Thanks for the reply. I tried the command you mentioned. Even then also
both services are not able to start/stop. They remain in the same status
On Thu, Feb 6, 2014 at 4:16 PM, Arman Khalatyan <arm2arm(a)gmail.com> wrote:
before service lnet stop you should run "lustre_rmmod".
I just using it inside lnet script
Regards
Arman,.
On Thu, Feb 6, 2014 at 11:18 AM, San B <forum.san(a)gmail.com> wrote:
> Ofcourse I've checked it, eth0 is present and it is up (used as internal
> network between the VMs). The other interface eth1 (Host Only adapter in
> VirtualBox) used to connect from Host to Guest(VM).
>
> The log messages were taken during boot.
> After logging in, "modprobe -v lnet" does not show anything(already
> loaded). But "modprobe -v lustre" shows
> insmod
>
/lib/modules/2.6.32-358.6.2.el6_lustre.g230b174.x86_64/updates/kernel/fs/lustre/mdc.ko
> insmod
>
/lib/modules/2.6.32-358.6.2.el6_lustre.g230b174.x86_64/updates/kernel/fs/lustre/osc.ko
> insmod
>
/lib/modules/2.6.32-358.6.2.el6_lustre.g230b174.x86_64/updates/kernel/fs/lustre/lov.ko
> insmod
>
/lib/modules/2.6.32-358.6.2.el6_lustre.g230b174.x86_64/updates/kernel/fs/lustre/lustre.ko
>
> The service status is:
>
> # service lnet status
> running
> # service lustre status
> partial
> # service lustre stop
> # service lustre status
> partial
> # service lnet stop
> Removing module lustre
> ERROR: Module osc has non-zero reference count.
> # service lnet status
> running
>
> So the reloading of lnet/lustre is not working here.
>
> On Wed, Feb 5, 2014 at 1:11 AM, Mohr Jr, Richard Frank (Rick Mohr) <
> rmohr(a)utk.edu> wrote:
>
>>
>> Are the log messages from when the system booted up? If so, did the
>> Lustre modules get loaded before the eth0 interface was activated?
>>
>> I would run "ifconfig" and some ping tests to verify that eth0 is
>> active. If so, try reloading the lnet/lustre modules to see if it starts
>> working.
>>
>> --
>> Rick Mohr
>> Senior HPC System Administrator
>> National Institute for Computational Sciences
>>
http://www.nics.tennessee.edu
>>
>>
>>
>> On Feb 4, 2014, at 1:37 PM, San B <forum.san(a)gmail.com>
>> wrote:
>>
>> > Hi,
>> >
>> > The lustre service on MGS server throws following message.
>> >
>> > # service lustre status
>> > partial
>> >
>> > This Lustre implementation is performed on VirtualBox VMs. Hence
>> No Infiniband connectivity. I tried to debug it further and found following
>> error messages:
>> >
>> > The /var/log/messages :
>> >
>> > Feb 4 18:02:11 ic2 kernel: LNet: HW CPU cores: 1, npartitions: 1
>> > Feb 4 18:02:11 ic2 kernel: alg: No test for crc32 (crc32-table)
>> > Feb 4 18:02:11 ic2 kernel: alg: No test for adler32 (adler32-zlib)
>> > Feb 4 18:02:11 ic2 kernel: padlock: VIA PadLock Hash Engine not
>> detected.
>> > Feb 4 18:02:11 ic2 kernel: Lustre: Lustre: Build Version:
>> 2.4.0-RC2-gd3f91c4-PRISTINE-2.6.32-358.6.2.el6_lustre.g230b174.x86_64
>> > Feb 4 18:02:11 ic2 kernel: LNetError:
>> 685:0:(socklnd.c:2822:ksocknal_startup()) Interface eth0 is down
>> > Feb 4 18:02:11 ic2 kernel: LNetError: 105-4: Error -100 starting up
>> LNI tcp
>> > Feb 4 18:02:11 ic2 kernel: LustreError:
>> 685:0:(events.c:806:ptlrpc_init_portals()) network initialisation failed
>> >
>> > # cat /etc/modprobe.d/lustre.conf
>> > options lnet networks="tcp0(eth0)"
>> >
>> > # modprobe -v lnet
>> > # modprobe -v lustre
>> > insmod
>>
/lib/modules/2.6.32-358.6.2.el6_lustre.g230b174.x86_64/updates/kernel/fs/lustre/mdc.ko
>> > insmod
>>
/lib/modules/2.6.32-358.6.2.el6_lustre.g230b174.x86_64/updates/kernel/fs/lustre/osc.ko
>> > insmod
>>
/lib/modules/2.6.32-358.6.2.el6_lustre.g230b174.x86_64/updates/kernel/fs/lustre/lov.ko
>> > insmod
>>
/lib/modules/2.6.32-358.6.2.el6_lustre.g230b174.x86_64/updates/kernel/fs/lustre/lustre.ko
>> > # df -h
>> > Filesystem Size Used Avail Use% Mounted on
>> > /dev/mapper/VolGroup-lv_root
>> > 8.5G 2.9G 5.2G 36% /
>> > tmpfs 246M 0 246M 0% /dev/shm
>> > /dev/sda1 485M 57M 403M 13% /boot
>> > /dev/sr0 3.5G 3.5G 0 100% /var/isoimages/rhel6.4
>> > /dev/sdb 7.9G 344M 7.2G 5% /mgs
>> > #
>> >
>> > Can someone point out what could be the wrong here?
>> >
>> > Thanks in advance
>> > _______________________________________________
>> > HPDD-discuss mailing list
>> > HPDD-discuss(a)lists.01.org
>> >
https://lists.01.org/mailman/listinfo/hpdd-discuss
>>
>>
>>
>>
>
> _______________________________________________
> HPDD-discuss mailing list
> HPDD-discuss(a)lists.01.org
>
https://lists.01.org/mailman/listinfo/hpdd-discuss
>
>