about that as 10Gb Interfaces didnt really exist in the XP Days but still works great! [ http://kb.vmware.com/selfservice/mi..._1_1&dialogID=136506672&stateId=0 0 136516516 ]
If you were running 5.0 or 5.1 u would have hit the E1000E bug with Server 2012 (http://kb.vmware.com/selfservice/mi...nguage=en_US&cmd=displayKC&externalId=2058692)
VMXNet is awesome with native support in Centos
dasaint
[H]ard|Gawd
- Joined
- Jun 1, 2002
- Messages
- 1,715
u should be good then for all VMXNet3 You can also do what they recommended by disabling the RSS in windows but im a bigger fan of using VMXNet3 over that