How to detect if SEA ent adapter is not working right in VIO AIX

Posted on 2011-04-26
Last Modified: 2013-11-17
This is a bit dumb and confusing question/inquiry.

I have a dual vio setup in my environment on which i have several Lpars.
I am planning on upgrading my Vio's from 2.1 to 2.2

My colleague told me before he left for vacation that there is an adapter issue on vio1 and NIC card for ent15(SEA) needs to be replaced, before i begin the vio upgrade process.

However i am unable to determine the error or issue that he saw in the system for any issues with ent15(SEA).

I need assistance in troubleshooting this SEA and seeing if there is something wrong with it.

Question by:assistunix
    LVL 68

    Assisted Solution

    Hi again,

    I think you're aware that a SEA is a virtual entity, not a real piece of hardware!

    SEA is a bridge between the real hardware and the virtual adapter(s) as defined in the VIOS partition's profile for interpartition logical lan access.

    To see the components involved issue

    lsdev -dev ent15 -attr

    Under "real adapter" you will see the "Physical adapter associated with the SEA", under "virt adapters" you will see a "List of virtual adapters associated with the SEA".
    The "real adapter" can be an etherchannel or link aggregation, not just a single card.

    Now for troubleshooting hardware issues you could try (assuming "ent5" as the real adapter):

    errlog -ls | grep -p ent5 | more

    and check for relevant error messages. "errlog" is the counterpart to "errpt" in native AIX, by the way.

    For further analysis you will most probably have to leave the IOSCLI environment, to be able to use "diag" or "smitty diag". To go to native AIX issue


    Please let me know what you found, maybe I could assist you a bit more then.



    Author Comment

    Hello wmp.

    The issue is that ent15 is in standy mode.
    I was going to fix that status with sudo chdev -l ent15 -a ha_mode=auto  but i was told not to do so by my colleague who went on vacation, who mentioned that the NIC for ent15 needs to be replaced.

    [padmin:/home/padmin]lsdev -dev ent15 -attr
    attribute     value    description                                                        user_settable

    accounting    disabled Enable per-client accounting of network statistics                 True
    ctl_chan      ent9     Control Channel adapter for SEA failover                           True
    gvrp          no       Enable GARP VLAN Registration Protocol (GVRP)                      True
    ha_mode       standby  High Availability Mode                                             True
    jumbo_frames  no       Enable Gigabit Ethernet Jumbo Frames                               True
    large_receive no       Enable receive TCP segment aggregation                             True
    largesend     0        Enable Hardware Transmit TCP Resegmentation                        True
    netaddr       0        Address to ping                                                    True
    pvid          601      PVID to use for the SEA device                                     True
    pvid_adapter  ent8     Default virtual adapter to use for non-VLAN-tagged packets         True
    qos_mode      disabled N/A                                                                True
    real_adapter  ent14    Physical adapter associated with the SEA                           True
    thread        1        Thread mode enabled (1) or disabled (0)                            True
    virt_adapters ent8     List of virtual adapters associated with the SEA (comma separated) True
    [padmin:/home/padmin]errlog -ls | grep -p ent14 | more

    I did not see any errors in errlog for ent14 the real adapter of ent15.

    Please advise on how further it can be determined to see there is an issue with ent15, that it requires the NIC for ent15 be replaced.

    Author Comment

    [root]lsdev -Cc adapter |grep -i ent15
    ent15     Available       Shared Ethernet Adapter

    [root]netstat -v |grep -i link
    Link Status: Up
    Link Status: Up

    Link Status: Up
    Link Status: Up
    Link Status: Up

    Link Status: Up
    Link Status: Up
    Link Status: Up

    [root]entstat -d ent15 |grep -i link

    Link Status: Up
    Link Status: Up
    Link Status: Up

    [root]entstat -d ent15 | grep -i media

    Media Speed Selected: Auto negotiation
    Media Speed Running: 1000 Mbps Full Duplex
    Media Speed Selected: Auto negotiation
    Media Speed Running: 1000 Mbps Full Duplex
    Media Speed Selected: Auto negotiation
    Media Speed Running: 1000 Mbps Full Duplex
    LVL 68

    Accepted Solution


    seems that this SEA has been manually set to "standby" to force a failover to the backup SEA adapter in the secondary VIOS.

    Switching it back to "auto" should initiate a failback to this adapter, and that's obviously what your colleague wants to avoid.

    But why?

    ent15 is a piece of software and cannot be "defective" in a narrow sense.

    The "NIC for ent15" is the real adapter ent14 (or what's a "NIC" in your terminology?), and you told me that you didn't find anything in the errorlog for this real adapter ent14.

    Running "diag" against a busy adapter is not possible (as long as there is a SEA defined its real part will be "in use" from an OS perspective, regardless of whether the SEA itself is in standby or not).

    So as long as there are no error messages and as long as the adapter is properly connected to a correctly configured switch port (with good cables) I really can't tell you why this card should have an issue.

    The most effective test would be setting ha_mode actually to "auto", to see what happens during failback.

    If the real card ent14 is indeed defective an almost immediate re-failover should occur, because the SEA ent15 should signal a "link loss" to the other VIOS via control channel.

    Even more effective: Ask your colleague (phone/email/homer pigeon/bottle post) why exactly (s)he would not want to activate this SEA again!

    Out of ideas now:



    Author Comment

    Thank you

    Write Comment

    Please enter a first name

    Please enter a last name

    We will never share this with anyone.

    Featured Post

    Highfive Gives IT Their Time Back

    Highfive is so simple that setting up every meeting room takes just minutes and every employee will be able to start or join a call from any room with ease. Never be called into a meeting just to get it started again. This is how video conferencing should work!

    Suggested Solutions

    Let's say you need to move the data of a file system from one partition to another. This generally involves dismounting the file system, backing it up to tapes, and restoring it to a new partition. You may also copy the file system from one place to…
    Why Shell Scripting? Shell scripting is a powerful method of accessing UNIX systems and it is very flexible. Shell scripts are required when we want to execute a sequence of commands in Unix flavored operating systems. “Shell” is the command line i…
    Learn how to navigate the file tree with the shell. Use pwd to print the current working directory: Use ls to list a directory's contents: Use cd to change to a new directory: Use wildcards instead of typing out long directory names: Use ../ to move…
    This tutorial goes over how to archive and restore FreeBSD jails that are managed by ezjail.

    779 members asked questions and received personalized solutions in the past 7 days.

    Join the community of 500,000 technology professionals and ask your questions.

    Join & Ask a Question

    Need Help in Real-Time?

    Connect with top rated Experts

    9 Experts available now in Live!

    Get 1:1 Help Now