[Ajay Kumar]

I was bounded by time deadline, thus, I deleted that slice and recreated so it working correctly now, if i still face this same issue, I would raise it here in this thread.

[Ajay Kumar]

I have also tried to purge all CUDA, NVIDIA drivers and install from scratch. This does not work as well.

Commands used:

sudo apt-get purge -y ‘*nvidia*’

sudo apt-get autoremove -y

sudo apt-get autoclean

sudo reboot

 

[Ajay Kumar]

slice_name=’GPU_Variant_Calling_FIU’
node_name=’Node3′
slice = fablib.get_slice(slice_name)
node = slice.get_node(node_name)
node.os_reboot()

This piece of code generated this error. Now that it’s live, I can access this node. Thank you very much, Komal.

[Ajay Kumar]

Thank you very much, Komal, you are always a big help while working with fabric. I am not sure, but maybe overloading tasks on GPUs might have caused it crashed and then, when we reboot, it vanish the network interface settings and detaches PCI devices.

It’s working pretty much well, thank you so much 😊 .

[Ajay Kumar]

Yes I did! but anyways, it started working, I guess there was some time constraint issue that time. Its working perfectly now.

[Ajay Kumar]

Is there issues going with fabric Jupyter lab to create cluster right now? It was working pretty fine yesterday.

[Ajay Kumar]

Thank you very much, now it works fine, double hands up for your help, Komal.

[Ajay Kumar]

My Slice ID: 09255c48-5512-4e3c-bdc6-ad7d4fd37d07
Output of ifconfig -a command:

(base) ubuntu@Node4:~$ ifconfig -a
docker0: flags=4099<UP,BROADCAST,MULTICAST> mtu 1500
inet 172.17.0.1 netmask 255.255.0.0 broadcast 172.17.255.255
ether a2:9a:8b:03:9c:61 txqueuelen 0 (Ethernet)
RX packets 0 bytes 0 (0.0 B)
RX errors 0 dropped 0 overruns 0 frame 0
TX packets 0 bytes 0 (0.0 B)
TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0

enp3s0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST> mtu 9000
inet 10.20.4.248 netmask 255.255.254.0 broadcast 10.20.5.255
inet6 fe80::f816:3eff:fe3a:e097 prefixlen 64 scopeid 0x20<link>
ether fa:16:3e:3a:e0:97 txqueuelen 1000 (Ethernet)
RX packets 541 bytes 53260 (53.2 KB)
RX errors 0 dropped 0 overruns 0 frame 0
TX packets 417 bytes 56772 (56.7 KB)
TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0

lo: flags=73<UP,LOOPBACK,RUNNING> mtu 65536
inet 127.0.0.1 netmask 255.0.0.0
inet6 ::1 prefixlen 128 scopeid 0x10<host>
loop txqueuelen 1000 (Local Loopback)
RX packets 114 bytes 9436 (9.4 KB)
RX errors 0 dropped 0 overruns 0 frame 0
TX packets 114 bytes 9436 (9.4 KB)
TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0

[Ajay Kumar]

Thank you so much, Komal, it worked for me.

Following on that, I noticed my interface (enp9s0) is not found, earlier it was there. I have used this interface to connect with other nodes in the cluster. Could you please help me to make it UP again?

(base) ubuntu@Node4:~$ ifconfig
docker0: flags=4099<UP,BROADCAST,MULTICAST> mtu 1500
inet 172.17.0.1 netmask 255.255.0.0 broadcast 172.17.255.255
ether a2:9a:8b:03:9c:61 txqueuelen 0 (Ethernet)
RX packets 0 bytes 0 (0.0 B)
RX errors 0 dropped 0 overruns 0 frame 0
TX packets 0 bytes 0 (0.0 B)
TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0

enp3s0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST> mtu 9000
inet 10.20.4.248 netmask 255.255.254.0 broadcast 10.20.5.255
inet6 fe80::f816:3eff:fe3a:e097 prefixlen 64 scopeid 0x20<link>
ether fa:16:3e:3a:e0:97 txqueuelen 1000 (Ethernet)
RX packets 179 bytes 18635 (18.6 KB)
RX errors 0 dropped 0 overruns 0 frame 0
TX packets 168 bytes 22384 (22.3 KB)
TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0

lo: flags=73<UP,LOOPBACK,RUNNING> mtu 65536
inet 127.0.0.1 netmask 255.0.0.0
inet6 ::1 prefixlen 128 scopeid 0x10<host>
loop txqueuelen 1000 (Local Loopback)
RX packets 110 bytes 8928 (8.9 KB)
RX errors 0 dropped 0 overruns 0 frame 0
TX packets 110 bytes 8928 (8.9 KB)
TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0

[Ajay Kumar]

Does anyone know, how to reboot a node even if ping and ssh not working to that same node?

[Author]

Posts