Container fails to start randomly
Container fails to start randomly
error pod id
840b98harmlsgb
wqaz2xufma32pt
eqyabu82t6l3y9
system log:
2024-02-29T02:38:20Z start container
2024-02-29T02:38:21Z error starting container: Error response from daemon: failed to create task for container: failed to create shim task: OCI runtime create failed: runc create failed: unable to start container process: error during container init: error running hook #0: error running hook: exit status 1, stdout: , stderr: Auto-detected mode as 'legacy'
Inconsistency detected by ld.so: ../sysdeps/x86_64/dl-machine.h: 534: elf_machine_rela_relative: Assertion ELFW(R_TYPE) (reloc->r_info) == R_X86_64_RELATIVE' failed!
nvidia-container-cli: detection error: driver rpc error: failed to process request: unknown`
15 Replies
Hi, I've encountered a new issue. When I create an RTX 4090 pod, it fails to launch. 😔 The system log is as follows. When I delete the pod and rebuild it, everything returns to normal. What could be the reason for this?
I use secure cloud and custom image
2024-02-27T08:13:42Z start container
2024-02-27T08:13:43Z error starting container: Error response from daemon: failed to create task for container: failed to create shim task: OCI runtime create failed: runc create failed: unable to start container process: error during container init: error running hook #0: error running hook: exit status 1, stdout: , stderr: Auto-detected mode as 'legacy'
Inconsistency detected by ld.so: ../sysdeps/x86_64/dl-machine.h: 534: elf_machine_rela_relative: Assertion ELFW(R_TYPE) (reloc->r_info) == R_X86_64_RELATIVE' failed!
nvidia-container-cli: detection error: driver rpc error: failed to process request: unknown
2024-02-27T08:13:58Z start container`Provide pod id so RunPod can investigate.
emmm, sorry, I forgot to record the ID. If this happens again, I will make sure to record it
hi, @ashleyk I have found another pod with the same issue, pod id : 840b98harmlsgb
I haven't deleted it yet
Great, RunPod can check when they come online, I dont work for RunPod.
May be a few hours because they are in the US.
thanks for you help👍 , Let me wait until the runpod developers are online😢
Maybe @Papa Madiator can help escalate for you when they come online.
Could you submit ticket on website it's better for hardware errors
https://www.runpod.io/
Rent Cloud GPUs from $0.2/hour
Save over 80% on GPUs. GPU rental made easy with Jupyter for Tensorflow, PyTorch or any other AI framework.
ok, let me find out how to submit a ticket.
when you press login
thanks!
@otakuhero could you also send your runpod email inside ticket on website?
going to grab that machine id so you can turn off pod
@otakuhero is pod issue happening on team account?
yes , on team account
I can shut down this pod now, right?
nvm do not have access to vie machine id. You can turn off pod but not delete it yet
emmm, any results? Can I delete this pod?:poddy: 😢
@Papa Madiator hi, I encountered the same problem again, two pods failed to start, pod id:wqaz2xufma32pt & eqyabu82t6l3y9, I'm not sure whether it's caused by my custom images or infrastructure such as physical machines....
