Hello,
Not sure if this is the correct repo to open issue, if not please close this issue and I can reopen the issue to a desired one.
Can you please share details on LSF exit code mapping, for instance, exit code 32 to UNIX signal mapping. The documentation for 32 code says EXIT but there could be several reasons for exit. Does exit code 32 mean SIGPROF?
Thanks

Hi,

I am writing this here to ask where is the LSF to K8S IBM connector guide in the official IBM website?
It looks like it was removed from the official docs.
https://www.ibm.com/docs/en/spectrum-lsf/10.1.0?topic=kubernetes-overview

Thank you.

Hello,
Does LSF allow processes launched by different users to share GPU on a single host?

Thanks

I am trying to download IBM Spectrum Computing Cloud Pak Technical Preview. But it seems there is a problem with download link from the following URL https://epwt-www.mybluemix.net/software/support/trial/cst/programwebsite.wss?siteId=548&h=&tabId=1091&p=

Hello,

I have read your LSF TP document. it's well written. can you please tell which container technology was used to build container in Openshift environment (Docker or podman?)

Hi,

I refered the "README for IBM Spectrum Computing LSF-Kubernetes Integration, Tech preview" but also the knowledge center guide of LSF Connector for Kubernetes and have tested the same environment.

After all configuration, we've encountered some issues of job submitting. When I submit the job through LSF, it work properly. However, if job is submitted through the Kubernetes, it shows pending/error state and K8S pod terminated. We tried allocate both cpu and gpu resources but doesn't work either.

Here's the each versions of docker, nvidia-docker, ICP for k8s distribution. Also I attached some logs.

Is it bug ? Or can I assume any configuration missing ?

[root@lsfsuite log]# nvidia-docker version
NVIDIA Docker: 2.0.3
Client:
Version: 18.09.3
API version: 1.39
Go version: go1.10.8
Git commit: 774a1f4
Built: Thu Feb 28 06:33:21 2019
OS/Arch: linux/amd64
Experimental: false

Server: Docker Engine - Community
Engine:
Version: 18.09.3
API version: 1.39 (minimum version 1.12)
Go version: go1.10.8
Git commit: 774a1f4
Built: Thu Feb 28 06:02:24 2019
OS/Arch: linux/amd64
Experimental: false

[root@lsfsuite log]# docker -v
Docker version 18.09.3, build 774a1f4

[root@lsfsuite log]# nvidia-docker -v
Docker version 18.09.3, build 774a1f4

[root@lsfsuite log]# kubectl describe daemonset platform-ui -n kube-system |grep ICP_VERSION
ICP_VERSION: 3.1.0

[root@lsfsuite log]# kubectl get nodes
NAME STATUS ROLES AGE VERSION

10.0.0.180 Ready etcd,management,master,proxy,worker 39d v1.11.1+icp-ee

These are all logs from batch-drive.log. There's nothing inside kubebridge.lsfmaster.log

[lsfadmin@lsfsuite log]$ cat batch-driver.log
Log file created at: 2019/11/27 14:44:12
Running on machine: lsfsuite
Binary: Built with gc go1.11.2 for linux/amd64
Log line format: [IWEF]mmdd hh:mm:ss.uuuuuu threadid file:line] msg
W1127 14:44:12.720253 30795 authorization.go:47] Authorization is disabled
W1127 14:44:12.720353 30795 authentication.go:55] Authentication is disabled
I1127 14:44:12.720360 30795 deprecated_insecure_serving.go:49] Serving healthz insecurely on [::]:10501

Hello,

I have read your LSF TP document. it's well written. However, I follow /lsf kubernetes/docker/lsf_ operator/Read. Some errors occurred while me was operating.

#kubectl logs pod-sched-master -n lsf
...
Error: LIM Process has exited due to a fatal error.
res/get_hostinfo() failed. Master Lim is down,try later
...

#kubectl logs pod-sched-agent-bgdcr -n lsf
...
Info: Running change_hostname()
Info: change_hostname: Change name from : pod-sched-agent-bgdcr to worker-10-119-221-133. hostname: you must be root to change the host name
...

#kubectl describute pods pod-sched-agent-bgdcr -n lsf
...
Events:
Warning unhealthy Liveness Probe failed: ls cannot access /tmp/lsf-alive ,no such a file or directory
Warning unhealthy readiness Probe failed: ls cannot access /tmp/lsf-alive ,no such a file or directory
...

These problems have bothered me for a long time. Can you help me？

thank you