Our company is using DL160 G6 servers. I wanted to know if it is possible to install IPMI or a similar technology? So far we have no possible way to reboot when the system freezes, go into BIOS etc. We would really love to have this option because the servers aren't right around the corner.

We have a port at the back of the server that sais MGMT but is not usable. I was wondering if we could upgrade this now or what other possibilites we have? Thank you so much in advance for any help! Sincerely Silvio Brossi. I have total problem with networking.

Driver Downloads Marvell Extranet Contact Support All Support. About Marvell Newsroom Marvell Blog Investor Relations Events Careers Contact Us Awards Trademarks Communications Policy Corporate Offices Sales Offices Directions Industry Associations Marvell Leadership Company.

Have HP DL160G6 with ESXi4 build 175625 onboard. 2 VMs with FreeBSD7.2. And time to time these VMs are stop to ping. I can ping host machine, i can connect to management with Client. But i cant go into console of any VMs from Client, i cant even reboot or shutdown VMs. Only Hard reset of host machine helps. After Hard reset i can start 2 of my VMs and they works fine about a day till next hang up.

ESXi starts from flash card. 2 months everything works fine, but now - almost every day i must do hard reboot of my host.

Anybody has such a problem? Now in process of retire my serv from hoster. Will have it in my hands tomorrow. Now download VM's to set them up on another hardware. Last HP diagnostics check show 'everything fine'. Server have last firmware on all possible components.

I added all necessary patthes to ESXi4, still bad. Today at night all VM's again become disconnected from the net. Client showns no errors. Ill try to see messages.log of ESXi when next hangup happens. Any VM do notging special or resource-critical during hangup. Still do not know what to do.

I will try to replace motherboart at first. If it happens - then it happens with chipset in my oppinion. Hello, I have installed Dell OpenManage 5.1 according to the guide from Dell. The guide tells you to install OpenIPMI 35.13 on ESX, but during the install I see messages telling me that IPMI 37 is already installed.

The install guide from Dell writes to install via install-force, but I'm not sure it is really replacing version 37 with version 35 and I'm actually not sure whether that is a bad thing at all. I have included some of the messages below. ========================================== Running the pre_install script: Module version 35.13 is not newer than 37 already found in the kernel. Pre_install failed, aborting install. You may override by specifying --force. Your DKMS tree now includes: openipmi, 35.13.RHEL3, 2.4.21-37.ELhugemem, i686: built openipmi, 35.13.RHEL3, 2.4.21-37.ELsmp, i686: built openipmi, 35.13.RHEL3, 2.4.21-37.EL, i686: built openipmi, 35.13.RHEL3, 2.4.21-37.EL, ia32e: built openipmi, 35.13.RHEL3, 2.4.21-4.ELhugemem, i686: built openipmi, 35.13.RHEL3, 2.4.21-4.ELsmp, i686: built openipmi, 35.13.RHEL3, 2.4.21-4.EL, i686: built openipmi, 35.13.RHEL3, 2.4.21-15.EL, ia32e: built openipmi, 35.13.RHEL3, 2.4.21-37.0.2.ELvmnix, i686: built Running module version sanity check.

Running the pre_install script: Module version 35.13 is not newer than 37 already found in the kernel. ========================================== Does anyone know whether or not I have downgraded my IPMI installation because I used install-force? It doesn't look like it downgraded IPMI, but I want to know for sure. How can I check the IPMI version currently used? Guess there should be no issues running Dell OpenManage with a higher version of IPMI, but maybe I'm an optimist. Kind regards, Jos Rossiau. BrunoLagos, Welcome to the Dell Community Forums.

There is software out there that supports some of the functions of IPMI on the Optiplex 620. Dell has it's own software called Dell Open Manage, which has a server side and client side component to manage systems on a network. The software works with other management applications as well as it's on client and server software. These applications help monitor the health of the PC's and allow remote startup and shutdown, as well as remote updates to be pushed. For more information you would need to contact your Dell account manager.

Message Edited by DELL-Elsbeth on 08:35 AM. I'm trying to get ESXi installed on a system with a SuperMicro X9SBAA-F motherboard. I've overcome the 'no keyboard' issue (due to all ports being USB 3.0) by installing over IPMI (also tried PXE) and using ks.cfg. However, I can never get ESXi installer to see my drive (WD RED 1.0TB) via the onboard 88SE9230 controller. I found a Japanese website which had a VIB that supposedly helps that and similar Marvell controllers, I rebuilt the ISO but still no luck.

The Chipset is over a year old, but surprised that it's not yet supported. Either that or I am missing something. Any suggestions to get the installer to see the drive are appreciated! Hi, I had a few X6250 about 4 months ago for TryBuy and SOL over IPMI was working great on them.

I just got a complete new set of 4 X6250 and X6450 and been have SOL problems from day one. Basically, i can work with IPMI no problem, but SOL just don't work. My question in general is where to find current data on the use of IPMI v1.5 remotely and a description of the packets sent over the network with respect to Traps from sensors. We need script data from UCS - C traps and as the headers are over 100 bytes in length with GUID and other connections data, I was trying to find a good description as how I can use varbind to pull specific data from them.

Our Application sits on a platform that does not provide us the access to snmptrapd or the snmptt daemons. Our Perl script will attempt to use IPMI and network connectivity to the CIMC to collect data and receive traps as we are caught with limitations on how we do this. My ask is for some direction to better understand the data transferred and how we can script out what's needed. Any details on the transmission of data here would be super.

Some details on how the traps are built and sent. When working locally we use the CIMC and its a great tool. All support is greatly appreciated. Gavin Had a spell error and fixed sorry. My question in general is where to find current data on the use of IPMI v1.5 remotely and a description of the packets sent over the network with respect to Traps from sensors. We need script data from UCS - C traps and as the headers are over 100 bytes in length with GUID and other connections data, I was trying to find a good description as how I can use varbind to pull specific data from them.

Our Application sits on a platform that does not provide us the access to snmptrapd or the snmptt daemons. Our Perl script will attempt to use IPMI and network connectivity to the CIMC to collect data and receive traps as we are caught with limitations on how we do this. My ask is for some direction to better understand the data transferred and how we can script out what's needed. Any details on the transmission of data here would be super. Some details on how the traps are built and sent. When working locally we use the CIMC and its a great tool. All support is greatly appreciated.

Gavin Had a spell error and fixed sorry. Hi sbs 2008 standard on dell T710. We are getting the following event; Log Name: System Source: IPMIDRV Date: 21:31:30 Event ID: 1004 Task Category: None Level: Warning Keywords: Classic User: N/A Description: The IPMI device driver attempted to communicate with the IPMI BMC device during normal operation.

However the communication failed due to a timeout. You can increase the timeouts associated with the IPMI device driver.

When I look under Device Manager-System Devices the Microsoft Generic IPMI Compliant Device entry seems to be fine ie there is no exclamation mark on it. What can I do from here to fix this issue? Many Thanks Regards. Are you getting a lot of these timeouts, or just a few sporadic ones? Probably want to check into the Dell knowledge base or contact their support resources if this is a persistance issue. This MS KB can give you some background and timeout adjustments if it is just a periodic nuisance. /kj Please remember to click “Mark as Answer” on the post that helps you, and to click “Unmark as Answer” if a marked post does not actually answer your question.

This can be beneficial to other community members reading the thread. Hello Tonight I gracefully shutdown all my VMs and gracefully rebooted the server with the ESX 3.5 CD in.

I upgraded succesfully to ESX 3.5 everything went fine. On boot I have 1 failed on ipmi. Seems to be ram according to some posts ive read. But this is not the problem since the entire system was fine before the upgrade.

After the upgrade when starting the VMs it starts fine it powers on then I see the windows logo and right after it goes to a black screen and nothing else. On a second attempt I tried to copy it from an exist ESX 3.5 using the converter and I get the error that the converter is unable to determine the Guest host OS. Are you aware of any problems like this? Have you ever experienced such an issue? Ive noticed that 1 VM powers on and boots fine but seems to be something wrong with it also when I go to the Event logs for example there was no logs written the last log is when I gracefully shutdown the system.

Let me know what you think could be the issue or if anything can be done to fix this situation Thank you. Hi there, I keep getting warning messages with CIM as below. And this is a problem because we use a python script for hardware monitoring. WARNING: IPMI SEL - Server: HP ProLiant ML350p Gen8 s/n. I know there is a ton of posts on clearning these using vCenter but cannot find how to do it using the CLI or the vSphere client. I believe it's something to do with either /var/log/ipmi/0/sel and/or /var/log/ipmi/0/sel.raw.

Both of these are only about 3k in size though. Can someone please let me know how to do this to prevent the error coming up? ESXi 5.0.0 623860 HP ML350p G8 Thanks for any help.

Hi there, I keep getting warning messages with CIM as below. And this is a problem because we use a python script for hardware monitoring. WARNING: IPMI SEL - Server: HP ProLiant ML350p Gen8 s/n. I know there is a ton of posts on clearning these using vCenter but cannot find how to do it using the CLI or the vSphere client. I believe it's something to do with either /var/log/ipmi/0/sel and/or /var/log/ipmi/0/sel.raw.

Both of these are only about 3k in size though. Can someone please let me know how to do this to prevent the error coming up? ESXi 5.0.0 623860 HP ML350p G8 Thanks for any help. Hi, I own a Linksys SRW2016 and i'm trying to configure some VLANs on it. The thing is the machines connected to this SRW2016 all have an IPMI interface which is a management interface for the server's BMC (this is supported on many recent servers). You may need to now a little about how this is done, because the server is catching packets to it's network interface and either redirecting it to the real interface or the BMC.

This means the network interface has 2 different mac adresses and each has it's own ip address. Which allows you to talk to the IPMI interface over the network. And this works fine. However with some servers you can assign a VLAN to this ipmi interface (with all Dell servers for example). And preferably I'd want to do this to divide up the traffic. But I'm absolutely unsure about how to configure this.

The way I see it packets from the OS are untagged, and those from the IPMI interface are tagged. But when I try to put the server's port in the additional VLAN I don't seem to be able to connect to it.

(I only created one additional VLAN for ipmi, it's ok for the OS to use the default) Does anyone have any experience with this kind of configuration? Or has any ideas on how to configure it? Note: i'm kind of new to vlan configuration and it's all a wee bit confusing for me, sorry about that:/ Message Edited by trebb on 02:48 AM. Hi, I own a Linksys SRW2016 and i'm trying to configure some VLANs on it. All Files Ftp Directory Vb Net Getsetting Visual Basic there. The thing is the machines connected to this SRW2016 all have an IPMI interface which is a management interface for the server's BMC (this is supported on many recent servers). You may need to now a little about how this is done, because the server is catching packets to it's network interface and either redirecting it to the real interface or the BMC.

This means the network interface has 2 different mac adresses and each has it's own ip address. Which allows you to talk to the IPMI interface over the network. And this works fine.

However with some servers you can assign a VLAN to this ipmi interface (with all Dell servers for example). And preferably I'd want to do this to divide up the traffic. But I'm absolutely unsure about how to configure this. The way I see it packets from the OS are untagged, and those from the IPMI interface are tagged. But when I try to put the server's port in the additional VLAN I don't seem to be able to connect to it. (I only created one additional VLAN for ipmi, it's ok for the OS to use the default) Does anyone have any experience with this kind of configuration?

Or has any ideas on how to configure it? Note: i'm kind of new to vlan configuration and it's all a wee bit confusing for me, sorry about that:/ Message Edited by trebb on 02:48 AM.

I have an IBM BladeCenter S chassis with three HS22 blade servers, each with embedded hypervisors that have been upgraded to ESXi 4.1, 260247. I'm running vCenter 4.1, 258902 When adding hosts to vCenter, each of them appears with a Red Alert, which seems to be related to these Events. Alarm 'Host IPMI System Event Log Status' on server changed from Grey to Red Alarm 'Host IPMI System Event Log Status' on server triggered an action Alarm 'Host IPMI System Event Log Status': an SNMP trap for entity server was sent How can I find out further information about what these messages mean? Below is the procedure to turn on / off promiscuous mode on the NICS. Question is, is there a query command I can run to determine what mode the NICS are in? The information below was located in the VMware support pages. My IBM eServer BladeCenter HS20-8843 is experiencing network connectivity problems.

The Serial Over LAN option is enabled in the Ethernet switch module (ESM). I see one of the following symptoms in ESX Server: I'm using a bond and some virtual machines lose network connectivity, either intermittently or persistently.

I'm using a bond and the service console loses network connectivity. I have dedicated one NIC port to the service console and the other to the virtual machines. The virtual machines cannot connect to the network.

Note: On HS20-8843 blades, this option is enabled by default. What causes these problems? The physical Ethernet adapter that is bound to the virtual switch needs to stay in promiscuous mode for proper network connectivity. One port of the built-in, dual-port Broadcom BCM5700 Ethernet adapter on HS20 blades may not stay reliably in promiscuous mode, under some circumstances. On HS20-8843 blades, this happens when the Serial Over LAN option is enabled. You may observe the above symtoms on other blade makes and models. To resolve the problem, disable IPMI on the Broadcom chip.

To do this, boot the server using a DOS floppy and run the Broadcom diagnostic utility, b57udiag.exe, with the following parameters: b57udiag -c 0 -ipmi 0 b57udiag -c 1 -ipmi 0 In the above example, -c indicates the device to be modified. 0 and 1 represent the first and second physical Ethernet adapters.

The -ipmi 0 attribute disables IPMI (1 enables it). Hi, I'm trying to install ESXi 4. The install 'hangs' while loading the ipmi_si_drv, but unlike other people having problems with this it does continue after 5-10 min.

Then a black screen with errors messages in red and then a greyed out install screen. Problem is the errors are displaying so fast nobody can read them.

The install screen is also disabled/greyed out so I can't get anywhere from there: Enter (install), Esc (cancel), R (repair). So my question is: How do I log the errors or get the ESXi to halt and display the error to me. It doesn't necessarily have to be IPMI that is the problem, although I think it is. Is there also a way to disable the loading of the ipmi_si_drv? I wanted to disable IPMI, but I can't do this through the BIOS. Other threads suggested upgrading the BIOS (the server was bought in 07/08). Although this will be my last resort.

Thanks in advance! Hi, we will use IPMI suspend and wakeup or Hosts. With IPMI Tool it works fine but when i will setup the settings in vCenter on Host - configuration - powermanagement there is a timeout. I've looked at the vCenter Database, there is a table: [VPX_HOST] with columns:,[IPMI_IP_ADDRESS],[IPMI_MAC_ADDRESS],[IPMI_USER_NAME],[IPMI_PASSWORD] there are no entries in this fields.

When i add there some informations the are gone after restart of vCenter. Where does the vCenter store the IPMI settings? Any other ideas? Hi, we will use IPMI suspend and wakeup or Hosts. With IPMI Tool it works fine but when i will setup the settings in vCenter on Host - configuration - powermanagement there is a timeout. I've looked at the vCenter Database, there is a table: [VPX_HOST] with columns:,[IPMI_IP_ADDRESS],[IPMI_MAC_ADDRESS],[IPMI_USER_NAME],[IPMI_PASSWORD] there are no entries in this fields. When i add there some informations the are gone after restart of vCenter.

Where does the vCenter store the IPMI settings? Any other ideas? Hello Guys: I have code problem. I use OS:Windows EmbeddedStandard7 Hardware have BMC chip, But i write code use call WMI IPMI class, but not find the name IPMIPrv(Reference:, and i checkDevice Manager os find Microsoft Generic IPMI Compliant Device, so why not findIPMIPrv,or what am I missing action?

Or whether os(win 7) not support IPMI? Please help mesolve my problem.

By Jessie Chen. Hi all, I have set up the ILOM alerts with SNMP traps (v2c,public) according to the ILOM user guide and pressed the 'send test alerts'. Nothing shows up in the SNMP trap monitor (using MG-SOFT software). When I add another alert of IPMI pet type (same destination ip as of SNMP trap alert) I see events in the trap monitor. It looks like the ILOM does not generate SNMP traps but only IPMI traps. In the ILOM user guide there is an explicit example of traps sent from the ILOM (in addition to the fact that the ILOM features include SNMP traps AS WELL as IPMI) so it seems this is possible.

My ILOM version is SP Firmware Version2.0.2.6 Does anyone have any idea of any step I have missed? Thanks in advance. Hi all, I have set up the ILOM alerts with SNMP traps (v2c,public) according to the ILOM user guide and pressed the 'send test alerts'. Nothing shows up in the SNMP trap monitor (using MG-SOFT software). When I add another alert of IPMI pet type (same destination ip as of SNMP trap alert) I see events in the trap monitor.

It looks like the ILOM does not generate SNMP traps but only IPMI traps. In the ILOM user guide there is an explicit example of traps sent from the ILOM (in addition to the fact that the ILOM features include SNMP traps AS WELL as IPMI) so it seems this is possible. My ILOM version is SP Firmware Version2.0.2.6 Does anyone have any idea of any step I have missed? Thanks in advance. I'm not having issues with the server except if i only have the system plugged in using power supply 2 then the server won't boot past the RAID boot. So i ran the diagnostics and this is what i get: IPMI - IPMI Firmware: Firmware Rev=0x01.74, Basebrd Mgt Ctlr.

Firmware Rev=0x01.01, Primary BP. Firmware Rev=0x01.01, Peripheral BP. IPMI - IPMI Firmware: Pass IPMI - IPMI System Event Log Check: Fail Result Code: 2900-0221 Msg:: Power supply sensor (Power Supply1) Power Supply A/C lost. IPMI - IPMI Sensor Data Record Check: Pass any thoughts? I'm not having issues with the server except if i only have the system plugged in using power supply 2 then the server won't boot past the RAID boot. So i ran the diagnostics and this is what i get: IPMI - IPMI Firmware: Firmware Rev=0x01.74, Basebrd Mgt Ctlr. Firmware Rev=0x01.01, Primary BP.

Firmware Rev=0x01.01, Peripheral BP. IPMI - IPMI Firmware: Pass IPMI - IPMI System Event Log Check: Fail Result Code: 2900-0221 Msg:: Power supply sensor (Power Supply1) Power Supply A/C lost. IPMI - IPMI Sensor Data Record Check: Pass any thoughts? Hi, got a modest 32 node cluster here and have i) added ipmi tools to the headnode ii) ManagementIpAddress using PS to the nodes iii) edited the ccppower.cmd script we can now power up nodes using the PS Start-HpcNode command quite happily However using the Cluster Manager UI the Start node action is still greyed out in the actions panel and is not even listed if I right click on a node as it does on my HPC2008 clusters.

Have I missed something or has this functionality changed with R2 or the R2 SP1? Hey, during setup of grid 11.0.2 I choosed not to used ipmi configuration. I am not sure, that I do need to setup the ipmi configuration now? How does the node fencing take place, without configuring the ipmi configuration? Will it work as in clusterware 10? What are the risks, if I do not configure IPMI?

If I do need the ipmi - is it necessary, that my linux machine is able to 'ping' (reach) the ip of the BMC? Is the ipmi driver using the network (switch) or does it communicate via internals on the motherboard? Hi Christian, you can configure IPMI later with the help of crsctl. Node fencing without IPMI works the same like in 10g. Node is evicted with the help of the voting disks.

If IPMI is configured then first it is tried to reboot the node via. IPMI, however all other mechanism stays in place (in the case the IPMI reboot should fail). IPMI has 2 advantages: a.) A little faster and you have a protocol of the reboot. B.) It works in all cases. I had 1 corner cases, where the system got stuck in a kernel I/O call (which was a kernel bug).

Due to this the node never could read from voting disks to get rebooted. In this case IPMI would have helped, and the node would have been rebooted. Risk: None I am aware of. IPMI: As said above not necessary (since not all servers have IPMI cards), but recommended. If your server have it, I personally would configure it.

And yes, the linux servers have to be able to communicate with the IP of the BMC (since the reboot is initiated by the 'other' node). Regards Sebastian.

This was a quite nasty case:user had locked himself out in lockdown-mode while trying to disable the IPMI loading. The VMs were powered off - no access of any type: no ssh, no viclient, no local console. Nothing.Trying to boot from ESX 4.1 i from stick got a PSOD because the same version was already instaled.

At that time I heard from the case in the german forum. I booted ESXi 4.0 from stick, fixed the ipmi loading problem and used Knoppix to remove the old installation and rebuild the RAID The original install and the ESX4.1i - stick and the ESX 4.0i-stick were all normal downloads - no OEM versions. Hello,I decided to open a new thread on this, as now I have a better understanding of what I need, but still confused of whether BMC can provide me with the full functionality of the requirements I have.What I am trying to do: I have a Linux client which sends PET SNMP traps via IPMI (Intelligent Platform Management Interface) protocol. I would like to have those PET SNMP traps arive as events to the BMC Impact Explorer (events console).What do I have so far: - A Linux machine with IPMP eabled. - BMC IM 3.5 installed. - BMC II for PATROL installed.What am I missing / confused about:1. It seems that the IPMI sends SNMP traps to the server on which the BMC II for PATROL agent is installed, but there is nothing on that server that listen to these traps.

Am I missing a component that listen to the SNMP traps and sends them over to the BMC II for PATROL? I thought that the IPMI should send the traps directly to the BMC II for PATROL. Or am I doing something wrong with the configuration of the BMC II for PATROL.?2. Is this the right way to integrate the IPMI with the BMC event console?Since we are talking about this issue for a long time, is it possible to send me a phone number of someone who is familiar with this area, so we can talk over the phone? It seems that talking on the phone will be much more efficient.Thanks! Hello,I decided to open a new thread on this, as now I have a better understanding of what I need, but still confused of whether BMC can provide me with the full functionality of the requirements I have.What I am trying to do: I have a Linux client which sends PET SNMP traps via IPMI (Intelligent Platform Management Interface) protocol. I would like to have those PET SNMP traps arive as events to the BMC Impact Explorer (events console).What do I have so far: - A Linux machine with IPMP eabled.

- BMC IM 3.5 installed. - BMC II for PATROL installed.What am I missing / confused about:1. It seems that the IPMI sends SNMP traps to the server on which the BMC II for PATROL agent is installed, but there is nothing on that server that listen to these traps. Am I missing a component that listen to the SNMP traps and sends them over to the BMC II for PATROL? I thought that the IPMI should send the traps directly to the BMC II for PATROL.

Or am I doing something wrong with the configuration of the BMC II for PATROL.?2. Is this the right way to integrate the IPMI with the BMC event console?Since we are talking about this issue for a long time, is it possible to send me a phone number of someone who is familiar with this area, so we can talk over the phone? It seems that talking on the phone will be much more efficient.Thanks! Device manager has an entry for the MS generic IPMI compliant device: Driver Date:, Driver Version: 6.1.7601.17514, Resources reports an I/O range and no conflicts. Although the general tab device status says: No drivers are installed for this device.

Running some test vbs scripts via WMI and WinRM show neitherthe IPMI provider or the Microsoft_IPMI driver are installed. There is no root/hardware namespace entry when enumerating namespaces, nor when searching registry. BMC is working via LAN access, and reports KCS interface is functioning. Reinstalling / updating IPMI driver always results in same driver being installed (as above). This article (suggests the IPMI driver expects the BIOS to expose BMC device info through the ACPI namespace/interface. So, is this a problem with the IPMI driver, or the ACPI, or something else? Any ideas on how to fix this?

Thank you, Darren. Hi, with our VMware Farm we have on the hostsystems performance problems.

Now we found on 2 hosts (running Vsphere 4.1 on IBM X3850M2) that the baseboard management controller had a failure. After solving this, its getting better. But my boss thinks there are more issues, and he wants me to disable IPMI on the hosts, because he thinks that the IPMI is also a possible reason for the performance.

Now my 2 questions: 1.st can IPMI bee a performance problem? 2.nd if yes, how can i disable it on Vsphere, as i know in ESX3.5 it was possible over the gui, now i cant find an optionin the VIClient or the command line - somebody a clue where to find it?

Tnx in advance Torsten. I think this post could be of help. Alex On ther performance side.I was looking at this presentation from intel 'and thought about that basically IPMI should not interfere with performance since it is a 'simple' monitor systems, but as from the latest specification it become more and more integrated with CIM and so, consequentially, 'active' in the process of manage the hardware system. Let see, for example, at the ability to mange power consuption through Energy Cap or RAM automatic throtthling, this IMO means that a un-correct configuration of the parameters could lead to performance issue to some hardware.in this case ESX. But as said this is just my thought and opinion.

I just got the PowerEdge 850 with BMC IPMI over LAN functionality. Does the BMC work only in a LAN enviroment, using private IPs? I tested it to work well in my home network, but once I got the system set up at the datacentre, and the BMC configured to use a public IP, as well as my provider's assigned netmask and gateway, I found that I was unable to access it from home using the supplied IPMI Shell CLI program. Would it be the same case for the optional DRAC4 card, or would that be accessible over WAN as well? Virtual media allows you to call up the DRAC 4 web interface and connect to a CDROM or floppy on the workstation as if it were on the server. Of course when dealing with an OS install some driver probe might do something unexpected. I will check on who may have actually done this and see what the results were.

The remote console on the DRAC is another feature.You get a server OS independent remotevideo/keyboard. The DRAC 4 user guide is found here: It will describe the DRAC 4 features more accurately than off the top of my head. I can add or find clarification for any questions. Message Edited by alfman850 on 06:51 PM. HP Integrity rx2660 server has problems with BMC/IPMC.

Hi Apobereyko, Try the following: Try to reset the BMC via the 'rb' command from the MPCM prompt.Reset the system via 'tc' command from the MPCM prompt.Reset the system via 'rs' command from the MPCM prompt.Pull a/c power cords for 1 minute and then reapply to system. MP will post for 30 seconds and then re-establish the connection into the MP. Now check theVFP output as well as sysrev output and the revisions should be present. Now boot the system. Regards, Rianui.

I have a WDM driver project in a ClearCase (source control/versioning file system) view(volume) that needs to run on a build server using the same view(volume). When I run the msbuild SDV command, it runs to completion on my desktop system, but seems to not really run on the build server system. I conclude this is because it takes 13hr 28min to run on my desktop but only 13min4sec to run on the build server.

The SDV summary is also very different in the two cases. I deleted the driver local sdv directory to try and force a full SDV run, but it still only took the 13minutes to run. I suspect that there are view private files, i.e files that are only visible to my desktop system that SDV needs that accounts for this behavior.

There are a number of SDV files and I need to only add the files to source control that are only read by SDV. (Note that when I copy the entire volume to a server system SDV does a full run as expected. So there must be some additional sdv-needed files. The possible files Isee are: runsdvui.cmd sdv-user.sdv I expect that I should not add these files as they seem user/logon specific. A staticdv.exe /clean is done first in both desktop and build server cases. The summary of the command output from the server build is: -------------- Build Server Outout ---------------- SDV checked 26 properties(s).

SDV performed 156 check(s) with: 138 Not Applicable 44 Other Run this command to view the results:msbuild driver project file /p:configuration=release configuration /p:platform=Win32 x64 arm /t:Sdv /p:inputs=/view EXEC: warning: 44 check(s) did not produce result(s). [p: h_w2k_platform driver ipmi build_ipmi.vcxproj] Merging results Nothing to merge for p: h_w2k_platform driver ipmi SDV exit code: 0 Done Building Project p: h_w2k_platform driver ipmi build_ipmi.vcxproj (sdv target(s)). Build succeeded. P: h_w2k_platform driver ipmi build_ipmi.vcxproj (sdv target) (1) - (Sdv target) - EXEC: warning: 44 check(s) did not produce result(s). [p: h_w2k_platform driver ipmi build_ipmi.vcxproj] 1 Warning(s) 0 Error(s) Time Elapsed 00:13:14.72 -------------- End Build Server Outout ---------------- A summary of the build from a desktop build is: --------------- Desktop Build Output ---------------- Check 'ipmi' for 'forwardedatbadirql'.Done SDV checked 26 properties(s). SDV performed 156 check(s) with: 1 Defect(s) 2 Warning(s) 48 Rule Passes 80 Not Applicable 24 Timeout 1 Other Run this command to view the results:msbuild driver project file /p:configu ration=release configuration /p:platform=Win32 x64 arm /t:Sdv /p:inputs =/view EXEC: warning: 25 check(s) did not produce result(s). [I: h_w2k_platform driv er ipmi build_ipmi.vcxproj] EXEC: warning: 13 check(s) should use the following workaround.

[I: h_w2k_pla tform driver ipmi build_ipmi.vcxproj] Workaround: msbuild driver project file /p:configuration=release configur ation /p:platform=Win32 x64 arm /t:Sdv /p:inputs=/refine Merging results Nothing to merge for I: h_w2k_platform driver ipmi SDV exit code: 0 Done Building Project I: h_w2k_platform driver ipmi build_ipmi.vcxproj (sdv t arget(s)). Build succeeded. I: h_w2k_platform driver ipmi build_ipmi.vcxproj (sdv target) (1) - (Sdv target) - EXEC: warning: 25 check(s) did not produce result(s). [I: h_w2k_platform dr iver ipmi build_ipmi.vcxproj] EXEC: warning: 13 check(s) should use the following workaround. [I: h_w2k_p latform driver ipmi build_ipmi.vcxproj] 2 Warning(s) 0 Error(s) Time Elapsed 13:27:51.50 --------------- End Desktop Build Output ---------------- Any idea why these two builds are behaving differently? Yes if you are missing libraries this will most definitely will cause more Not Applicable results. Yes the SDV Controller in VS 2012 will try to write to this file.

If you make your file writable then the controller will overwrite the thread property which you have set to 0 and thus you will not be able to share this file on machines with a different number of logical processors. For a workaround I would suggesthaving a script that copies the sdv-default.xml file from a share not the driver folder and then runs StaticDV on the command line. We will investigate this issue in the Windows Next timeframe. Thanks for the feedback. I am trying to upgrade a test server (Supermicro 6015P-TR) to ESXi 5 from ESXi 4.1 and the reboot fails with it hanging at 'Initializing ACPI.' If I reset the IPMI card it passes this point. The upgrade then completed.

But each reboot presents the same problem. Anyone else successfully using ESXi on Supermicro 6015P-tr or X7DBP boards with a IPMI controller? Wondering if this could be a problem with the IPMI card or if it is more widespread. Motherboard BIOS and IPMI firmware are latest versions. Have tried cold booting, reset BIOS defaults, changing drive from AHCI to legacy compatible. I am trying to upgrade a test server (Supermicro 6015P-TR) to ESXi 5 from ESXi 4.1 and the reboot fails with it hanging at 'Initializing ACPI.' If I reset the IPMI card it passes this point.

The upgrade then completed. But each reboot presents the same problem. Anyone else successfully using ESXi on Supermicro 6015P-tr or X7DBP boards with a IPMI controller? Wondering if this could be a problem with the IPMI card or if it is more widespread. Motherboard BIOS and IPMI firmware are latest versions.

Have tried cold booting, reset BIOS defaults, changing drive from AHCI to legacy compatible. I used ITA to push firmware upgrades to 10 Dell PE 1950s running Windows 2003 server SP2 9 out of 10 completed perfectly and came back up fine after the reboot However, the 10th generated an error when trying to update the RAC 5 firmware: ERROR: IPMI is not available. Ensure that IPMI is enabled and the supported IPMI driver is installed I tried running just the update EXE within Windows but it failed with the same error again Note: The EXE I'm running is: RAC_FRMW_WIN_R175502.EXE This box is configured identically as my other 9 servers, both hardware and OS/software So I'm scratching my head as to why it's complaining about IPMI Note: I don't use IPMI for any management. I'm using SNMP only for hardware monitoring/alerts Any ideas?

This error has me worried a bit because I have approx 130 more servers to push firmware upgrades to, in the coming weeks. Thanks for any info/help -GreenHills.

Thanks for your reply Jeff Here are my replies to your post: The BIOS update completed successfully. I pushed it along with BMC/DRAC and other firmware as part of the 5.40 bundle. This same process worked fine on all my other systems I upgraded. Same thing with the ESM firmware (aka Dell SAS Backplane Firmware) It was part of my bundle I was pushing to the client machine. This upgrade completed successfully on this box and the others.

It's not clear how Windows updates would affect the RAC firmware process. Do you happen to have any references which show this connection? Ok I've searched the documentation, read about IPMI, looked into the DRAC/MC stuff, etc and for the life of me can't figure out if this is possible. We have a PowerEdge 1855 server with 10 blades. Currently we only have the KVM pass through module (it's a very basic module that lets you select the blade you want to work on by depressing the 'KVM' button on the front of the blade).

Does anyone know if you can select the KVM button remotely on a blade using IPMI, or the DRAC/MC tools? I haven't been able to find anything that might suggest you can do this.

But then again I'm not expert. Currently we have all the blades setup for console redirection over the DRAC/MC however that doesn't work so well when you want to install an OS. For example Linux. Once you start the kickstart. It sits on the video interface (assuming that's the console) and waits for input. If this doesn't make any sense at all please let me know -- I can post more details. I would love to have some command like: chassis kvmbutton enable Or something like that which would activate the KVM button on that particular blade (maybe it's in IPMI?).

Thanks in advance. I can tell you with very good confidence that you'll need an Analog or IP KVM module in your Blade chassis. We struggled for weeks with this and after we had purchased a KVM (for $600 I think) we installed all ten blades in under a day with Kickstart. It's a breeze once you can get control of the keyboard and mouse. Without that and only console redirection. You are hosed.

We also ran into problems with the IP Based KVM and using a RHEL 4 machine as the admin station (running the KVM viewing software). Seems the KVM that we got has a self signed cert that the JVM (1.5.0 update 3) doesn't like and won't work with javaws. Let me know if you need more details. Well, I got an ASUS KFN4-D16 (dual socket AMD Opteron board) and added an ASUS ASMB3 miniPCI IPMI addon card. The card (and its functions) did not show up in ESXi (there was some hope, that more sensors will show up).

But I can use it to boot/reboot/stop the server, watch fan/temp and I can use it for Serial Over LAN into BIOS (including updates). Unfortunately, this only covers BIOS and the very early stage of an ESXi installation. IMHO there is no serial support in ESXi. You might have more luck with your card. Ok, so 2 nights ago my Dell R200 server became unresponsive (failed to respond to ping or ssh). I power cycled it using IPMI, but it never came back online. Had the data center check it out and theycouldntget any type of display to come up when trying to connect to the front or back of the server and no diagnotic lights.

There is a single amber light, but nothing on the A, B, C, or D lights. I can connect through IPMI and get basic sensor info and confirm the chassis power is on. I am getting cpu fan speeds when i check the sensors and the DC has confirmed the fans are spinning up like normal when its turned on. They tried taking out ram to see if any of it was bad, but no difference. No beep codes either. I am 7 hours away, so its kind of hard for me to troubleshoot it and am searching for ideas.

Anyone have any ideas or suggestions? Its out of warranty, so unfortunately there isnt a simple/cheap way to just try out new parts. I already have the BMC users guide, and have my BMC set up.

What I need and cant find anywhere is User How To information on the IPMI utilities. Ive downloaded and installed the ipmiutil on two computers; one a Workstation joined to the servers domain and an independent computer on the network (not joined to the domain) but for the life of me I cant make them work. I need to be able to have the IPMI on the workstation start the server. Heres the scenario: My wife and I operate a (2 person) online background screening operation, which uses the server for, among other things, our databases.

We have to travel often on business, and while were away we operate via RDS/RDG to our two workstations which are joined to the servers domain. In the event of a power failure the workstations will automatically restart when power returns, but the server (a PowerEdge T110 II) doesnt seem to have that capability. We can, however, access one of our workstations via LogMeIn and restart the server via IPMI/BMC. Unfortunately weve yet been able to make the IPMI work, and thats where I need the help. Creacontech, I should be able to solve both issues you stated. First off, the T110II does have the ability to power on after a failure. If you access the BIOS (F2 at startup) then select System Security, from there set AC Power Recovery to ON.

The three states you have to choose from, and there meanings are Last - the system returns to the last power state. On - turns on the system after power is restored. Off - allows the system to remain off after power is restored. So you will need to set it to ON. That will allow the server to power back on, once power is restored. For the other issue, you can install OpenManage Server Administrator (OMSA) (64bit or 32bit depending on OS) on the server, as well as OpenManage Essentialson the workstation, and be able to monitor and manage the servers remotely from there. That includes the ability to configure it to send email alerts.

This video will show you some of the functions of Essentials. This link will take you to the User Guide for Essentials, as well as for OMSA. I hope this helps. Let me know if this helps answer your questions.

Hello, I got two PE860 with DRAC 4/P Cards. I connected them to ITA via SNMP and my colleague installed OMSA on its Suse Linux 10 OS. So ITA shows as Agentnames: DRAC, Inventory Collector Agent, SA and SA (Storage Management). But I am missing IPMI. During the servers BIOS setup we enabled IPMI over LAN with a acccountname and a password. But the ITA IPMI connection test fails. And IPMI settings were configured for the server range in ITA.

I thought I could configure IPMI parameters over the webinterface of the DRAC card (at least that worked with other DRAC5 cards). But the DRAC4 webinterface shows no options for IPMI. What am I missing? Must we install an IPMI client like Open IPMI on SuseLinux? IPMI goes OOB - out-of-band, that is, it bypasses the OS.

Some users like to maintain separate infrastructure for systems management, they choose separate dedicated monitoring using IPMI. OMSA allows you to do in-band management. It is more powerful, in terms that it provides lot more information and capabilities than IPMI alone. ITA supports SNMP, CIM and IPMI all for discovery and monitoring (with varying capabilities, depending on the choice of protocol). Message Edited by OMNavigator on 03:55 AM. Hello, I wan't to configure my Catalyst 3750.

My network is: 10.1.0.0 /16 There 4 different sections 1. 10.1.1.0 - 255 for Server's 2. 10.1.100.0 - 255 for IPMI Interface's 3.

10.1.2.0 - 255 for Client's (1) 4. Hi, How do you enable Wake-On-LAN support in vCenter? When I choose 'Enter Standby Mode' for a host in vCenter I get the error as per the attached JPG. The part I am most interested in is 'Or, configure vCenter to use Wake-On-LAN'. How is this done? I realise I could configure IPMI under Configuration, Power Management but I specifically want to test DPM using Wake-On-LAN only.

According to the vSphere documentation there are three ways to use DPM. HP iLO, IPMI and Wake-On-LAN.

The stuff that can be configured in the Power Management section is for iLO and IPMI. I specifically want to test it with boring old Wake-On-LAN - which used to work just fine using ESX 3.5 on exactly the same hardware without any configuration required. When I click on Configuration, Network Adapters all the network adapters in the server say they support Wake-On-LAN. What am I missing? Cheers, David. I have 2 motherboards (from 25), which log unknown error some times. I didn't find the description of it in various documents ('Alert Standard Format (ASF) Specification, ver.

2.0', 'IPMI PET to Platform Message Registry Mapping, ver. 1.2.0', 'IPMI Platform Event Trap Format Specification, ver. 1.0', 'IPMI Specification Second Generation, ver. 2.0', WSEventAndPetTableReplacement.xls). Motherboard: DQ965GF EventSensorType:Fh (POST Progress/Error) EventType:6Fh EventOffset:0h Severity: Non-recoverable Event Data: 40 14 00 00 00 00 00 00 Entity: 26 (Other chassis board) Can anyone give me explanation of what happens with these two m/boards and there I can get the list of Platform Event Codes. Filling in the Asset Tag value (and for that matter any other blank values in the FRU) have ZERO impact on this issue. Looks like the issue here is a bug in ESXi as it interacts with this motherboard.

I'd sure like to see this fixed, and it'll affect my purchase of vShere licensing, as we are evaluating the product set now. Disabling IPMI entirely in the VMWare configuration is presently the only way to make these platforms functional. That VMWare fails with cryptic messages about the disk being full, rather than properly handle the issue, gives an appearance of poor software quality, and questionable software quality assurance testing to the entire product.

Surely you can do better. I just got two brand-new R200 and first thing I did with both is go in the BMC/IPMI configuration and enable IPMI over LAN and configure an IP address.

However, I can't reach it. I can't even ping it. Both of them. LAN connection is fine as I can PXE boot the machines. Then I changed one of the BMCs to DHCP and, surprise, I see that it asks for an IP and gets one. Nevertheless, I can't reach it on that IP. No ping, no telnet, no http.

What's the deal here, what am I doing wrong? My experience with BMC is only with HP machines so far. There it works immediately after switching it on. I've got two datacentres, each has its own VirtualCentre server and cluster, all hosts are on esx4.1u1, vCenter servers are 4.1.

Both sites have all HP servers with ILO configured and working. On one site I can use the vSphere client to go to Configuration - Power Management - Properties and add the username, password, bmc ip and mac address. I can also use a powershell script to do this as part of my build. This has worked fine on at least 20 servers. On my other site when I try and add the details through the client or powershell I get the error 'a specified parameter was not correct'. Any ideas what the problem might be?

So, I have a new server with a Supermicro MB, and the BMC is set for failover, with no way to change this from the BIOS. This means that since no cable is plugged into the IPMI port, the BMC module listens on the standard ethernet connection and picks up an IP address from the DHCP server, creating a huge security problem. As far as I can tell, the only way to change this is using the ipmitool, which is available in the Community package repository; however I can find no documentation on how to use it under Arch, and the instructions for Debian dont work. In particular # modprobe ipmi_devinf # modprobe ipmi_si # ipmitool chassis status Could not open device at /dev/ipmi0 or /dev/ipmi/0 or /dev/ipmidev/0: No such file or directory. I found this answer on superuser.com: Yes, It is possible. You will have to plug in the DRAC Interface, and connect to it using a web browser on another machine.

You would do that from a PC thats on the same network as that DRAC interface. The DRAC NIC has a default IP address of 192.168.0.120, Gateway 192.168.0.1 and mask of 255.255.255.0 Once connected to that address, the username Root, password Calvin should work, if the settings are fresh. Once in, you should be able to do whatever you need using the Console If that IP doesnt come up for you, grab a free net scanning tool off the web and just look for a new address on your network. Is this correct? I know my server doesnt have seperate drac interface. I havent been home in a couple of days otherwise I would have checked it out myself. Also I dont really need to manage the server remotely just from within my home wired network.

Hi everybody, I have a PowerEdge T320 with iDRAC Basic Management. The installed OS is Ubuntu 14.04 64 bit (kernel 3.13.0-32-generic). My problem is that I want to monitor Power consumption. All guides says that iDRAC is IPMI2.0 compliant. So I installed ipmitool and freeipmi-tools, but when I run # ipmitool sdr list full the output is. Pwr Consumption 0 Watts ok.

When i run # ipmi-oem dellget-power-consumption-statistics average The output is: Last Minute Average Power: 0 W Last Hour Average Power: 0 W Last Day Average Power: 0 W Last Week Average Power: 0 W When I run # ipmi-oem dellget-power-consumption-data The output is: dell:get-power-consumption-data failed: Requested Sensor, data, or record not present. What is the problem? Is it really that Basic Management doesnt support power monitoring? Do I have to upgrade the license? Thank you in advance.

Hi, I have X4200M2 Server, which is going in standby mode. I had exactly the same issue with identical hardware ID's that were embedded on a Gateway Profile 6 main board. The drivers can be found @ Intel's site on the link provided below. This cleared up both the IPMI and missing PCI driver for Windows 7 x64. The drivers included in the pack cover both x86 X64 versions XP and up. Strangely enough the drivers are found in the Intel pro set Lan drivers pack.

Hi, Several sources mention that the Microsoft IPMI driver and the Intel IPMI driver should not be loaded at the same time. If both drivers are installed, resource contention problems result in unreliable operation. I wish to use the Intel driver, and have tried to uninstall the Microsoft IPMI driver (IPMIdrv.sys), but all my efforts have failed so far. The obvious way would seem to be tostart Device Manager,go to System Devices,right-click Microsoft Generic IPMI Compliant Device,select Properties, andon driver tab, click Uninstall.

When I do this, I do not get any error message, but when the machine is rebooted, Microsoft Generic IPMI Compliant Device reappears. I also tried using DevCon.exe, but I got the same result as earlier. I also tried some dirty tricks, but they did not work as well. I am aware I can choose to stop (disable via Device Manager) the Microsoft IPMI driver, but I really wish to uninstall it completely (the proof being that IPMIdrv.sys should not show up when driverquery command is run). Intel published a document titled Resolving IPMI Driver Conflicts in Windows Server 2003 R2, but as the name suggests, it is meant forServer 2003 R2, and the techniques mentioned do not work forServer 2008 R2. Any pointers will be greatly appreciated.

Hello I have a a few ESI5 running all well. I am trying to setup the IPMI/iLO and I am getting a little confused. I have looked online and I am sure I missing something simple. I can access Fujitsu Server View at 192.168.1.201 with a user name and password with no problem. I can power on the blade etc, no problem. When I enter in the user/pass/BMC IP amd BMC mac, I do so and then presented with this message: 'One or more IPMI/iLO parameters are incorrect.

Check that the username, password, MAC address, and IP address are valid for host xxxx, and that the BMC is correctly configured.' I am not sure what I am doing wrong. I have checked the BIOS on each blade and I cannot see anything about IMPI. I have a BX600 with 10 x BX620S Thank you.

I am trying to setup the Power Mangment portion of our VMware ESX4 servers running on a Dell R710 but not having any luck. ESX4 requires IPMI to be configured first before I can place the servers in standby and bring them back out. Setup Ports 1 and 2 are configured for the iDrac 6 express and are configured for the vMware vConsole. Also the vConsole and idrac6 are configured on the same vlan with trunking turned off, and are in the same subnet range. VCenter Server is running on Windows Server 2008 with the firewall off and on the same subnet as the vConsole and iDrac6 vCenter Server 4.0.0,162856 vSphere Client 4.0.0,162856 ESX 4.0.0,175625 I have attached a picture of the network configuration Issue I am able to successfully connect to the idrac interface from my machine, but i can not ping the idrac ip address from the ESX4 server. However, I can ping other idrac ip address that are on my other two ESX servers that I am having the same issue with. I can also succesfully query the IPMI on any of the servres from my machine.

When I attempt to configure the IPMI / iLO Settings i get the following 'The request failed because the remote server 'vcenter server name' took too long to respond. (The command has timed out as the remote server taking to long to respond.) Any help is appreciated, Thanks Dan. I was wondering if someone could point me in the right direction. I have a Dell 1650 that bluescreened one morning. The message was: ______________ ***the system has halted Please contact your hardware vendor.

____________ So then I ran system diags through Server assistant, and there were no errors. I then ran Dell diags through boot disks, version 5012 I believe. They returned the following errors: IPMI - IPMI System Event Log Check: Fail Result Code: 2900-0221 Msg: System Boot: Power supply sensor (Power Supply2) A/C lost. IPMI - IPMI System Event Log Check: Fail Result Code: 2900-0221 Msg: System Boot: Power supply sensor (Power Supply1) A/C lost. IPMI - IPMI System Event Log Check: Fail Result Code: 2900-0221 Msg: Jan 02 18:: Memory sensor (0x02) Uncorrectable ECC error Card 0 Bank 0. Any idea what this error means? I'm not sure what an IPMI system event log check is, but it soundsto me like this error will keep occurring as long as itexist inwhatever log is being checked, even if the hardware is good?

Am I off base with that? If there still is a hardware problem, I'm guessingit lies in the motherboard memory slots, as I have swapped memory and this error keeps occuring (unless of course this error is just stating that an error has happened in the past) Thanks in advance! Nathan PS - This happened 3-4 weeks ago, and there have been no more bluescreens or freezes since. I was wondering if someone could point me in the right direction.

I have a Dell 1650 that bluescreened one morning. The message was: ______________ ***the system has halted Please contact your hardware vendor. ____________ So then I ran system diags through Server assistant, and there were no errors. I then ran Dell diags through boot disks, version 5012 I believe. They returned the following errors: IPMI - IPMI System Event Log Check: Fail Result Code: 2900-0221 Msg: System Boot: Power supply sensor (Power Supply2) A/C lost. IPMI - IPMI System Event Log Check: Fail Result Code: 2900-0221 Msg: System Boot: Power supply sensor (Power Supply1) A/C lost.

IPMI - IPMI System Event Log Check: Fail Result Code: 2900-0221 Msg: Jan 02 18:: Memory sensor (0x02) Uncorrectable ECC error Card 0 Bank 0. Any idea what this error means?

I'm not sure what an IPMI system event log check is, but it soundsto me like this error will keep occurring as long as itexist inwhatever log is being checked, even if the hardware is good? Am I off base with that? If there still is a hardware problem, I'm guessingit lies in the motherboard memory slots, as I have swapped memory and this error keeps occuring (unless of course this error is just stating that an error has happened in the past) Thanks in advance! Nathan PS - This happened 3-4 weeks ago, and there have been no more bluescreens or freezes since. Hello, I have a problem with one of my WAVE-574. I get many of the following messages: WAAS-SYS-4-900000: IPMI message handler: BMC returned incorrect response, expected netfn b cmd 23, got netfn b cmd 2c I already figured out that it could be a problem with a managementblade in the vmware.

But I have no virtual blades configured. Please take a look at the following output: name#sh virtual-blade vmstat No virtual blade has been started yet. Does anyone have an idea what to do? Kind Regards Markus. Hi, I have created a Red Hat Enterprise Linux Server release 5.3 Beta (Tikanga) Virtual Machine on HP DL 380G5. I installed ipmitool-1.8.11 manually and when I tried to start the ipmi service, I am getting the following errors, code[root@xxxxxxx root]# /etc/init.d/ipmi start Starting ipmi drivers: [FAILED] [root@xxxxxxx root]# /etc/init.d/ipmi status ipmi_msghandler module not loaded.

Ipmi_si_drv module not loaded. Ipmi_devintf module not loaded. /dev/ipmi0 does not exist.

/code When I cheked in /var/log/messages Oct 6 17:10:29 xxxxxx kernel: IPMI System Interface driver. Oct 6 17:10:29 xxxxx kernel: ipmi_si: Unable to find any System Interface(s) Oct 6 17:36:35 xxxxx kernel: ipmi device interface # modprobe ipmi_si_drv FATAL: Module ipmi_si_drv not found. # uname -a Linux aspint7-sol.ind.hp.com 2.6.18-118.el5 #1 SMP Sat Oct 4 00:21:41 EDT 2008 i686 i686 i386 GNU/Linux ++Does that mean that my kernel does not support ipmi_si? I tried installing the IPMI on the ESX3.5 which has this virtual machine and it is working fine. But I am facing issues only when I try installing ipmi on virtual machine.

Hello, i have a problem with my Data Domain IPMI Interface. Firstly our Data Domain is just a Test-Box, so nothing special. But a Customer contacted me, because his Data Domain IPMI Interface was no longer accessable after an upgrade from DD OS 5.4.2.X to 5.5.0.8. So i thought i recreate that problem on our Data Domain (DD 2500, ver.

And I ran into the same problem as the customer described. The problem is as follows: With Webbrowser: Error: Failed to Connect SSH: connect to host 192.168.176.224 port 22: Connection refused Telnet with a bunch of ports: (Port 80) connect to address 192.168.176.224: Connection refused (Port 22) connect to address 192.168.176.224: Connection refused (Port 443) connect to address 192.168.176.224: Connection refused Ping to IPMI is possible. I already restartet the DD and I powerd it off and unplugged it from power as well. Nothing changed. The only thing that worked is, to connect to DD GUI with webbrowser, go to Maintenance-IPMI and choose 'Login to Remote System'. When I log in to the IPMI Interface, it works! But it is supposed to do so?

I mean, before the update everything was fine; I was able connect via SSH or Webbrowser to my IPMI and restart the box as I want! One strange thing to add: I can access a html-page on the IPMI over webbrowser. Its the index.html. So I can connect to 192.168.176.224/index.html but I can do absolutely nothing there except for navigating a little bit through the menues, but still can't log in there or power on/off the machine. I already talked to customer support but they won't confess a bug yet. I Can definetly say that there is no firewall between the machines, that could lead to that connection refused-statement. Again I testet it with only a consumer switch between the IPMI-Interface and a server; Same here with connection refused.

My thoughts on that are, that there is bug with IPMI by upgrading the system to 5.5.0.8. I think the IPMI shielded itself against everything and there is just one little option, that isn't restored after upgrade. While trying to install ESX 4.0ux the installer will hang for ~30 minutes at 52% (72.IPMI). I have found a few documents with possible solutions to them on the pre-install (noIPMIENabled, noACPI, etc.) but have been unable to prevent the IPMI driver from loading during the initial install.

I do have a script that runs that 'fixes' IPMI after the initial install but it's still not what I would like to have happen. Is there a%pre command I can run in a kickscript that will prevent it from attempting to load the driver? Any suggestions would be welcome.

While trying to install ESX 4.0ux the installer will hang for ~30 minutes at 52% (72.IPMI). I have found a few documents with possible solutions to them on the pre-install (noIPMIENabled, noACPI, etc.) but have been unable to prevent the IPMI driver from loading during the initial install. I do have a script that runs that 'fixes' IPMI after the initial install but it's still not what I would like to have happen. Is there a%pre command I can run in a kickscript that will prevent it from attempting to load the driver? Any suggestions would be welcome. I have a need to receive SNMP traps from random devices, some of which are not amenable to being added as a network platform because they can't be queried via SNMP.

Or if they can, they simply don't have the OIDs HQ wants to check for when adding the new platform.One example of such a device is the Sun iLOM on their x86 machines. IPMI isn't a suitable tool as they aren't all running Solaris and the contrib.

IPMI plugin appears to be meant for use with the local host -- we use some as VMware ESXi hosts, for example, where we can't really run a HQ agent. We need to receive traps from these as there doesn't appear to be any other way of being alerted to hardware failure.(There are other devices with similar issues on our network, receiving traps is a one-size-fits-all fix that works for us.)At the moment we're using What's Up Gold, and that seems happy receiving traps from devices without needing to be able to do more than ping them.Other than this we're pretty impressed with HQ, but if I can't solve this with HQ then my bosses aren't going to be willing to go with it.Matt. I eventually figured this out for myself: the newer Sun gear is using a completely different set of MIBs. Once I found the necessary OIDs I was able to make a new plugin to grab that data, and added the log_tracker plugin type to it so I can receive traps and alert on them.Of course it looks like every model -- and possibly every iLOM version or host!

-- will have these at different points in the MIB. Is there perhaps any way to do lookups within SNMP tables so I can just specify a key to find the right index -- for example, '1.3.6.1.4.1.42.2.70.101.1.1.8.1.4.x' will have the text value '/SYS/MB/T_AMB0', knowing 'x' points to the correct OID for the ambient temperature sensor, so if I could somehow tell HQ to do that lookup on each host rather than having to do it by hand and create a unique plugin each variant then that would be fantastic. Members, I have a Dell 1950 Type II server with a couple of 146GB SCSI SAS drives in it and I was reading that SATA II (up to 2TB) HDs can go into the 1950 Type II to replace the SAS drives. However, there is some confusion (on my part) as to if the drive sleds I have need an interposer board or not, and if the sleds I have are compatible with standard SATA II drives.

Most of my sleds are marked Y6939 or G7267 on the black groves on the outside of the tray part of it. I would presume that once installed I must enter the System BIOS or SAS Controller BIOS screens to inform the controller that the drives are SATA II not SAS and/or to set other options too. However, here are my hardware questions?

1) I have mostly drive sleds with the model number Y6939 on them, and am curious if these sleds will work? Is it as simple as unbolting the SAS drive and bolting in the SATA II drive and sliding the tray into the 1950? 2) If I have SATA III drives (presuming they will run at SATA II speeds as the Controller is still presumably SATA II regardless of my putting SATA III drives in), they will work just as well in the 1950 Type II using the proper drive sleds?

3) Can I have one SATA II/SATA III drive in one bay and a SAS drive in the second bay? 4) If a SATA II drive is installed (presuming the proper sled and BIOS configuration instantiated), the SATA drives are capable of being bootable I presume? 5) Outside of the SCSI/HD subject: Does anyone know what SLIC tables were standard with 1950 Type II servers? I am planning on trying to do a SLIC enabled install of either Windows 7 Desktop or Windows 7 Server (I know it has a different name 2008 server or something).

Alternatively, I may just install Linux and then run Windows 7 within a virtualization environment (for example, I heard that the 1950 supports ESXI 5.5 free version, so maybe I will start with that). 6) Outside SCSI HD subect: How difficult is it to use IPMI to install Linux or Windows for example if I have no monitor to setup IPMI?

If not, I will just obtain a monitor for the purposes of getting IPMI setup? Thanks in advance to all that read and reply to these interrogatories. Where do I find the ipmitool command (or equivalent) for OEL/OL 5.6 (aka RHEL5r6)? Busy building a set of dev blade servers using OEL 5.6 (registered with ULN) and this is what I see in the channel: root@dev1 /root yum list grep ipmi freeipmi.x86_64 0.5.1-6.0.1.el5 installed freeipmi.i386 0.5.1-6.0.1.el5 ol5_x86_64_latest freeipmi-bmc-watchdog.x86_64 0.5.1-6.0.1.el5 ol5_x86_64_latest freeipmi-devel.i386 0.5.1-6.0.1.el5 ol5_x86_64_latest freeipmi-devel.x86_64 0.5.1-6.0.1.el5 ol5_x86_64_latest freeipmi-ipmidetectd.x86_64 0.5.1-6.0.1.el5 ol5_x86_64_latestThe freeipmi.x86_64 package does not include the ipmitool command. There is a source rpm and source code available for it on the web - but I'm not inclined to built s/w from source on Oracle Linux.

Is there an alternative command that one can use to configure the IPMI interface? Or am I missing a rpm to install, common sense, or something else?;-) Will appreciate comments/suggestions/feedback. I'm trying to test out DPM on a Dell PowerEdge 2950. FOr this to work I need to setup IPMI so I can have a server go into safemode and exit out of safe mode. I've been googling like mad but have been unable to find an answer on how to get this to work. Whenever I try to put my host in standby I get this message: vCenter has determined that it cannot resume host lt;hostnamegt; from standby; therefore, the enter standby task was stopped. Confirm that IPMI/ILO is correctly configured for host lt;hostnamegt.

Or, configure vCenter to use WoL, ensuring there are at least two or more hosts in the cluster, and try the task again. A couple things about this error: One this is a two host cluster (for testing purposes). Two, I'm using Dell hardware so I have to use IPMI not ILO. I have logged into my DRAC page and have enabled IPMI over LAN, but I don't know what else to do here, nor do I know where any settings concerning this are found in vCenter or ESX. Any help would be appreciated. Hi, I have created a Red Hat Enterprise Linux Server release 5.3 Beta (Tikanga) Virtual Machine on HP DL 380G5. I installed ipmitool-1.8.11 manually and when I tried to start the ipmi service, I am getting the following errors, lt;codegt;[root@xxxxxxx root]# /etc/init.d/ipmi start Starting ipmi drivers: # /etc/init.d/ipmi status ipmi_msghandler module not loaded.

Ipmi_si_drv module not loaded. Ipmi_devintf module not loaded. /dev/ipmi0 does not exist. Lt;/codegt; When I cheked in /var/log/messages Oct 6 17:10:29 xxxxxx kernel: IPMI System Interface driver. Oct 6 17:10:29 xxxxx kernel: ipmi_si: Unable to find any System Interface(s) Oct 6 17:36:35 xxxxx kernel: ipmi device interface root@xxxxx# modprobe ipmi_si_drv FATAL: Module ipmi_si_drv not found. Root@xxxx # uname -a Linux aspint7-sol.ind.hp.com 2.6.18-118.el5 #1 SMP Sat Oct 4 00:21:41 EDT 2008 i686 i686 i386 GNU/Linux ++Does that mean that my kernel does not support ipmi_si? I tried installing the IPMI on the ESX3.5 which has this virtual machine and it is working fine.

But I am facing issues only when I try installing ipmi on virtual machine. Hi, I have created a Red Hat Enterprise Linux Server release 5.3 Beta (Tikanga) Virtual Machine on HP DL 380G5.

I installed ipmitool-1.8.11 manually and when I tried to start the ipmi service, I am getting the following errors, lt;codegt;[root@xxxxxxx root]# /etc/init.d/ipmi start Starting ipmi drivers: # /etc/init.d/ipmi status ipmi_msghandler module not loaded. Ipmi_si_drv module not loaded. Ipmi_devintf module not loaded. /dev/ipmi0 does not exist. Lt;/codegt; When I cheked in /var/log/messages Oct 6 17:10:29 xxxxxx kernel: IPMI System Interface driver.

Oct 6 17:10:29 xxxxx kernel: ipmi_si: Unable to find any System Interface(s) Oct 6 17:36:35 xxxxx kernel: ipmi device interface root@xxxxx# modprobe ipmi_si_drv FATAL: Module ipmi_si_drv not found. Root@xxxx # uname -a Linux aspint7-sol.ind.hp.com 2.6.18-118.el5 #1 SMP Sat Oct 4 00:21:41 EDT 2008 i686 i686 i386 GNU/Linux ++Does that mean that my kernel does not support ipmi_si? I tried installing the IPMI on the ESX3.5 which has this virtual machine and it is working fine. But I am facing issues only when I try installing ipmi on virtual machine. We have vSphere Essentials using the vCenter Server Appliance (VCSA) with two Dell PowerEdge R805 hosts.

Both have identical hardware configurations and are generally working fine. Because these systems don't include the full Dell Remote Access Controller (DRAC or iDRAC) option, we'd hoped to utilize the Baseboard Management Controller (BMC) for IPMI feature set for remote power controls from within vSphere client, rather than some third-party Windows-based IPMI utility we'd never use otherwise. I've confirmed they are both configured correctly and I can connect using a third-party IPMI utility to both. On one host, I was able to fill out the settings under Host - Configuration tab - Software - Power Management and when clicking Properties provide the Username, Password, BMC IP Address, and BMC MAC Address, and was able to save when I clicked OK.

On the other host, I get a 'Connection Error' dialog and it states the request failed because the remote server took too long to respond. Hallo I am using a Supermicro Server Board MBD-H8SGL-F.

(8C CPU und 16GB RAM) with the latest ESXi 4.1. My Problem is that after one day (more or less) I am losing all the Hardware Sensors. (Only the CPU Sensor is stil there) After a host reboot I can see the sensors agin for 'on day'. It looks like that ther is a communication Problem between the IPMI Chip and the ESXi. BIOS and IPMI Firmeware are the latest available. The 'support' of supermicro says there is a patch for this problem that maybe work.

PATCH = ESXi401.zip So my question is: Is this patch also possibel to install on ESXi4.1 and if yes how should I install this patch (CLI?) Thanks for your answeres db2flasher. I have discovered that the MS ipmidrv.sys silently does not load if the IPMI hardware has register spacing on 4-byte boundaries instead of 1-byte boundaries. Apparently the driver does not think that an IPMI interface is present, so it does not log anything. This is not resolved by the ipmidrv.sys from the service pack. The same hardware can be recognized and used by using the Intel imbdrv.sys driver for Windows, or by using the Linux IPMI driver, because these drivers recognize the IPMI register spacing as advertised in the SMBIOS data.

This information should be detected by the IPMI driver by reading the SMBIOS data in the IPMI record (DMI type 38). The byte at offset 16 in this record contains the register spacing in its upper 2 bits as follows: 0 = 1-byte spacing, 1 = 4-byte spacing, 2 = 16-byte spacing. The effect of the register spacing governs the difference between the KCS command and status registers, for instance: 0x0ca2, 0x0ca3 = 1-byte register spacing 0x0ca2, 0x0ca6 = 4-byte register spacing 0x0ca2, 0x0cb2 = 16-byte register spacing Please correct this problem, as described above. If questions, feel free to email me. Thank you, Andy Cress Kontron America, Inc.

We use Win2008 R2 and Win2008 R2 SP1. I do not think that the version matters until register spacing is resolved in ipmidrv.sys by using the DMI type 38 record.

The list of vendors came from various customers who use IPMI on those platforms, most of those were reported against non-Microsoft OSs which support any register spacing. I believe that Dell, IBM, and Supermicro have several different IPMI firmware sources, so some have 1-byte spacing, and some have 4-byte spacing. But nearly all Kontron and Pigeon Point platforms have 4-byte spacing.

I have a Kontron KTC5520/EATX motherboard in my lab that is one example. Hello, Were you able to get this working yet? We're getting ready to order this card for some of our SuperMicro boxes. Our dive into it came when we were not seeing PS failures on remote servers (that's a bad thing ).

I tried unsuccessfully to ascertain the PS status info from the servers without any additional hardware, but was unable to do so. It looks like our best bet is going to be going with these cards. I was hoping the addition of the card would allow ESX to see the IPMI data. Once you get yours configured, I'm wondering if you're able to see this data in the ESX console.

Worst case scenario, the card itself should be able to send traps and/or email to preconfigured places. It's my understanding that to set the cards up initially, you have to boot off of a CD, which will allow you to configure the IP address and such of the interface. Once that's done, you just boot the server like normal, and make sure you have the managment interface connected to the network. The card is supposed to make ascertaining this information completely OS dependent. Ideally, you could take a machine that's turned off, power it up, mount an ISO, and put an OS on it without even having to be in the same state. This makes me happy. Let me know how your experience turned out.

I have a VMware cluster running on a blade server and all but one of the HOSTS report this error Alert Host IPMI System Event Log status. I did a Google search and came up with this fix VMware KB: The Host IPMI System Event Log Status alarm is triggered repeatedly in VMware vCenter Server 4.x and 5.x In step 4 i get this The system event log is empty.

Any help with this would be great. Resolution To determine why the log has filled up, investigate the hardware. To resolve this issue, stop the alarm from triggering repeatedly and clear the IPMI System Event log and reset the sensors. To clear the log and reset the sensors: Open vCenter Server using vSphere Client.In the vCenter Inventory, select the ESXi/ESX host.Click the Hardware Status tab.Click System Event log under View.Click Reset Event Log. The red alert is removed from the System Event log.Click Reset Sensors to reset the host sensors.Introduced in ESXi 5.1 Update 2 (Build Number: 1483097) and ESXi 5.5 Patch 1 (Build Number: 1474528), there is a new localcli command line to clear the IPMI SEL logs:localcli hardware ipmi sel clearTo execute this command on the ESXi 5.1 or 5.5 host:Connect to the ESXi host via SSH.

For more information, see Using Tech Support Mode in ESXi 4.1 and ESXi 5.x (1017910).Execute this command:localcli hardware ipmi sel clear. Hola, te cuento que yo he tenido el mismo problema, pues debes de realizar lo siguiente por que el mio lo he solucionado asi y bastante logica. Procura que la hora de tu Host Hardware sea la correcta, modifica y guardala, la hora de tu vcenter server, la hora del software esxi ntp este bien sincronizada. Despues realizas esto y eliminas la alarma. To clear the IPMI System Event.log file and reset the sensors: Open vCenter Server using vSphere Client.In the vCenter Inventory, select the ESXi/ESX host.Click the Hardware Status tab.Click System Event log under View.Click Reset Event Log. The red alert is removed from the System Event log.Click Reset Sensors to reset the host sensors. Saludos y me cuentas.

I have a problem to install DOM 5.4 on ESX35U2: # srvadmin-services.sh start Starting Systems Management Device Drivers: Building dell_rbu using DKS: Check /var/log/dell_rbu.dks.log for build error. Starting ipmi driver: Starting Systems Management Device Drivers: Building dell_rbu using DKS: Check /var/log/dell_rbu.dks.log for build error. Starting ipmi driver: Starting DSM SA Shared Services: OK Starting DSM SA Connection Service: # omreport about system summary Error! I have a problem to install DOM 5.4 on ESX35U2: # srvadmin-services.sh start Starting Systems Management Device Drivers: Building dell_rbu using DKS: Check /var/log/dell_rbu.dks.log for build error.

Starting ipmi driver: Starting Systems Management Device Drivers: Building dell_rbu using DKS: Check /var/log/dell_rbu.dks.log for build error. Starting ipmi driver: Starting DSM SA Shared Services: OK Starting DSM SA Connection Service: # omreport about system summary Error! Windows2008_R2_Std Dell r710 server model and installed The following warning is shown in the event log will contack you. -------------------------------------------------------------------------------------- ID: 1004 Source: IPMIDRV Message: The IPMI device driver attempted to communicate with the IPMI BMC device during normal operation. However the communication failed due to a timeout. You can increase the timeouts associated with the IPMI device driver. ------------------------------------------------------------------------------------------------------------------------------------------- 1) Occur in the system event log and Why?

2) I do not want to modify the registry. What are the other solutions? 3) If I ignore the event logs, and performance issues? Hi, I assume you have already found this article: Event ID 1004 — IPMI Driver Functionality The registry key helps increase the timeout duration but it may not able to solve the error if it simply fail to communicate with the IPMI BMC device. I suggest you to check if the driver loaded successfully as the article mentioned: Click Start, then click on Start Search. Type Device Manager and click Enter to open the Device Manager window. Open the System devices node and locate Microsoft Generic IPMI Compliant Device.

Note whether there is an exclamation mark beside the device name. If there is no exclamation mark, the driver has loaded successfully. If IPMI is actually not needed, you can uninstall it as mentioned in this article (for Windows 2003 so just for reference) The command to remove it: Rundll32 ipmisetp.dll, RemoveTheDeviceIf you have any feedback on our support, please send to tnfsl@microsoft.com. Still very new with esxi. Tryng to figure out what I'm doing but I'm liking what I'm seeing so far except one. Apparently my SuperMicro Mobo can't run ESXI if my IPMI is enabled.

So I have to disable it when I boot using this command. Shift O noipmiEnabled That lets me boot the ESXI which is cool except one problem. I think that's a temporary fix because if I rebooted the machine, I'll have to redo it again. So how do I do it permanently to disable the IPMI? I found the solution from this link: Once booted, connect with vSphere and add uncheck VMkernel.Boot.impiEnabled Wasn't quite sure what it means.

Can anyone help me out with the step by step on how to do it? I'm very puzzled.

Contents • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • CPUs [ ] Marvell has a license and designs its CPUs based on the ARMv5, ARMv6 and ARMv7 instruction sets. Marvell CPUs are embedded into various Marvell SoCs. Options ahci marvell_enable = 1 and run update-initramfs -u or dracut -f /boot/initramfs-$(uname -r).img $(uname -r) shutdown enable Marvell and see if it works for you Linux driver [ ] On the website, you can find the source code and documentation for Marvell SATA Storage Controllers 6121/6122/6141/6145 that explains how to build and install the Linux driver for these controllers. The code source version is v1.0.0.9 for Marvell SATA Controllers and was developed by a Marvell developer.. Archived from on 2016-03-03.

] Linux PCI ID Repository [ ] The Linux PCI ID Repository is a public repository of all known ID's used in PCI devices, that is ID's of vendors, devices, subsystems and device classes. These ID lists (usually in form of a pci.ids file) are used in various programs (The Linux Kernel, The PCI Utilities, new version of the XFree86 etc.) to display full device names instead of the numeric ID's. The PCI ID for Marvell 88SE61xx chipset are available The April 8, 2008, there is only PCI ID for Marvell 88SE6101, 88SE6121, 88SE6141 and 88SE6145 chips. PCI ID for Marvell 88SE6102 and 88SE6111 are added April 9, 2008. After the April 9, 2008 and when the PCI ID list will be up to date in the Linux distributions, the Marvell 88SE6102 and 88SE6111 chips will be recognize by Linux but will not work because actually there is no drivers for them. As of the present date [11 January 2012], the Linux kernel supports at least the 88SE6102 chips. In the, Marvell's Libertas chipset runs an on an and interfaces over a with the actual operating system.

• Marvell Alaska MV88E1041S an -family • Marvell Libertas WNIC-family 88E80xx (Yukon) Gigabit Ethernet chipsets [ ] The following utilize 88E80xx (Yukon) Gigabit Ethernet chipsets: Marvell 88E80xx (Yukon) Gigabit Ethernet are chips. Marvell doesn't provide a lot of information for these chipsets on their website. The only page for their Ethernet chipsets is:. •, Linux PCI ID Repository Website, accessed April 9, 2008.

• source: •, accessed January 14, 2010 •, Xensource, accessed April 8, 2008. •, Xensource, accessed April 8, 2008. •, kerneltrap.org Website, accessed April 12, 2008. • • Source: • • pata_marvell: Undo the regressions Jeff caused • [You need to load pata_marvell before AHCI on current kernels (this is fixed in.28rc) for PATA support.

Comment #34 From Alan • How to get a Marvell 88SE6145 (and probably 6121) sata controller working on new kernels • Delvare, Jean.. Kernel Bug Tracker.

Retrieved 2012-01-11. •, MSI Website, accessed April 8, 2008. •, Intel Website, accessed April 8, 2008. •, Intel Website, accessed April 8, 2008.

•, Nnov Website, accessed April 8, 2008. •, Intel Website, accessed October 11, 2008. •, Asus Website, accessed April 8, 2008. •, Asus Website, accessed April 8, 2008. •, Asus Website, accessed April 8, 2008.

•, Asus U.S.A. Website, accessed October 19, 2008. •, Asus Website, accessed December 10, 2009. •, Asus Website, accessed April 8, 2008.

•, Asus Website, accessed April 8, 2008. • •, Asus U.S.A. Website, accessed August 8, 2008. •, Asus Website, accessed April 8, 2008.

•, Asus Website, accessed April 8, 2008. • (E3327 First Edition V1 ed.), ASUSTek Computer, Inc., 2007, retrieved 2009-01-07 •, Intel Website, accessed October 14, 2008. Retrieved 2015-07-16. Retrieved 2015-07-16. •, Linux PCI ID Repository Website, accessed April 9, 2008.

Code: 04:00.0 SATA controller: Marvell Technology Group Ltd. 88SE9230 PCIe SATA 6Gb/s Controller (rev 11) (prog-if 01 [AHCI 1.0]) Subsystem: Marvell Technology Group Ltd.