Failed fetching headers

[keyolk]

Hi.
I'm trying to use kubectl-trace on my single node microk8s env

My system is

$ cat /etc/os-release
NAME="CentOS Linux"
VERSION="8 (Core)"
ID="centos"
ID_LIKE="rhel fedora"
VERSION_ID="8"
PLATFORM_ID="platform:el8"
PRETTY_NAME="CentOS Linux 8 (Core)"
ANSI_COLOR="0;31"
CPE_NAME="cpe:/o:centos:centos:8"
HOME_URL="https://www.centos.org/"
BUG_REPORT_URL="https://bugs.centos.org/"

CENTOS_MANTISBT_PROJECT="CentOS-8"
CENTOS_MANTISBT_PROJECT_VERSION="8"
REDHAT_SUPPORT_PRODUCT="centos"
REDHAT_SUPPORT_PRODUCT_VERSION="8"

$ uname -r
5.8.17

$ ls -l /lib/modules/$(uname -r)
total 3448
lrwxrwxrwx  1 root root     19 Nov  1 12:56 build -> /home/keyolk/src/linux-stable
drwxr-xr-x 13 root root    141 Nov  1 13:57 kernel
-rw-r--r--  1 root root 874945 Nov  1 14:00 modules.alias
-rw-r--r--  1 root root 839183 Nov  1 14:00 modules.alias.bin
-rw-r--r--  1 root root   8196 Nov  1 13:55 modules.builtin
-rw-r--r--  1 root root  10534 Nov  1 14:00 modules.builtin.bin
-rw-r--r--  1 root root  62087 Nov  1 13:55 modules.builtin.modinfo
-rw-r--r--  1 root root 278851 Nov  1 14:00 modules.dep
-rw-r--r--  1 root root 389143 Nov  1 14:00 modules.dep.bin
-rw-r--r--  1 root root    405 Nov  1 14:00 modules.devname
-rw-r--r--  1 root root  99528 Nov  1 13:55 modules.order
-rw-r--r--  1 root root    521 Nov  1 14:00 modules.softdep
-rw-r--r--  1 root root 422542 Nov  1 14:00 modules.symbols
-rw-r--r--  1 root root 510124 Nov  1 14:00 modules.symbols.bin
lrwxrwxrwx  1 root root     19 Nov  1 13:55 source -> /home/keyolk/src/linux-stable

$ ls /usr/src/kernels/$(uname -r)
Kconfig  Makefile  default_cpio_list  gen_init_cpio.c  gen_initramfs.sh  include  initramfs_data.S

$ zcat /proc/config.gz | grep -iE 'bpf|btf'
CONFIG_CGROUP_BPF=y
CONFIG_BPF=y
CONFIG_BPF_LSM=y
CONFIG_BPF_SYSCALL=y
CONFIG_ARCH_WANT_DEFAULT_BPF_JIT=y
CONFIG_BPF_JIT_ALWAYS_ON=y
CONFIG_BPF_JIT_DEFAULT_ON=y
CONFIG_NETFILTER_XT_MATCH_BPF=m
# CONFIG_BPFILTER is not set
CONFIG_NET_CLS_BPF=m
CONFIG_NET_ACT_BPF=m
CONFIG_BPF_JIT=y
CONFIG_BPF_STREAM_PARSER=y
CONFIG_LWTUNNEL_BPF=y
CONFIG_HAVE_EBPF_JIT=y
CONFIG_LSM="lockdown,yama,loadpin,safesetid,integrity,selinux,smack,tomoyo,apparmor,bpf"
CONFIG_DEBUG_INFO_BTF=y
CONFIG_BPF_EVENTS=y
# CONFIG_BPF_KPROBE_OVERRIDE is not set
CONFIG_TEST_BPF=m

$ kubectl-trace version
git commit: d34d1d586b110af718aeadc9f3213c78e543a961
build date: 2019-09-20 00:00:13 +0900 KST

What I did first run it without "--fetching header"

$ kubectl trace run dev-keyolk003-ncl -e "tracepoint:syscalls:sys_enter_* { @[probe] = count() }"
trace b89c8f3c-1e76-11eb-8da5-f670c888798e created

$ kubectl trace get
NAMESPACE       NODE                    NAME                                                    STATUS  AGE
default         dev-keyolk003-ncl       kubectl-trace-df593bfc-1e71-11eb-abba-f670c888798e      Failed  34m

$ kubectl trace logs kubectl-trace-b89c8f3c-1e76-11eb-8da5-f670c888798e
if your program has maps to print, send a SIGINT using Ctrl-C, if you want to interrupt the execution send SIGINT two times
/bpftrace/include/clang_workarounds.h:14:10: fatal error: 'linux/types.h' file not found
exit status 1

then tried with "--fetch-headers"

$ kubectl trace run dev-keyolk003-ncl --fetch-headers -e "tracepoint:syscalls:sys_enter_* { @[probe] = count() }"
trace df593bfc-1e71-11eb-abba-f670c888798e created

$  kubectl trace get
NAMESPACE       NODE                    NAME                                                    STATUS  AGE
default         dev-keyolk003-ncl       kubectl-trace-df593bfc-1e71-11eb-abba-f670c888798e      Failed  22s

$ kubectl get pod
NAME                                                       READY   STATUS       RESTARTS   AGE
kubectl-trace-df593bfc-1e71-11eb-abba-f670c888798e-hx4m9   0/1     Init:Error   0          35s
kubectl-trace-df593bfc-1e71-11eb-abba-f670c888798e-5vpsv   0/1     Init:Error   0          26s

$ kubectl describe pod kubectl-trace-df593bfc-1e71-11eb-abba-f670c888798e-5vpsv
Name:         kubectl-trace-df593bfc-1e71-11eb-abba-f670c888798e-5vpsv
Namespace:    default
Priority:     0
Node:         dev-keyolk003-ncl/10.113.85.167
Start Time:   Wed, 04 Nov 2020 16:46:48 +0900
Labels:       controller-uid=446013c5-4152-40c7-b325-bc711ea66115
              iovisor.org/kubectl-trace=kubectl-trace-df593bfc-1e71-11eb-abba-f670c888798e
              iovisor.org/kubectl-trace-id=df593bfc-1e71-11eb-abba-f670c888798e
              job-name=kubectl-trace-df593bfc-1e71-11eb-abba-f670c888798e
Annotations:  cni.projectcalico.org/podIP: 10.1.198.149/32
              cni.projectcalico.org/podIPs: 10.1.198.149/32
              iovisor.org/kubectl-trace: kubectl-trace-df593bfc-1e71-11eb-abba-f670c888798e
              iovisor.org/kubectl-trace-id: df593bfc-1e71-11eb-abba-f670c888798e
Status:       Failed
IP:           10.1.198.149
IPs:
  IP:           10.1.198.149
Controlled By:  Job/kubectl-trace-df593bfc-1e71-11eb-abba-f670c888798e
Init Containers:
  kubectl-trace-init:
    Container ID:   containerd://151cbfeafe487c560277a6074b15f1a4e8fa4024da6c8357beb32302762430e3
    Image:          quay.io/iovisor/kubectl-trace-init:latest
    Image ID:       quay.io/iovisor/kubectl-trace-init@sha256:5909add1e8b7cb1cca740bb587fc5f742c468e2b3dad49789660a47f14e7a1f5
    Port:           <none>
    Host Port:      <none>
    State:          Terminated
      Reason:       Error
      Exit Code:    1
      Started:      Wed, 04 Nov 2020 16:46:50 +0900
      Finished:     Wed, 04 Nov 2020 16:46:55 +0900
    Ready:          False
    Restart Count:  0
    Limits:
      cpu:     1
      memory:  1G
    Requests:
      cpu:        100m
      memory:     100Mi
    Environment:  <none>
    Mounts:
      /etc/lsb-release.host from lsb-release (ro)
      /etc/os-release.host from os-release (ro)
      /lib/modules from modules-dir (rw)
      /lib/modules.host from modules-host (ro)
      /usr/src/ from linux-headers-generated (rw)
      /var/run/secrets/kubernetes.io/serviceaccount from default-token-hqfpf (ro)
Conditions:
  Type              Status
  Initialized       False
  Ready             False
  ContainersReady   False
  PodScheduled      True
Volumes:
  program:
    Type:      ConfigMap (a volume populated by a ConfigMap)
    Name:      kubectl-trace-df593bfc-1e71-11eb-abba-f670c888798e
    Optional:  false
  usr-src-host:
    Type:          HostPath (bare host directory volume)
    Path:          /usr/src
    HostPathType:
  modules-host:
    Type:          HostPath (bare host directory volume)
    Path:          /lib/modules
    HostPathType:
  sys:
    Type:          HostPath (bare host directory volume)
    Path:          /sys
    HostPathType:
  lsb-release:
    Type:          HostPath (bare host directory volume)
    Path:          /etc/lsb-release
    HostPathType:
  os-release:
    Type:          HostPath (bare host directory volume)
    Path:          /etc/os-release
    HostPathType:
  modules-dir:
    Type:          HostPath (bare host directory volume)
    Path:          /var/cache/linux-headers/modules_dir
    HostPathType:
  linux-headers-generated:
    Type:          HostPath (bare host directory volume)
    Path:          /var/cache/linux-headers/generated
    HostPathType:
  default-token-hqfpf:
    Type:        Secret (a volume populated by a Secret)
    SecretName:  default-token-hqfpf
    Optional:    false
QoS Class:       Burstable
Node-Selectors:  <none>
Tolerations:     node.kubernetes.io/not-ready:NoExecute op=Exists for 300s
                 node.kubernetes.io/unreachable:NoExecute op=Exists for 300s
Events:
  Type    Reason     Age   From               Message
  ----    ------     ----  ----               -------
  Normal  Scheduled  44s   default-scheduler  Successfully assigned default/kubectl-trace-df593bfc-1e71-11eb-abba-f670c888798e-5vpsv to dev-keyolk003-ncl
  Normal  Pulling    43s   kubelet            Pulling image "quay.io/iovisor/kubectl-trace-init:latest"
  Normal  Pulled     42s   kubelet            Successfully pulled image "quay.io/iovisor/kubectl-trace-init:latest" in 951.832792ms
  Normal  Created    42s   kubelet            Created container kubectl-trace-init
  Normal  Started    42s   kubelet            Started container kubectl-trace-init

$ kubectl logs kubectl-trace-df593bfc-1e71-11eb-abba-f670c888798e-5vpsv -c kubectl-trace-init
+ LSB_FILE=/etc/lsb-release.host
+ OS_RELEASE_FILE=/etc/os-release.host
+ TARGET_DIR=/usr/src
+ HOST_MODULES_DIR=/lib/modules.host
++ uname -r
+ KERNEL_VERSION=5.8.17
+ [[ ! -e /lib/modules/.installed ]]
+ check_headers /lib/modules.host
+ modules_path=/lib/modules.host
++ uname -m
+ arch=x86_64
+ kdir=/lib/modules.host/5.8.17
+ [[ x86_64 == \x\8\6\_\6\4 ]]
+ arch=x86
+ [[ ! -e /lib/modules.host/5.8.17 ]]
+ [[ ! -e /lib/modules.host/5.8.17/source ]]
+ [[ ! -e /lib/modules.host/5.8.17/build ]]
+ return 1
+ install_headers
++ awk '/^NAME *= */ { gsub(/^NAME *= */, ""); print }' /etc/os-release.host
WARNING: Cannot find distro-specific headers for "CentOS Linux". Fetching generic headers.
+ distro='"CentOS Linux"'
+ case $distro in
+ echo 'WARNING: Cannot find distro-specific headers for "CentOS Linux". Fetching generic headers.'
+ install_generic_linux_headers
+ BUILD_DIR=/linux-generic-5.8.17
+ SOURCES_DIR=/usr/src/linux-generic-5.8.17
+ [[ ! -e /usr/src/linux-generic-5.8.17/.installed ]]
+ echo 'Installing kernel headers for generic kernel'
+ fetch_generic_linux_sources
++ echo 5.8.17
Installing kernel headers for generic kernel
++ awk '-vFS=[-+]' '{ print $1 }'
+ kernel_version=5.8.17
Fetching upstream kernel sources for 5.8.17.
++ echo 5.8.17
++ awk -vFS=. '{ print $1 }'
+ major_version=5
+ [[ 5.8.17 == *.0 ]]
+ echo 'Fetching upstream kernel sources for 5.8.17.'
+ mkdir -p /linux-generic-5.8.17
+ curl -sL https://www.kernel.org/pub/linux/kernel/v5.x/linux-5.8.17.tar.gz
+ tar --strip-components=1 -xzf - -C /linux-generic-5.8.17
tar: invalid magic
tar: short read

real    0m5.067s
user    0m0.005s
sys     0m0.008s

Any hints to solve it?

[keyolk]

seems that symlink /lib/modules/$(uname -r)/build make problem
after updating it, it solved : )

[lu-you]

I don't quite understand what you mean, could you give me detailed instructions

Thanks very much!

[lu-you]

I don't quite understand what you mean, could you give me detailed instructions

Thanks very much!

Hi. I'm trying to use kubectl-trace on my single node microk8s env

My system is

$ cat /etc/os-release
NAME="CentOS Linux"
VERSION="8 (Core)"
ID="centos"
ID_LIKE="rhel fedora"
VERSION_ID="8"
PLATFORM_ID="platform:el8"
PRETTY_NAME="CentOS Linux 8 (Core)"
ANSI_COLOR="0;31"
CPE_NAME="cpe:/o:centos:centos:8"
HOME_URL="https://www.centos.org/"
BUG_REPORT_URL="https://bugs.centos.org/"

CENTOS_MANTISBT_PROJECT="CentOS-8"
CENTOS_MANTISBT_PROJECT_VERSION="8"
REDHAT_SUPPORT_PRODUCT="centos"
REDHAT_SUPPORT_PRODUCT_VERSION="8"

$ uname -r
5.8.17

$ ls -l /lib/modules/$(uname -r)
total 3448
lrwxrwxrwx  1 root root     19 Nov  1 12:56 build -> /home/keyolk/src/linux-stable
drwxr-xr-x 13 root root    141 Nov  1 13:57 kernel
-rw-r--r--  1 root root 874945 Nov  1 14:00 modules.alias
-rw-r--r--  1 root root 839183 Nov  1 14:00 modules.alias.bin
-rw-r--r--  1 root root   8196 Nov  1 13:55 modules.builtin
-rw-r--r--  1 root root  10534 Nov  1 14:00 modules.builtin.bin
-rw-r--r--  1 root root  62087 Nov  1 13:55 modules.builtin.modinfo
-rw-r--r--  1 root root 278851 Nov  1 14:00 modules.dep
-rw-r--r--  1 root root 389143 Nov  1 14:00 modules.dep.bin
-rw-r--r--  1 root root    405 Nov  1 14:00 modules.devname
-rw-r--r--  1 root root  99528 Nov  1 13:55 modules.order
-rw-r--r--  1 root root    521 Nov  1 14:00 modules.softdep
-rw-r--r--  1 root root 422542 Nov  1 14:00 modules.symbols
-rw-r--r--  1 root root 510124 Nov  1 14:00 modules.symbols.bin
lrwxrwxrwx  1 root root     19 Nov  1 13:55 source -> /home/keyolk/src/linux-stable

$ ls /usr/src/kernels/$(uname -r)
Kconfig  Makefile  default_cpio_list  gen_init_cpio.c  gen_initramfs.sh  include  initramfs_data.S

$ zcat /proc/config.gz | grep -iE 'bpf|btf'
CONFIG_CGROUP_BPF=y
CONFIG_BPF=y
CONFIG_BPF_LSM=y
CONFIG_BPF_SYSCALL=y
CONFIG_ARCH_WANT_DEFAULT_BPF_JIT=y
CONFIG_BPF_JIT_ALWAYS_ON=y
CONFIG_BPF_JIT_DEFAULT_ON=y
CONFIG_NETFILTER_XT_MATCH_BPF=m
# CONFIG_BPFILTER is not set
CONFIG_NET_CLS_BPF=m
CONFIG_NET_ACT_BPF=m
CONFIG_BPF_JIT=y
CONFIG_BPF_STREAM_PARSER=y
CONFIG_LWTUNNEL_BPF=y
CONFIG_HAVE_EBPF_JIT=y
CONFIG_LSM="lockdown,yama,loadpin,safesetid,integrity,selinux,smack,tomoyo,apparmor,bpf"
CONFIG_DEBUG_INFO_BTF=y
CONFIG_BPF_EVENTS=y
# CONFIG_BPF_KPROBE_OVERRIDE is not set
CONFIG_TEST_BPF=m

$ kubectl-trace version
git commit: d34d1d586b110af718aeadc9f3213c78e543a961
build date: 2019-09-20 00:00:13 +0900 KST

What I did first run it without "--fetching header"

$ kubectl trace run dev-keyolk003-ncl -e "tracepoint:syscalls:sys_enter_* { @[probe] = count() }"
trace b89c8f3c-1e76-11eb-8da5-f670c888798e created

$ kubectl trace get
NAMESPACE       NODE                    NAME                                                    STATUS  AGE
default         dev-keyolk003-ncl       kubectl-trace-df593bfc-1e71-11eb-abba-f670c888798e      Failed  34m

$ kubectl trace logs kubectl-trace-b89c8f3c-1e76-11eb-8da5-f670c888798e
if your program has maps to print, send a SIGINT using Ctrl-C, if you want to interrupt the execution send SIGINT two times
/bpftrace/include/clang_workarounds.h:14:10: fatal error: 'linux/types.h' file not found
exit status 1

then tried with "--fetch-headers"

$ kubectl trace run dev-keyolk003-ncl --fetch-headers -e "tracepoint:syscalls:sys_enter_* { @[probe] = count() }"
trace df593bfc-1e71-11eb-abba-f670c888798e created

$  kubectl trace get
NAMESPACE       NODE                    NAME                                                    STATUS  AGE
default         dev-keyolk003-ncl       kubectl-trace-df593bfc-1e71-11eb-abba-f670c888798e      Failed  22s

$ kubectl get pod
NAME                                                       READY   STATUS       RESTARTS   AGE
kubectl-trace-df593bfc-1e71-11eb-abba-f670c888798e-hx4m9   0/1     Init:Error   0          35s
kubectl-trace-df593bfc-1e71-11eb-abba-f670c888798e-5vpsv   0/1     Init:Error   0          26s

$ kubectl describe pod kubectl-trace-df593bfc-1e71-11eb-abba-f670c888798e-5vpsv
Name:         kubectl-trace-df593bfc-1e71-11eb-abba-f670c888798e-5vpsv
Namespace:    default
Priority:     0
Node:         dev-keyolk003-ncl/10.113.85.167
Start Time:   Wed, 04 Nov 2020 16:46:48 +0900
Labels:       controller-uid=446013c5-4152-40c7-b325-bc711ea66115
              iovisor.org/kubectl-trace=kubectl-trace-df593bfc-1e71-11eb-abba-f670c888798e
              iovisor.org/kubectl-trace-id=df593bfc-1e71-11eb-abba-f670c888798e
              job-name=kubectl-trace-df593bfc-1e71-11eb-abba-f670c888798e
Annotations:  cni.projectcalico.org/podIP: 10.1.198.149/32
              cni.projectcalico.org/podIPs: 10.1.198.149/32
              iovisor.org/kubectl-trace: kubectl-trace-df593bfc-1e71-11eb-abba-f670c888798e
              iovisor.org/kubectl-trace-id: df593bfc-1e71-11eb-abba-f670c888798e
Status:       Failed
IP:           10.1.198.149
IPs:
  IP:           10.1.198.149
Controlled By:  Job/kubectl-trace-df593bfc-1e71-11eb-abba-f670c888798e
Init Containers:
  kubectl-trace-init:
    Container ID:   containerd://151cbfeafe487c560277a6074b15f1a4e8fa4024da6c8357beb32302762430e3
    Image:          quay.io/iovisor/kubectl-trace-init:latest
    Image ID:       quay.io/iovisor/kubectl-trace-init@sha256:5909add1e8b7cb1cca740bb587fc5f742c468e2b3dad49789660a47f14e7a1f5
    Port:           <none>
    Host Port:      <none>
    State:          Terminated
      Reason:       Error
      Exit Code:    1
      Started:      Wed, 04 Nov 2020 16:46:50 +0900
      Finished:     Wed, 04 Nov 2020 16:46:55 +0900
    Ready:          False
    Restart Count:  0
    Limits:
      cpu:     1
      memory:  1G
    Requests:
      cpu:        100m
      memory:     100Mi
    Environment:  <none>
    Mounts:
      /etc/lsb-release.host from lsb-release (ro)
      /etc/os-release.host from os-release (ro)
      /lib/modules from modules-dir (rw)
      /lib/modules.host from modules-host (ro)
      /usr/src/ from linux-headers-generated (rw)
      /var/run/secrets/kubernetes.io/serviceaccount from default-token-hqfpf (ro)
Conditions:
  Type              Status
  Initialized       False
  Ready             False
  ContainersReady   False
  PodScheduled      True
Volumes:
  program:
    Type:      ConfigMap (a volume populated by a ConfigMap)
    Name:      kubectl-trace-df593bfc-1e71-11eb-abba-f670c888798e
    Optional:  false
  usr-src-host:
    Type:          HostPath (bare host directory volume)
    Path:          /usr/src
    HostPathType:
  modules-host:
    Type:          HostPath (bare host directory volume)
    Path:          /lib/modules
    HostPathType:
  sys:
    Type:          HostPath (bare host directory volume)
    Path:          /sys
    HostPathType:
  lsb-release:
    Type:          HostPath (bare host directory volume)
    Path:          /etc/lsb-release
    HostPathType:
  os-release:
    Type:          HostPath (bare host directory volume)
    Path:          /etc/os-release
    HostPathType:
  modules-dir:
    Type:          HostPath (bare host directory volume)
    Path:          /var/cache/linux-headers/modules_dir
    HostPathType:
  linux-headers-generated:
    Type:          HostPath (bare host directory volume)
    Path:          /var/cache/linux-headers/generated
    HostPathType:
  default-token-hqfpf:
    Type:        Secret (a volume populated by a Secret)
    SecretName:  default-token-hqfpf
    Optional:    false
QoS Class:       Burstable
Node-Selectors:  <none>
Tolerations:     node.kubernetes.io/not-ready:NoExecute op=Exists for 300s
                 node.kubernetes.io/unreachable:NoExecute op=Exists for 300s
Events:
  Type    Reason     Age   From               Message
  ----    ------     ----  ----               -------
  Normal  Scheduled  44s   default-scheduler  Successfully assigned default/kubectl-trace-df593bfc-1e71-11eb-abba-f670c888798e-5vpsv to dev-keyolk003-ncl
  Normal  Pulling    43s   kubelet            Pulling image "quay.io/iovisor/kubectl-trace-init:latest"
  Normal  Pulled     42s   kubelet            Successfully pulled image "quay.io/iovisor/kubectl-trace-init:latest" in 951.832792ms
  Normal  Created    42s   kubelet            Created container kubectl-trace-init
  Normal  Started    42s   kubelet            Started container kubectl-trace-init

$ kubectl logs kubectl-trace-df593bfc-1e71-11eb-abba-f670c888798e-5vpsv -c kubectl-trace-init
+ LSB_FILE=/etc/lsb-release.host
+ OS_RELEASE_FILE=/etc/os-release.host
+ TARGET_DIR=/usr/src
+ HOST_MODULES_DIR=/lib/modules.host
++ uname -r
+ KERNEL_VERSION=5.8.17
+ [[ ! -e /lib/modules/.installed ]]
+ check_headers /lib/modules.host
+ modules_path=/lib/modules.host
++ uname -m
+ arch=x86_64
+ kdir=/lib/modules.host/5.8.17
+ [[ x86_64 == \x\8\6\_\6\4 ]]
+ arch=x86
+ [[ ! -e /lib/modules.host/5.8.17 ]]
+ [[ ! -e /lib/modules.host/5.8.17/source ]]
+ [[ ! -e /lib/modules.host/5.8.17/build ]]
+ return 1
+ install_headers
++ awk '/^NAME *= */ { gsub(/^NAME *= */, ""); print }' /etc/os-release.host
WARNING: Cannot find distro-specific headers for "CentOS Linux". Fetching generic headers.
+ distro='"CentOS Linux"'
+ case $distro in
+ echo 'WARNING: Cannot find distro-specific headers for "CentOS Linux". Fetching generic headers.'
+ install_generic_linux_headers
+ BUILD_DIR=/linux-generic-5.8.17
+ SOURCES_DIR=/usr/src/linux-generic-5.8.17
+ [[ ! -e /usr/src/linux-generic-5.8.17/.installed ]]
+ echo 'Installing kernel headers for generic kernel'
+ fetch_generic_linux_sources
++ echo 5.8.17
Installing kernel headers for generic kernel
++ awk '-vFS=[-+]' '{ print $1 }'
+ kernel_version=5.8.17
Fetching upstream kernel sources for 5.8.17.
++ echo 5.8.17
++ awk -vFS=. '{ print $1 }'
+ major_version=5
+ [[ 5.8.17 == *.0 ]]
+ echo 'Fetching upstream kernel sources for 5.8.17.'
+ mkdir -p /linux-generic-5.8.17
+ curl -sL https://www.kernel.org/pub/linux/kernel/v5.x/linux-5.8.17.tar.gz
+ tar --strip-components=1 -xzf - -C /linux-generic-5.8.17
tar: invalid magic
tar: short read

real    0m5.067s
user    0m0.005s
sys     0m0.008s

Any hints to solve it?

That's a problem I think I've solved！

kubectl trace ，There are two parameters，Look at help, I'm ignoring the alias here：

kubetrace run -h | grep head

kubectl trace run pod/nginx nginx -e "tracepoint:syscalls:sys_enter_* { @[probe] = count(); } --init-imagename=quay.io/custom-init-image-name --fetch-headers"
--fetch-headers Whether to fetch linux headers or not
--init-imagename string Custom image for the init container responsible to fetch and prepare linux headers (default "quay.io/iovisor/kubectl-trace-init:latest")

quay.io/iovisor/kubectl-trace-init:latest This image Can help us to get kernel headers; But because of the system，So you need to modify this script and re-create an image based on your own system！

scripts name is "fetch-linux-headers.sh";Modify several system paths
Before the change：
LSB_FILE="/etc/lsb-release.host"
OS_RELEASE_FILE="/etc/os-release.host"
TARGET_DIR="/usr/src"
HOST_MODULES_DIR="/lib/modules.host"
After the change：
LSB_FILE="/etc/lsb-release.host"
OS_RELEASE_FILE="/etc/os-release"
TARGET_DIR="/usr/src"
HOST_MODULES_DIR="/lib/modules"

The last rebuild your image；Then Specify the parameters

for example:

kubetrace run --serviceaccount=kubectltrace node/your_nodename --fetch-headers --init-imagename quay.io/iovisor/kubectl-trace-init:test --tracer bcc --program tcpconnect -a -n default --args "-p 9831" --args "-t"