Changing a Network Interface on Solaris 8

This is the joke:
1- Solaris 8 was first installed with a NIC not recognized by Solaris (in-board)
2- A compatible PCI Card has been installed (3Com 3C905B-TX).
3- Old Card has been desactived in the BIOS. No IRQ conflict.
4- At reboot, the Configuration Assistant recognizes the device...
5- At boot, the system does not manage to configure the interface.
The reason comes from that the interface is not recognized by
ifconfig (does not appear in result of "ifconfig -a").
The message "BAd File Number" is obtained with "ipconfig elxl0 plumb".
Questions: - Why?
- Is there a solution? (apart to reinstall the system...)
Hints:
- the interface should be named "elxl0"
(because it is the name of the 3Com interface on other Solaris PC's).
Is it right ????
- the driver name elxl does not appear in the PCI devices of "prtconf -D"
- but the file /kernel/drv/elxl exists !
Waiting for enlighting, thanks very much!!!!

Try to change the network binding on the server
http://technet.microsoft.com/en-us/library/cc732472(WS.10).aspx
Try to delete the hidden network adapters as well. 
Open Device Manager, from the File menu, expand the View and select the Show Hidden Devices option. This will enable and show any old un-used devices in the window
http://www.gfi.com/blog/how-to-remove-hidden-network-adapters-from-virtual-machines/
Thanks,
MAS
Please mark as helpful if you find my comment helpful or as an answer if it does answer your question. That will encourage me - and others - to take time out to help you.

Similar Messages

  • PRIF-33 and CRS-02307 while changing public network interface, RAC

    Hi,
    I'm working on an Oracle Database 11g Enterprise Edition Release 11.2.0.3.0 - 64bit Production With the Partitioning, Real Application Clusters, Automatic Storage Management. It's a 3 nodes RAC: mgvdb01/02/03
    After the installation I had to change IP (different IP and Subnet) to the RAC:
    From 172.17.1.0/24 to 10.19.201.0/24.
    Node from 172.17.1.31/32/33 to 10.19.201.31/32/33.
    The same for VIP(s): from 172.17.1.131/132/133 to 10.19.201.131/132/133.
    The oifcfg iflist shows correct ips configuration:
    [oracle@mgvdb01 bin]$ ./oifcfg iflist -p -n
    eth0  10.19.201.0  PRIVATE  255.255.255.0
    eth1  172.17.100.0  PRIVATE  255.255.255.0
    I'm following doc 276434.1 from Metalink: How to Modify Public Network Information including VIP in Oracle Clusterware, starting from Case III "Changing public network interface, subnet or netmask".
    But at the first operation I've a problem:
    [oracle@mgvdb01 bin]$ ./oifcfg getif
    eth0  172.17.1.0  global  public  **THIS HAS TO BE CHANGED**
    eth1  172.17.100.0  global  cluster_interconnect
    [oracle@mgvdb01 bin]# ./oifcfg delif -global eth0/172.17.1.0
    PRIF-33: Failed to set or delete interface because hosts could not be discovered
      CRS-02307: No GPnP services on requested remote hosts.
    PRIF-32: Error in checking for profile availability for host mgvdb02
      CRS-02306: GPnP service on host "mgvdb02" not found.
    PRIF-32: Error in checking for profile availability for host mgvdb03
      CRS-02306: GPnP service on host "mgvdb03" not found.
    [oracle@mgvdb01 bin]$ ./oifcfg delif -node mgvdb01 eth0/172.17.1.0
    [oracle@mgvdb01 bin]$ ./oifcfg setif -node mgvdb01 eth0/10.19.201.0:public
    PRIF-33: Failed to set or delete interface because hosts could not be discovered
      CRS-02307: No GPnP services on requested remote hosts.
    PRIF-32: Error in checking for profile availability for host mgvdb02
      CRS-02306: GPnP service on host "mgvdb02" not found.
    PRIF-32: Error in checking for profile availability for host mgvdb03
      CRS-02306: GPnP service on host "mgvdb03" not found.Then I restart Clusterware service, but issuing
    [oracle@mgvdb01 bin]$ ./oifcfg getif
    eth0  172.17.1.0  global  public
    eth1  172.17.100.0  global  cluster_interconnectnothing seems to be changed.
    This is blocking following operations, the ones in Case IV.
    Do you have any suggestion?
    Thanks in advance,
    Samuel

    Hi,
    I'm working on an Oracle Database 11g Enterprise Edition Release 11.2.0.3.0 - 64bit Production With the Partitioning, Real Application Clusters, Automatic Storage Management. It's a 3 nodes RAC: mgvdb01/02/03
    After the installation I had to change IP (different IP and Subnet) to the RAC:
    From 172.17.1.0/24 to 10.19.201.0/24.
    Node from 172.17.1.31/32/33 to 10.19.201.31/32/33.
    The same for VIP(s): from 172.17.1.131/132/133 to 10.19.201.131/132/133.
    The oifcfg iflist shows correct ips configuration:
    [oracle@mgvdb01 bin]$ ./oifcfg iflist -p -n
    eth0  10.19.201.0  PRIVATE  255.255.255.0
    eth1  172.17.100.0  PRIVATE  255.255.255.0
    I'm following doc 276434.1 from Metalink: How to Modify Public Network Information including VIP in Oracle Clusterware, starting from Case III "Changing public network interface, subnet or netmask".
    But at the first operation I've a problem:
    [oracle@mgvdb01 bin]$ ./oifcfg getif
    eth0  172.17.1.0  global  public  **THIS HAS TO BE CHANGED**
    eth1  172.17.100.0  global  cluster_interconnect
    [oracle@mgvdb01 bin]# ./oifcfg delif -global eth0/172.17.1.0
    PRIF-33: Failed to set or delete interface because hosts could not be discovered
      CRS-02307: No GPnP services on requested remote hosts.
    PRIF-32: Error in checking for profile availability for host mgvdb02
      CRS-02306: GPnP service on host "mgvdb02" not found.
    PRIF-32: Error in checking for profile availability for host mgvdb03
      CRS-02306: GPnP service on host "mgvdb03" not found.
    [oracle@mgvdb01 bin]$ ./oifcfg delif -node mgvdb01 eth0/172.17.1.0
    [oracle@mgvdb01 bin]$ ./oifcfg setif -node mgvdb01 eth0/10.19.201.0:public
    PRIF-33: Failed to set or delete interface because hosts could not be discovered
      CRS-02307: No GPnP services on requested remote hosts.
    PRIF-32: Error in checking for profile availability for host mgvdb02
      CRS-02306: GPnP service on host "mgvdb02" not found.
    PRIF-32: Error in checking for profile availability for host mgvdb03
      CRS-02306: GPnP service on host "mgvdb03" not found.Then I restart Clusterware service, but issuing
    [oracle@mgvdb01 bin]$ ./oifcfg getif
    eth0  172.17.1.0  global  public
    eth1  172.17.100.0  global  cluster_interconnectnothing seems to be changed.
    This is blocking following operations, the ones in Case IV.
    Do you have any suggestion?
    Thanks in advance,
    Samuel

  • Changing a Network Interface in DAG Replication Network after going P2V

    I have a DAG member suffering severe hardware failure so I have gone P2V with it.  It's functioning fine, except that the DAG Network Interfaces have changed - The MAPI Interface was on "Local Area Network" on the physical server, and is now
    on Local Area Connection 7 on the VM.   However, the DAG settings still point to "Local Area Network" and will not let me change it.  Same thing for the replication network.
    I have lost redundancy due to this, so I need to get it repaired pretty promptly.
    Removing and re-adding the subnet in the DAG network has no effect - it still references the old physical NIC instead of the new virtual NIC, despite the new virtual NIC having the same IP as the old physical NIC.

    Try to change the network binding on the server
    http://technet.microsoft.com/en-us/library/cc732472(WS.10).aspx
    Try to delete the hidden network adapters as well. 
    Open Device Manager, from the File menu, expand the View and select the Show Hidden Devices option. This will enable and show any old un-used devices in the window
    http://www.gfi.com/blog/how-to-remove-hidden-network-adapters-from-virtual-machines/
    Thanks,
    MAS
    Please mark as helpful if you find my comment helpful or as an answer if it does answer your question. That will encourage me - and others - to take time out to help you.

  • Network interfaces in solaris 10g guests

    Dear all,
    I installed solaris 10 on vm server but it doesn't recognize the virtual network interfaces in the system, i tried all the network command in the OS but no way i was able to solve it.
    Please advice.

    You can see all installed interfaces by running 'grep network /etc/path_to_inst'. An example:
    grep network path_to_inst
    "/node@1/pci@8,600000/network@1" 0 "ge"
    "/node@1/pci@9,700000/network@1,1" 0 "eri"
    "/node@1/pci@9,700000/pci@2/network@0" 0 "ce"
    "/node@1/pci@9,700000/pci@2/network@1" 1 "ce"
    "/node@1/pci@9,700000/pci@2/network@2" 2 "ce"
    "/node@1/pci@9,700000/pci@2/network@3" 3 "ce"
    "/node@1/pci@9,600000/network@2" 4 "ce"
    shows, ge0, eri0, ce0 - ce4

  • Intel network interface with Solaris 10 x86

    Hi,
    I have been Googling all evening to try and find a solution to this, but here is the problem:
    I have a PC built around an Asus P4C800-E system board, and I have Solaris 10 x86 01/2006 installed. I would like to get the Ethernet interface (Intel adaptor, running on a CSA bus) up and running - but for some reason I can't seem to get the driver to bind to the proper address (pci8086,1019 - from what I have been able to find out): It instead insists on binding to pci1043,80f7. Even though pci8086,1019 is listed in /etc/device_aliases, it won't play nice.
    I know the NIC is working, and have tried it under other OSes (including Linux).
    Although Solaris x86 "sees" something there, and the interface is plumbed (it can even tell when the network link is up or down), I can't get the interface to work - pings do not work, etc.
    Is there a way I can convince the driver to bind to pci8086,1019 instead of pci1043,80f7?
    Any assistance would be greatly appreciated.
    Joovilhar.

    <table border="0" align="center" width="90%" cellpadding="3" cellspacing="1"><tr><td class="SmallText"><b>Joovilhar wrote on Wed, 08 March 2006 02:14</b></td></tr><tr><td class="quote">
    I can't seem to get the driver to bind to the proper address (pci8086,1019 - from what I have been able to find out): It instead insists on binding to pci1043,80f7. Even though pci8086,1019 is listed in /etc/device_aliases, it won't play nice.
    </td></tr></table>
    Is 1043,80f7 the device's pci subsystem vendor and subsystem device id?
    And 8086,1019 is the normal vendor/device id?
    In this case everything would work as expected.
    What is listed by "/usr/X11/bin/scanpci -v" for the ethernet controller?
    <table border="0" align="center" width="90%" cellpadding="3" cellspacing="1"><tr><td class="SmallText"><b>Quote:</b></td></tr><tr><td class="quote">
    I know the NIC is working, and have tried it under other OSes (including Linux).
    Although Solaris x86 "sees" something there, and the interface is plumbed (it can even tell when the network link is up or down), I can't get the interface to work - pings do not work, etc.
    </td></tr></table>
    Try to boot the kernel with various settings of the "acpi-user-options" variable,
    to disable / enable ACPI. In previous Solaris x86 releases, when
    something went wrong configuring an interrupt vector for a nic device,
    the typical end result is a nic that is unable to receive incomming
    packets. Maybe your Asus P4C800-E is still problematic with ACPI, with
    S10 x86 1/2006, and needs the acpi-user-options workaround.
    http://blogs.sun.com/roller/page/danasblog?anchor=configurin g_solaris_acpi_at_boot
    To change the "acpi-user-options" variable, edit the grub boot command
    line and edit the "multiboot" line, e.g. append "-Bacpi-user-options=0x2".

  • Network Interface Configuration Solaris 7

    Hi,
    I'm new on Solaris platform. How does one change the gateway IP address to a new gateway IP address. I've tried "route change" command but no success. Any suggestion is very much appreciated.
    Kala
    Telikom PNG Ltd

    Changing the IP address on a server can be quite involved if there are applications running on the server that "know" about the machines IP address ( typically databases ).
    Ordinarily, if you are only changing the IP address you could use the ifconfig command. You might also need to change the netmask value.
    Be careful, if your doing this remotely, say via telnet, once you run the ifconfig command and the IP changes, you will loose your connection to the server. It's best to try this sort of thing whilst your on the system console. If there is any mistake in your command and the interface is configured incorrectly or goes down, you will not be able to reconnect unless your on the system console.
    If you want the IP address change to be persistent ie after reboots, you will need to edit the /etc/hosts file and add the new IP address in. If your using DNS and/or NIS/NIS+ then these files will also have to be changed.

  • Network Interface Configuration

    For Solaris 9,
    <<Q1>> What is the following interfaces name means? le, hme? Which is the Ethernet port?
    <<Q2>> Where is the networking configuration file stored in the Solaris 9?
    <<Q3>>How to configure the network interfaces in Solaris 9?
    <<Q4>> Did we need to restart the network after we have edit the network configuration? If, how to restart the network? (e.g. in Debian /etc/init.d/networking restart).
    Please assist.

    You also posted this same question at another forum web site.
    It appears your answer is going to be over there.
    There's now no sense in duplicating it here.
    ['Network Interface Configuration' at www.LinuxQuestions.Org|http://www.linuxquestions.org/questions/solaris-opensolaris-20/network-interface-configuration-652192/]

  • Setting network interface IP address

    Hello everyone,
    Is it possible to change a network interface IP address(eg set my local area connection IP) purely in Java or should JNI be used?
    Thanks in advance.

    I doubt there is a single, cross-platform way in purely Java, but short of using JNI you could probably use something like Runtime.exec() (assuming your user has proper permissions) and a corresponding command-line utility (like ifconfig on Unix/Linux, ipconfig on Windows).
    Edit From a brief look, it looks like ipconfig won't do the equivalent on Windows as ifconfig will do.
    Edited by: endasil on 16-Sep-2009 3:37 PM
    Edit Looks like the netsh command on Windows does some of those things:
    netsh interface ip set address local static 123.123.123.123 255.255.255.0 Source: [Wikipedia: netsh|http://en.wikipedia.org/wiki/Netsh]
    Edited by: endasil on 16-Sep-2009 3:39 PM

  • [SOLVED] Network interface name keeps changing

    My network interface name keeps changing between wlp2s0 and wlan0. This is a real pain, since I never know what name will be generated at boot. Most of the time it's called wlp2s0, so this is what netctl is set up to handle by itself. At boots were wlan0 is generated I have to connect manually.
    I've tried making the name static by adding a rule to the /etc/udev/rules.d/10-network.rules. The rule i added was:
    SUBSYSTEM=="net", ACTION=="add", ATTR{address}=="48:5d:60:7e:64:e9", NAME="wlp2s0"
    Wlp2s0 is generated at boot alright, but I still have to connect manually with this solution. As if the "new" wlp2s0 is not recognized as the old one (same MAC). What gives?
    This is on an Asus Eee 1001PXD, with the Qualcomm Atheros AR9285.
    Last edited by madr (2013-05-18 20:09:34)

    I've now been able to replicate the issue, and run the command:
    journalctl -b
    I did find the output:
    mai 18 02:30:02 <hostname> systemd-udevd[126]: error changing net interface name wlan0 to wlp2s0: Device or resource busy
    I haven't been able to process the whole output yet (I'll continue tomorrow if necessary), but I've found some things that I hope will clearify things. I get similar outputs to the following a lot throughout:
    mai 18 02:30:03 <hostname> dhcpcd[237]: wlan0: carrier lost
    mai 18 02:30:03 <hostname> dhcpcd[237]: wlan0: waiting for carrier
    mai 18 02:30:04 <hostname> systemd[1]: [email protected]: control process exited, code=exited status=1
    mai 18 02:30:04 <hostname> systemd[1]: Failed to start Automatic wireless network connection using netctl profiles.
    mai 18 02:30:04 <hostname> systemd[1]: Unit [email protected] entered failed state.
    mai 18 02:30:04 <hostname> systemd[1]: Starting Network.
    mai 18 02:30:04 <hostname> systemd[1]: Reached target Network.
    EDIT:
    I see now that this is later in time, so that's not so strange after all. I'll look some more.
    EDIT2:
    The error code that a name change i impossible is the first time the term 'wlan0' occurs in the log. However there's a line much earlier that talks about wlp2s0:
    mai 18 02:29:59 <hostname> systemd[1]: Expecting device sys-subsystem-net-devices-wlp2s0.device...
    But as I said, if this information doesn't help, I'll have a closer look at the output tomorrow. Getting late here in Norway now.
    Last edited by madr (2013-05-18 01:20:50)

  • How to change the Ethernet Interface name in Solaris 10

    I have to install Oracle 10G RAC on three nodes with different ethernet cards, but the ethernet interface name must be the same which is the precondition for Oracle RAC installation.
    So I want to know how can I change the ethernet name in Solaris.
    For example, the ethernet name in the OS is "ce0", how can i change the interface name "ce0" to "e1000g0".
    bash-3.00# dladm show-dev
    ce0 link: unknown speed: 1000 Mbps duplex: full
    ce1 link: unknown speed: 100 Mbps duplex: half
    ce2 link: unknown speed: 100 Mbps duplex: half
    ce3 link: unknown speed: 100 Mbps duplex: half
    ce4 link: unknown speed: 100 Mbps duplex: half
    ce5 link: unknown speed: 0 Mbps duplex: unknown
    Thanks in advance.

    s-wilson wrote:
    You can't. The ce or e1000g refers to the driver as well as the adapter itself. The only exception I am aware of is: the ipge and e1000g which both refer to the Intel Pro 1000.I'm pretty sure I've renamed a driver in the past (and all references to it in name_to_major, path_to_inst, driver_aliases, and minor_perm) and had it function just fine after 'plumb'ing it up. However I just tested this by trying to turn a 'pcn' driver into a 'foobar' driver on a vmware box. It looks like everything works except some internal bits of the driver continue to create one file with 'pcn' in the name instead of 'foobar'. And since this is Solaris 10, the /devices filesystem is dynamic and read-only. I can't seem to force the change.
    So while this may be possible with some drivers (and maybe only on older versions of the OS), it doesn't seem to be generally possible for all drivers on Solaris 10.
    Darren

  • Solaris command to show all network interfaces in a server (plumb/unplumb)?

    Hi there,
    Is there a Solaris command to show all the network interfaces in a server
    (plumb/unplumb etc...).
    I have a V880 that has an eri0 on board, plumbed up. It also has a couple
    of ge0, ge1, and ce0, and a qfe card.
    ifconfig -a only shows eri0.
    (I am using something like "dladm show-link", but this is really for
    the link-aggregation).
    How do I show the remaining network interfaces?
    Thanks
    Tony

    You can see all installed interfaces by running 'grep network /etc/path_to_inst'. An example:
    grep network path_to_inst
    "/node@1/pci@8,600000/network@1" 0 "ge"
    "/node@1/pci@9,700000/network@1,1" 0 "eri"
    "/node@1/pci@9,700000/pci@2/network@0" 0 "ce"
    "/node@1/pci@9,700000/pci@2/network@1" 1 "ce"
    "/node@1/pci@9,700000/pci@2/network@2" 2 "ce"
    "/node@1/pci@9,700000/pci@2/network@3" 3 "ce"
    "/node@1/pci@9,600000/network@2" 4 "ce"
    shows, ge0, eri0, ce0 - ce4

  • Getting the network interface up in solaris 10

    Hi,
    I am a solaris newbie and am trying to get the network interface up on solaris 10. I have set the required details in
    /etc/hostname.if
    /etc/hosts
    /etc/inet/ipnodes
    /etc/inet/netmasks
    /etc/defaultrouter
    yet on the startup i get a message "failed to bring up IPv4 interfaces : interfacename"
    And when the server starts up after giving the command "ifconfig if up"
    the network interface comes up.
    I need to get the interface up on the startup itself.
    can anyone help me on this ?????
    Thanks in advance,

    ... and of course, you need compatible hardware.
    But you failed to tell anyone what Ethernet adapter you're trying to use.
    Is it on the HCL ? http://www.sun.com/bigadmin/hcl/
    If it requires third party driver software, did you install that as well ?
    You failed to tell anyone whether you're installing to Solaris x86 or to Solaris SPARC.
    ( For example, a PeeCee NIC will generally just not work in a SPARC system. )

  • Solaris 10 u11 : Is it possible to mix IPMP with vertual network interface

    Hi
    I have a system with 4 Ethernet interfaces , I'd like to try to have virtual network interface on two of them . then configure IPMP between each two.
    - Is it possible to mix IPMP and VNIC ?
    - online , I found posts about having vnic on solaris by creating additional interface file (ex, /etc/hostname.bge0:1 ) and also the IPMP create such interface?
    - How can I know which IPMP version installed on the system ? and how to know I'm reading the right configuration guide?
    - on what basis should i choose the type of IPMP for my system (Link-based or Probe-based)
    Thanks in advance
    HEBA

    Hi.
    IPMP can coexist with VNIC.
    IPMP - part of Solaris. So you not need check version of IPMP. Just read Docs for Solaris.
    http://docs.oracle.com/cd/E23823_01/html/816-4554/ipmptm-1.html#scrolltoc
    It's not clear what Vnic you plan use. Is it VLAN or just additional IP on interface ?
    For test address of IPMP you should configure interface like bge0:1 with additional options.
    Regards.

  • Cannot see network interface counters under Solaris 10 with Sun SMA

    Hi,
    Noticed this a few weeks ago. When I do a discovery of a Sun server running the Net-SNMP SMA that is a zone, I can see the network interfaces:
    .1.3.6.1.2.1.2.2.1.2.1 = STRING: lo0:1
    .1.3.6.1.2.1.2.2.1.2.2 = STRING: ipge1:1
    .1.3.6.1.2.1.2.2.1.3.1 = INTEGER: softwareLoopback(24)
    .1.3.6.1.2.1.2.2.1.3.2 = INTEGER: ethernetCsmacd(6)
    .1.3.6.1.2.1.2.2.1.4.1 = INTEGER: 8232
    .1.3.6.1.2.1.2.2.1.4.2 = INTEGER: 1500
    .1.3.6.1.2.1.2.2.1.5.1 = Gauge32: 127000000
    .1.3.6.1.2.1.2.2.1.5.2 = Gauge32: 100000000
    But the counters all come up as 0:
    .1.3.6.1.2.1.2.2.1.8.1 = INTEGER: up(1)
    .1.3.6.1.2.1.2.2.1.8.2 = INTEGER: up(1)
    .1.3.6.1.2.1.2.2.1.9.1 = Timeticks: (0) 0:00:00.00
    .1.3.6.1.2.1.2.2.1.9.2 = Timeticks: (0) 0:00:00.00
    .1.3.6.1.2.1.2.2.1.10.1 = Counter32: 0
    .1.3.6.1.2.1.2.2.1.10.2 = Counter32: 0
    .1.3.6.1.2.1.2.2.1.11.1 = Counter32: 0
    .1.3.6.1.2.1.2.2.1.11.2 = Counter32: 0
    .1.3.6.1.2.1.2.2.1.12.1 = Counter32: 0
    .1.3.6.1.2.1.2.2.1.12.2 = Counter32: 0
    .1.3.6.1.2.1.2.2.1.13.1 = Counter32: 0
    .1.3.6.1.2.1.2.2.1.13.2 = Counter32: 0
    .1.3.6.1.2.1.2.2.1.14.1 = Counter32: 0
    .1.3.6.1.2.1.2.2.1.14.2 = Counter32: 0
    .1.3.6.1.2.1.2.2.1.15.1 = Counter32: 0
    .1.3.6.1.2.1.2.2.1.15.2 = Counter32: 0
    .1.3.6.1.2.1.2.2.1.16.1 = Counter32: 0
    .1.3.6.1.2.1.2.2.1.16.2 = Counter32: 0
    .1.3.6.1.2.1.2.2.1.17.1 = Counter32: 0
    .1.3.6.1.2.1.2.2.1.17.2 = Counter32: 0
    .1.3.6.1.2.1.2.2.1.18.1 = Counter32: 0
    This happens consistently across all zones we have running. We can monitor traffic for a zone via the hwardware node SNMP daemon, but certainly seems like we should be able to also monitor network interface utilization via SNMP from any zone as well.
    Has anyone encountered this / know of a fix or workaround for this? We have a valid Sun support contract but they unfortunately have been REALLY BAD about answering questions related to the SMA Net-SNMP agent thus far.
    - Max

    <table border="0" align="center" width="90%" cellpadding="3" cellspacing="1"><tr><td class="SmallText"><b>tcbarrett wrote on Fri, 28 October 2005 15:37</b></td></tr><tr><td class="quote">
    SunOS Release 5.10 Version s10_69 64-bit
    </td></tr></table>
    Why are you trying to run an old obsolete pre-release version of Solaris 10?
    I'd suggest that you install the final released product.

  • Trying to set network interface at 100/full on Solaris 11

    Hi all,
    I hope someone can help me.
    I've built two boxes running Solaris 11 in my lab. Connected to a hub, all seemed well.
    Now, I've connected them to the network here and I'm having the following issue, The machines should be connecting to the network at 100Mbps/full. Instead, it seems they are trying to connect at 10Mbps/full, which is causing a problem.
    When I check the speed on the NIC (interface=net2), speed is set to 10Mb :
         dladm show-linkprop net2
    Returns speed=10.
    PROPERTY     PERM     VALUE
    speed               r-          10
    As you can see from the above, permission is set to read only.
    When I try to set the link speed as follow :
         dladm set-linkprop -p speed=100 net2
    I get the following error
         'Cannot set link property speed on net2 - operation not supported.'
    Which I assume is because of the read-only property/
    I have also tried setting the property old-style with ndd :
         /usr/sbin/ndd -set /dev/<device-name> adv_100fdx_cap 1
    And get 'Operation not supported'.
    And finally, I've entered the settings in /etc/system itself (and rebooted) :
        set net2:adv_100fdx_cap=1
    All to no avail.
    Note the following blog post I've found here indicating someone else had issues with mode/speed in Solaris 11 (unfortunately I'm still stumped!) :
    http://blog.thilelli.net/post/2011/02/19/Solaris-11-Express%3A-Problem-6
    My networks man is adamant that the network settings are correct, i.e. the network is set to 100Mbps, and looking at his switch settings he seems to be correct.
    Does anyone have any ideas as to where I can go from here? Much appreciated....

    Despite what your site's network staff have said, this is likely to be their issue.
    If you are connecting to a managed switch, that entire peripheral may have specific settings on some or all of its ports.
    I suggest you let that switch do whatever it expects to do and remove all customizations to your NIC's configuration.   Let it auto-negotiate.
    You should also ignore that link you have in your initial post.   The ERI network interface that is described there was a very specific network chipset and used on a very small specific number of Sun SPARC boxes.  It was known to have issues with proper negotiation (it would usually work but sometimes just get mucked up).
    See this Symantec link http://www.symantec.com/business/support/index?page=content&id=TECH37566
    and the following two Oracle notes:
    (PDF) http://docs.oracle.com/cd/E19455-01/806-5579-10/806-5579-10.pdf
    http://docs.oracle.com/cd/E19253-01/816-4554/fwawf/index.html

Maybe you are looking for