Hi experts.
we found one of our VM's frozen and digging deep into "Alert/Events" in virtual center we find the ESX had failed with NVRAM: write failed.
Can you please help us troubleshoot what went wrong?
Events button on VM shows this error
Error message from ESX XXX.XXXX.XXXX.com:NVRAM: write failed.
-Srinivas
Forgot to mention ESX server version in hurry:
we have all ESX servers running version 5.1.
-Srinivas
Hi Sriven,
can you post more information about your environment and versions/build please... othervise its just shot in the dark.
Please attach vmkernel.log from your host and vmware.log from VM.
This error could be caused by many things, faulty hardware Disk, RAM, HBA or maybe its just corrupted file or full disk...
Please have a look at host sensors under Hardware tab or through Configuration tab - > Health Status ...
Also check if there is sufficient disk space on host datastore and try to perform some hardware diagnostic.
Regards,
P.
Hi, i backup with two options (memory dump and quiesce file system).
I notice that the Vsphere client riports "VM TOOLS NOT RUNNING" during quiesce, is it correct?
Best praticte to backup service "high memory" as Exchange, is NOT quiesce?
Hi,
Try to make the same choice in the Web Client 
You will see that this is a mistake in the Classic Client, you simply cannot make a memory snapshot and quiesce the file system as well, it would be pointless.
Since you are choosing the first option you will get a fully consistent state including system memory, there is no need for quiescing, and since you have quite a bit of RAM and possibly change rate this also explains the long time your snapshots are taking.
As the VM gets "stunned" to be able to cope with the change you will probably notice some service interruption, this maps perfectly to the symptom of you seeing the tools as not running.
If you really want to quiesce the system, do NOT check the memory option. You then will get a snapshot that takes file changes into consideration using Microsoft VSS as a basis.
The only "disadvantage" without backing up the memory state of the system is that Windows will do a hard reboot after a restore as it comes with empty system memory, yet again I don't know of any enterprise backup solution that would actually dump the memory of a system into the backup.
so for those of us already stuck with m1015 or LSI counterpart, then for spindle disk perhaps a solution can be to leave each disk as a JBOD/pass thru and then add each disk individually to windows VM and do RAID1 from there. a little messy but perhaps performance will be slightly better. i have one of these cards in a HP microserver with a couple 512GB ssd. the sata disk raid keeps popping events for latency in esxi5.1 and at times the users have complained the server is slow and it looking like that could be part of the issue. granted its a microserver, but its only running win2k8 VM that is a fileserver for 3 users.
Thanks, i'm using trilead and i can flag TWO options:
INCLUDE MEMORY DUMP OF THE VM
QUIESCE THE FILE SYSTEM IN THE VIRTUAL MACHINE
i'm getting lots of Exchange errors while doing backup.
The server is an Exchange server.
What you think, quiesce or not quiesce?
Pros and cons?
Thanks
Quiesce but don't choose the memory option.
Pro: Your file system will still be consistent and the backup should be much faster without showing the errors
Con: A restore will give you the the impression of a system crash during the initial boot, simply ignore it 
With Exchange, Databases and Active Directory vendors usually also provide some sort of agent to make backup and restore of single mailboxes possible.
Done...faster!
Why trilead says that i have to take snapshot with memory?
Glad it worked, white box motherboards with onboard SATA insist that the drive containing the OS be on SATA port 0, just the way it is.
Another piece of bizarreness from Adaptec is that its /var/log/arcerror.txt file keeps growing with entries like this:
FFD1AB70 | Fri 09May2014 (tz=0)
FFD1AB70 | 23:34:48.707529000 DG_SMIS: ArcAdapter.cpp: Line #: 12440
StorLib::getSystemConfig(char** pXmlOUT)
ArcSystem::buildChildren(Ret & retOUT)
ArcAdapter::buildChildren(Ret &retOUT)
ArcAdapter::smpPassThrough()
*** FSA API Error: FsaSendSMPCommand() fsaStatus=661 ***
What is that mean to mean?
I have several 3Ware 9650SE (4 port) cards and I want to run them with Supermicro's H8SCM board. I have had no problems running ESXI 5.1 on this platform. and Indeed, I have installed 5.5 without any issues so far. I loaded up windows 8 on a VM and I am loading up WS2012R2 which is the ultimate goal. My question is this: Has anyone had any success running this configuration or in particular this RAID controller card?. I have read that there is a problem with it on http://www.experts-exchange.com/Software/.VMWare/Q_28328401.html but I have yet to experience this issue. I am using 2.27.08.036-1OEM.500.0.0.472560.x86_64.vib for the lsi driver file and 5.50 update 1 of esxi.
Hi all,
is there any common reference to read about tier1, tier2 technical term?
data center tier, capacity management tier, application tier
thx
I will like to know if "Round Robin" multipathing policy is supported with boot LUNs when booting from SAN.
Environment is : IBM blades (HS22) in Bladecenter H chassis, ESXi 5 U3, IBM V7000 SAN (true active active), SRM 5.
Hi All,
Both Floodping and unicast tests of the NIC I/O Certification failed with the following error :
2014-05-09 09:41:42 UTC [May 9 2014 09:41:42 : TESTHASH ] [0] ERROR: Failed to execute processSetup for the Setup : Failed to execute handleHostVMObj for the Setup : SetIP: Internet Systems Consortium DHCP Client V3.0.5-RedHat
2014-05-09 09:41:42 UTC Copyright 2004-2006 Internet Systems Consortium.
2014-05-09 09:41:42 UTC All rights reserved.
2014-05-09 09:41:42 UTC For info, please visit http://www.isc.org/sw/dhcp/
2014-05-09 09:41:42 UTC
2014-05-09 09:41:42 UTC Listening on LPF/eth0/00:0c:29:0c:15:a4
2014-05-09 09:41:42 UTC Sending on LPF/eth0/00:0c:29:0c:15:a4
2014-05-09 09:41:42 UTC Sending on Socket/fallback
2014-05-09 09:41:42 UTC
2014-05-09 09:41:42 UTC [May 9 2014 09:41:42 : TESTHASH ] [0] WARN: found objects that need to be cleaned up
2014-05-09 09:41:44 UTC [May 9 2014 09:41:42 : VIRTUALMAC] [0] INFO: Cleaning up VM: vtafVM2_192.168.168.161_27747.............
2014-05-09 09:41:44 UTC [May 9 2014 09:41:44 : MAIN ] [0] ERROR: Use of uninitialized value $mac in exists at /opt/vmware/VTAF/TestLib/Sphere/VM/VirtualMachine.pm line 3923.
2014-05-09 09:41:44 UTC
2014-05-09 09:41:44 UTC [May 9 2014 09:41:44 : UTILITY ] [0] ERROR: Use of uninitialized value $mac in exists at /opt/vmware/VTAF/TestLib/Sphere/VM/VirtualMachine.pm line 3923.
2014-05-09 09:42:15 UTC ......................................................................................................................................................................................................................................................................................
2014-05-09 09:42:15 UTC [May 9 2014 09:42:15 : UTILITY ] [0] INFO: ----------------- Test Summary ----------------
2014-05-09 09:42:15 UTC [May 9 2014 09:42:15 : UTILITY ] [0] INFO: LogDirectory : /xsigo/TestSessions/new_cert.d/FloodPing/20140509093530/nic55-cert/VTAF/Test/Networking/Driver/Functional/FloodPing/FloodPing_May_09_2014_09.35.34_27747/logs/
2014-05-09 09:42:15 UTC [May 9 2014 09:42:15 : UTILITY ] [0] INFO: TestResult : FAIL
2014-05-09 09:42:15 UTC [May 9 2014 09:42:15 : UTILITY ] [0] INFO: ZipDirectory : /xsigo/TestSessions/new_cert.d/FloodPing/20140509093530/nic55-cert/VTAF/Test/Networking/Driver/Functional/FloodPing/FloodPing_May_09_2014_09.35.34_27747/zip/
2014-05-09 09:42:15 UTC [May 9 2014 09:42:15 : UTILITY ] [0] INFO: StartTime : May 9 2014 09:35:35
2014-05-09 09:42:15 UTC [May 9 2014 09:42:15 : UTILITY ] [0] INFO: EndTime : May 9 2014 09:42:15
2014-05-09 09:42:15 UTC [May 9 2014 09:42:15 : UTILITY ] [0] INFO: Duration : 0 day(s) 0 hr(s) 6 min(s) 40 secs
2014-05-09 09:42:15 UTC [May 9 2014 09:42:15 : UTILITY ] [0] INFO: -----------------------------------------------
2014-05-09 09:42:15 UTC [May 9 2014 09:42:15 : MAIN ] [0] ERROR: Trouble creating test instance, Failed to execute processSetup for the Setup : Failed to execute handleHostVMObj for the Setup : SetIP: Internet Systems Consortium DHCP Client V3.0.5-RedHat
2014-05-09 09:42:15 UTC Copyright 2004-2006 Internet Systems Consortium.
What can be done to fix this issue?
Thanks for your help.
Regards,
Senthil
This below documentation gives you all the requirements for Boot from SAN for ESXi
I have checked this a bit more thoroughly, the difference between E1000 and VMXNet3 Cards are that VMXNet3 is an implementation of paravirtualized driver. Hence it handles the I/O more efficiently.
The paravirtualization provides specially defined 'hooks' to allow the guest(s) and host to request and acknowledge these tasks, which would otherwise be executed in the virtual domain (where execution performance is worse). A successful paravirtualized platform may allow the virtual machine monitor(VMM) to be simpler (by relocating execution of critical tasks from the virtual domain to the host domain), and/or reduce the overall performance degradation of machine-execution inside the virtual-guest.
Above quote is from wikipedia.
We have VM's held in certain containers\folders and I wondered if you can in someway show the total number of devices in the folder? I would like to be able to look at the folder structure at at the top level and tell how many VM's are in each container\folder in vSphere. Failing that if someone can tell me another way to view the numbers that would be great. Thanks
Hello,
bear in mind that if the in the snapshot the memory option is checked, the virtual machine's memory must be dumped to disk.
This would take a long time if the virtual machine has to much memory. While the memory is dumping to disk, the virtual machine will be unresponsive.
The memory option in the snapshot creation is enabled by default.
The best options as commented by Frank, is to choose quiesce but not memory option.
Enabling the memory options permits to return to the running virtual machine state (with the same memory content) as it was when the snapshot was taken.
Check the "Creating a snapshot" section in this vmware's kb, to be able to gain more knowledge about this topic.
Best regards,
Pablo
Like object
Error That i see:
TRILEAD BACKUP
The Backup has finished successfully, but Quiesce snapshot could not be performed. ESX said: Vim25Api51.LocalizedMethodFault - An error occurred while saving the snapshot: Failed to quiesce the virtual machine.
ESXI Hostd
DatastoreBrowserImpl::PopulateSearchResult: File: '/vmfs/volumes/53467ede-3e8a7bfa-ece5-3440b5ac1f80/srv-exchange/srv-exchange-000002.vmdk' is owned by user 'root'
In Event Viewer seeing a lot of error of VSS of Mailbox Replication.
Any suggestion?
