Knowledge Base

This patch updates the esx-base VIB to resolve the following issues:

• PR 755467: On Dvportgroup, a Promiscuous port might not work in Promiscuous mode if the DVMirror sessions are reconfigured.
• PR 775353: After applying a host Profile that has the ESXi firewall on and the Fault Tolerance rule blocked, a warning message similar to the following might be unnecessarily displayed:
  Ruleset faultTolerance doesn't match the specification
• PR 778127: For some ESXi systems, vSphere client might display information about a non-existent power supply source. This patch improves handling of sensor data from the IPMI sensor data repository (SDR) to resolve the issue.
• PR 816486: While using VMware SVGA 3D driver if you set 3D enable option, then the text labels in group boxes are overwritten by rectangular frame on WPF applications. This issue was observed on ESX hosts running Atlas Client applications.
• PR 817713: The System Event Log in the vSphere Client might be empty if ESXi 5.0.x is run on certain physical servers.
  The host's IPMI logs (/var/log/ipmi/0/sel) might also be empty.
  An error message similar to the following might be written to /var/log/messages:
  Dec 8 10:36:09 esx-200 sfcb-vmware_raw[3965]: IpmiIfcSelReadAll: failed call to IpmiIfcSelReadEntry cc = 0xff
• PR 818092: Virtual machine fails with monitor panic if paging is disabled. An error messages similar to the following is written to vmware.log:
  vcpu-0| MONITOR PANIC: vcpu-1:VMM64 fault 14: src=MONITOR
  vcpu-0| rip=0xfffffffffc262277 regs=0xfffffffffc008c50
• PR 819959: An ESXi host might fail with a purple diagnostic screen at Kseg_ReleaseVA() or Kseg_ReleasePtr() with a NULL "va" or pointer due to an unclean exception handling code path in the ESXi network stack.
• PR 821460: vm-support utility cannot collect partition table listing of vmfs5 volumes by using fdisks -lu. The vm-support utility is enhanced to collect outputs for partedUtil getptbl and partedUtil getUsableSectors. This helps in getting partition information of vmfs5 volumes.
  
• PR 823777: Host profile creation fails and the ESXi host logs messages similar to the following to /var/log/vmware/hostd:
  Error: 'nmp.nmpprofile.FixedPspPolicy: Unrecognized Fixed PSP configured path iqn.2000-04.com.qlogic:qle4062c.yk10ny9cl5yk.1-00c0dd1c341f,iqn.1992-04.com.emc:cx.ckm00094800328.a2,t,1-naa.6006016090d0260071c460297bc1df11'
  This issue occurs if you have configured a preferred path for iSCSI LUNs with fixed path policy.
• PR 827707: An ESXi host is unaware of unmapped luns from DS8300 storage array, fails to mark the paths as dead, and does not remove these unmapped luns. Commands to unmapped luns on DS8300 fail with sense key 0x0b and the ESXi host is unable to record a Permanent Device Loss (PDL) case. PDL is now discovered based on ASC sense data and not based on sense key.
• PR 829486: Some HP servers experience a situation where the PCC (processor Clocking Control or Collaborative Power Control) communication between the VMware ESXi kernel (vmkernel) and the server BIOS does not function correctly. As a result, one or more PCPUs might remain in SMM (System Management Mode) for many seconds. When the vmkernel notices a PCPU is not available for an extended period of time, a purple diagnostic screen occurs dsiplaying messages similar to the following:
  
  PCPU 39 locked up. Failed to ack TLB invalidate (total of 1 locked up, PCPU(s): 39).
  0x41228efc7b88:[0x41800646cd62]Panic@vmkernel#nover+0xa9 stack: 0x41228efe5000
  0x41228efc7cb8:[0x4180064989af]TLBDoInvalidate@vmkernel#nover+0x45a stack: 0x41228efc7ce8
  
  @BlueScreen: PCPU 0: no heartbeat, IPIs received (0/1).
  ...
  0x4122c27c7a68:[0x41800966cd62]Panic@vmkernel#nover+0xa9 stack: 0x4122c27c7a98
  0x4122c27c7ad8:[0x4180098d80ec]Heartbeat_DetectCPULockups@vmkernel#nover+0x2d3 stack: 0x0
  ...
  NMI: 1943: NMI IPI received. Was eip(base):ebp:cs [0x7eb2e(0x418009600000):0x4122c2307688:0x4010](Src 0x1, CPU140)
  Heartbeat: 618: PCPU 140 didn't have a heartbeat for 8 seconds. *may* be locked up
• PR 836555: When you upgrade the ESX host from 4.1 to 5.0 update1, firewall rules change from being disabled to enabled. This happens after you create a host profile, modify it and then apply the same profile to the host.
  
• PR 838575: The network scheduler uses a single queue to allocate bandwidth between all virtual machines sharing the same network resource pool. If multiple virtual machines send network data at the same time, the available bandwidth might get exhausted before it is allocated proportionally to all virtual machines.
  The improvised algorithm now uses a time based scheduling scheme to optimally allocate bandwidth to all virtual machines sharing the same network resource pool.
• PR 839658: When you use the vmkfstools –q option with the vm-support utility, the generated file might not list the attributes of the virtual machine vmdk files.
  
  
• PR 841778: When attempting to take a quiesced snapshot of a virtual machine, if the snapshot operation fails towards the end of its completion, the redo logs created as part of the snapshot are not consolidated. The redo logs might consume a lot of datastore space.
  With this patch if the quiesced snapshot operation fails, the redo log files are consolidated.
• PR 842015: When a virtual machine with snapshots is deleted, independent or non-independent virtual disks that were detached from the virtual machine but are part of an existing snapshot might also be deleted.
• PR 842481: Netdump fails during core dump collection after a purple-diagnostic screen appears if the vSwitch security option Mac Address Changes is set as Reject.
• PR 844260: VMFS might issue I/Os to a volume when a VMFS heartbeat reclaim operation is in progress or a virtual reset operation is performed on an underlying device. As a result, alert and warning messages similar to the following are logged:
  
  ALERT: ScsiDeviceIO: SCSIAsyncDeviceCommand:3082: Failed command 0x2a to quiesced partition naa.9999999999
  
  WARNING: ScsiDeviceIO: 2360: Failing WRITE command (requiredDataLen=512 bytes) to write-quiesced partition naa.9999999999
  
  With this patch, the alert messages are removed and warnings are changed to log messages.
  
• PR 845859: When a Linux kernel crashes, the linux kexec feature might be used to enable booting into a special kdump kernel and gathering crash dump files. An SMP Linux guest configured with kexec might cause the virtual machine to fail with a monitor panic during this reboot. Error messages such as the following might be logged:
  vcpu-0| CPU reset: soft (mode 2)
  vcpu-0| MONITOR PANIC: vcpu-0:VMM fault 14: src=MONITOR rip=0xfffffffffc28c30d regs=0xfffffffffc008b50
  
• PR 849959: If the SMBIOS version of the ESXi 5.0 system is of version 2.6 or later, the SMBIOS UUID reported by the ESXi 5.0 host might be different from the actual SMBIOS UUID. The byte order of the first 3 fields of the UUID is not correct.
• PR 851071: This issue occurs on servers that have access to LUNs which are claimed by VMW_SATP_LSI module. A memory leak that exists in VMW_SATP_LSI module forces the module to run out of memory. Error messages similar to the following are logged to vmkernel.log file:
  
  Feb 22 14:18:22 [host name] vmkernel: 2:03:59:01.391 cpu5:4192)WARNING: Heap: 2218: Heap VMW_SATP_LSI already at its maximumSize. Cannot expand.
  Feb 22 14:18:22 [host name] vmkernel: 2:03:59:01.391 cpu5:4192)WARNING: Heap: 2481: Heap_Align(VMW_SATP_LSI, 316/316 bytes, 8 align) failed. caller: 0x41800a9e91e5
  Feb 22 14:18:22 [host name] vmkernel: 2:03:59:01.391 cpu5:4192)WARNING: VMW_SATP_LSI: satp_lsi_IsInDualActiveMode: Out of memory.
  
  The memory leak in the VMW_SATP_LSI module has been resolved with this patch.
• PR 851138: In ESXi, when vmxnet3 is used as vNIC in some virtual machines and you turnoff packet coalescing, the ESXi host might fail with a purple screen as the virtual machine is booting up.
  
• PR 852729: The datastores residing on the LUNs of the storage controllers might be inactive and a virtual machine might be inaccessible in vSphere client when you encounter a storage fault. The datastores remain inactive until a manual rescan is performed. This happens when the management agent, hostd, fails to handle esx.clear.storage.redundancy.restored vob message properly.
• PR 856566: Attempts to create a diagnostic partition using the vSphere client might fail with the following error message:
  Partition format unknown is not supported.
  This issue occurs if a valid partition format does not exist while creating a diagnostic partition and partition format “unknown” is returned.
  
  
• PR 858424: On an ESXi host, a secondary Fault Tolerance virtual machine installed with VMXNET 3 adapter might fail. Error messages similar to the following are written to vmware.log:
  
  Dec 15 16:11:25.691: vmx| GuestRpcSendTimedOut: message to toolbox timed out.
  Dec 15 16:11:25.691: vmx| Vix: [115530 guestCommands.c:2468]: Error VIX_E_TOOLS_NOT_RUNNING in VMAutomationTranslateGuestRpcError(): VMware Tools are not running in the guest
  Dec 15 16:11:30.287: vcpu-0| StateLogger::Commiting suicide: Statelogger divergence
  Dec 15 16:11:31.295: vmx| VTHREAD watched thread 4 "vcpu-0" died
  
  This issue does not occur on a virtual machine installed with E1000 adapter.
  
• PR 858535: When creating a quiesced snapshot of a Microsoft Windows Server 2008 R2 virtual machine, if you specify the working directory, the snapshot operation might fail with the following error message:
  Snapshot guest failed: Failed to quiesce the virtual machine.
  
  
• PR 859249: If the SMBIOS version of the ESXi 5.0 system is of version 2.6 or later, the SMBIOS UUID reported by the ESXi 5.0 host might be different from the actual SMBIOS UUID. The byte order of the first 3 fields of the UUID is not correct.
• PR 861418: When an ESXi host uses more than approximately 1024 dvPorts, the virtual machine stops responding.
  The following warning messages are seen in the vmkwarnings:
  vmkwarning.log:2012-04-20T14:16:42.999Z cpu4:2898)WARNING: Heap: 2639: Heap dvsLargeHeap (65582032/67108864): Maximum allowed growth (1527808) too small for size (3719168)vmkwarning.log:2012-04-20T14:16:42.999Z cpu4:2898)WARNING: Heap: 2900: Heap_Align(dvsLargeHeap, 3715488/3715488 bytes, 8 align) failed. caller: 0x418025d29338
• PR 863267: Changing MTU multiple times might lead to netGPHeap depletion and memory leak, causing ESXi host to stop responding with a purple screen due to an issue with the Intel async driver.
  
  The vmkernel.log file might contain heap exhaustion messages similar to the following:
  
  WARNING: Heap: 2525: Heap netGPHeap already at its maximum size. Cannot expand
  WARNING: Heap: 2900: Heap_Align(netGPHeap, 8192/8192 bytes, 64 align) failed. caller: 0x41802c10a03f
  WARNING: NetPort: 1244: failed to enable port 0x2000002: Out of memory
  NetPort: 1426: disabled port 0x2000002
  Uplink: 5240: vmnic2: Failed to enable the uplink port 0x2000002: Out of memory
  <3>ixgbe: vmnic2: ixgbe_alloc_tx_queue: allocated tx queue 1
  <3>ixgbe: vmnic2: ixgbe_alloc_tx_queue: allocated tx queue 2
  <3>ixgbe: vmnic2: ixgbe_alloc_tx_queue: allocated tx queue 3
  <6>ixgbe 0000:03:00.0: vmnic2: changing MTU from 9000 to 1500
  
• PR 864236: VMFS journal replay might fail with the following message on an ESXi 5.0 host:
  J3: 3167: Failed to replay extended transaction on 4a1aa282-32d04c23-03a2-001517ab207b possibly pending online upgrade
  WARNING: HBX: 4336: Replay of journal on vol 'san1_vmfs3' failed: Bad parameter
  
  
• PR 864963: The reboot following a successful reboot of ESXi 5.0 U1 from FCoE SAN or FC SAN or iSCSI SAN during an upgrade from ESXi 5.0 might result in the loss of configuration for Cisco UCS N20-B6625-2 blade. Also, during a fresh installation of ESXi 5.0 U1 and a subsequent reboot, the UCS blade loses the configuration and reverts back to its previous state.
  You can face a similar issue while performing any of the following actions:
  • Installing third party nics
  • Installing the security patches
  • Shutting down ESXi 5.0 U1 for maintenance
  • Rebooting ESXi 5.0 U1 for maintenance
  • Shutting down ESXi 5.0 U1 for any unplanned power shortage issue
• PR 873057: If you modify the /etc/rc.local file in ESXi 4.1.x hosts, and upgrade these ESXi 4.x hosts to ESXi 5.0.x, the upgraded hosts might not be able to re-connect to vCenter Server due to liscensing related issues. Error messages similar to the following might be displayed:
  The Unlicensed license for Host xxx does not include VMware DRS. Upgrade the license
  
  The vpxd log files might contain entries similar to the following:
  
  012-04-26T11:58:37.069-04:00 [05420 warning 'Default' opID=C8581389-00000EF2] [LicMgr] Trying to remove licenses. Host was never registered with the license manager.
  012-04-26T11:58:37.069-04:00 [05420 trivia 'QueryServiceprovider' opID=C8581389-00000EF2] Clearing uncommitted generations on this thread for LicenseManager
  012-04-26T11:58:37.069-04:00 [05420 trivia 'Vpxprofiler' opID=C8581389-00000EF2] Ctr: CheckVCLicense/TotalTime = 0 ms
  012-04-26T11:58:37.069-04:00 [05420 error 'Default' opID=C8581389-00000EF2] [LicMgr] feature drs not included in license assigned to host-129707.
  012-04-26T11:58:37.069-04:00 [05420 trivia 'Vpxprofiler' opID=C8581389-00000EF2] Ctr: CheckingHostFeatures/TotalTime = 0 ms
  
• PR 873251: During ESXi boot from a Fibre Channel over Ethernet (FCoE) LUN, the FCoE device discovery takes a longer time than expected. You might not be able to restore the bootbank from the FCoE boot disk with the restore-bootbank jumpstart plug-in.
  
• PR 874446: The vmkernel.log and the dmesg commandmight show the following message:
  C1E enabled by the BIOS.
  The message may be shown even when C1E has been disabled by the BIOS, and it may not be shown even when C1E has been enabled by the BIOS.
  
• PR 876212: If the metadata associated with slowpath packet is copied without checking whether enough data is mapped, then the metadata moves into an unmapped area resulting in a page fault.
  Mapping necessary data to include the metadata before copying it solves the issue.
  
• PR 876555: When you edit the virtual machine config options and set the virtual device bios filename to an empty string through the API, the virtual machine then fails to power on.
  
• PR 877568: When you set the iSCSI initiator Login timeout value it has to be inherited for the discovery phase, this does not happen in ESX 5.0 U1.
  
• PR 884739: When the user attempts to upgrade ESX 4.1 Classic to ESXi 5.0 which has extended partitions, the ESX host fails to detect the extended partitions and might not upgrade it. Including pre-check scripts to detect extended partitions solves this issue.
  
• PR 886880: When you add a virtual disk to a virtual machine that resides on a Storage DRS enabled datastore and if the size of the virtual disk is greater than the free space available in the datastore, SDRS might migrate another virtual machine out of the datastore to allow sufficient free space for adding the virtual disk. Storage vMotion operation completes but the subsequent addition of virtual disk to the virtual machine might fail and an error message similar to the following might be displayed:
  Insufficient Disk Space
  
• PR 888067: When using the Multiple-NIC vMotion feature with vSphere 5, if vMotions operations continue for a long time, unicast flooding is observed on all interfaces of the physical switch. Approximately 5 minutes after vMotion starts, the source and destination host start receiving high amounts of network traffic.
  
• PR 888095: IBM ULTRIUM-HH5 tape device does not support SCSI commands with ordered set attribute and might fail with a SCSI device status
  Check Condition and sense key 0x05/0x49/0x00 INVALID MESSAGE ERROR when SCSI requests with ordered set attribute are sent to the device.
  
• PR 890932: When you install Solaris 11 GA on ESX host from the virtual machine console and configure it with a LSI logic parallel controller, the ESX host always displays the following warning messages:
  unknown ioc_status = 4
  incomplete write – giving up
  
• PR 891155: If you attempt to load CBRC module in ESX which has limited amount of memory, the ESX host might fail with a purple diagnostic screen. When two functions attempt to change the value of a memory counter at the same time instance this issue occurs.
  
• PR 892131: When you apply a host profile that modifies only the MTU value for a standard vSwitch, the new MTU configuration is not applied on vSwitches of the new destination host.
  
• PR 892904: If a file metadata is corrupted on a VMFS-3 Volume, ESXi host might fail with a purple diagnostic screen while trying to access such a file. VMFS file corruption are extremely rare but can be caused by external storage issues.
  
• PR 896075: After moving one or more files into a directory, an attempt to delete the directory or any of the files in directory might fail. In such a case, the vmkernel.log file will contain entries similar to the following :
  2012-06-25T21:03:29.940Z cpu4:373534)WARNING: Fil3: 13291: newLength 85120 but zla 2
  2012-06-25T21:03:29.940Z cpu4:373534)Fil3: 7752: Corrupt file length 85120, on FD <281, 106>, not truncating
  
• PR 898200: Host profiles check compliance against the speed and duplex of a physical NIC. If the speed and duplex of a physical NIC of a ESXi host matches that of the Host profile, the ESXi host shows as compliant, even if the physical NIC is set to Auto-negotiate and the Host profile is set to Fixed. Also, physical NICs set to auto-negotiate cannot be changed to Fixed by using Host profiles if the speed and duplex settings of the ESXi host and host profile is the same.
  
• PR 899608: When you execute the commands esxcfg-info, esxcfg-resgrp -l, or vm-support on an ESXi host by using ESXi Shell or SSH, error messages similar to the following might be logged in the syslog.log file:
  2012-08-01T09:40:12Z esxcfg-info: ResourceGroup: Skipping CPU times for : Vcpu Id 11777 Times due to error. max # of processors: 4 < 11777
  2012-07-03T00:57:28Z esxcfg-resgrp: ResourceGroup: Skipping CPU times for : Vcpu Id 55340 Times due to error. max # of processors: 1 < 55340
  
  
• PR 900618: A storage vMotion operation on ESXi 5.0 by default sets disk.enableUUID to true for a Windows 2008 virtual machine, thus enabling application quiescing. Subsequent quiesce snapshot operation will fail till the virtual machine undergoes a power cycle.
  
• PR 909891: After completing a storage VMotion operation, ESXi disconnects the mirror device it created for the operation. However, under certain erroneous conditions, it might reference an uninitialized pointer which might result in ESXi host failing with a purple diagnostic screen.
  
• PR 912768: Description: A stateless ESXi host might have the PartnerSupported acceptance level even if the image profile that it boots from does not have the PartnerSupported acceptance level.
  
• PR 914442: This release adds log entries to check the maximum number of open file descriptors of sfcb processes.
  You can check the limits for open file descriptors in the cim logs, by performing the following steps:
  1. Set the CIM log level to 6 by using the # esxcfg-advcfg -s 6 /UserVars/CIMLogLevel command 2. Restart the sfcbd service by using the # /etc/init.d/sfcbd-watchdog restart command.
  2. Verify that log file in the folder /var/log/messages contains entries similar to the following for the maximum limits for open file descriptor:
     sfcb-HTTP-Daemon[30847]: --- Limit of maximum open file descriptors: soft Limit - 512 Hard Limit - 1024
     
• PR 921357: Guest Timer Calibration for Solaris 10 U10 is incorrect, including support for Guest Timer Calibration solves this issue.
  
• PR 923155: If an independent disk is deleted from a virtual machine, attempts to create a quiesced snapshot of a virtual machine might fail as the disk mode data for a given SCSI node might be stale. Error message similar to the following might be displayed:
  Status: An error occurred while quiescing the virtual machine. See the virtual machine's event log for details.
  The log files might contain entries similar to the following:
  ToolsBackup: changing quiesce state: STARTED -> DONE
  SnapshotVMXTakeSnapshotComplete done with snapshot 'back': 0
  SnapshotVMXTakeSnapshotComplete: Snapshot 0 failed: Failed to quiesce the virtual machine. (40).
  
  
• PR 926779: When the ESXi host attempts to create a world group heap without releasing the memory associated with world slot the process fails due to a world slot memory leak.
  
• PR 938082: Attempts to PXE boot virtual machines that use the VMXNet3 network adapter, by using the Microsoft Windows Deployment Services (WDS) might fail with messages similar to the following:
  Windows failed to start. A recent hardware or software change might be the cause. To fix the problem:
  1. insert you Windows installation disc and restart your computer.
  2. Choose your language setting, and then click "Next."
  3. Click "Repair your computer."
  If you do not have the disc, contact your system adminsitrator or computer manufacturer for assistance.
  
  Status: 0xc0000001
  
  Info: The boot selection failed because a required device is inaccessible.
  
• PR 940124: When a host with SMBIOS 2.6 or higher is upgraded from ESXi 5.0 to ESXi 5.1, virtual machines ask for information on whether the virtual machine is moved or copied when you power them on. Message displayed is similar to the following:
  Virtual Machine Message
  This virtual machine might have been moved or copied. In order to configure certain management and networking features, VMWare ESX needs to know if this virtual machine was moved or copied. If you don't know, answer "I...
  • Cancel
  • I moved it
  • I copied it