You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The problem occur only when using kubectl trace from the master branch and not when installing kubectl-trace with krew.
Steps to reproduce the problem:
Install kubectl-trace from source:
Run go get -u github.com/iovisor/kubectl-trace/cmd/kubectl-trace
Make sure that which kubectl-trace points to $GOPATH/bin/kubectl-trace
Run the following command:
kubectl trace -n $NS --serviceaccount=$SA run $NODE -e 'tracepoint:syscalls:sys_enter_exec* { printf("%-10u %-5d ", elapsed / 1000000, pid); join(args->argv); }' -a
Where $NS is set to the namespace, $SA the service account, $NODE the hostname of the node.
At the moment I am seeing the following output:
trace 03463dbb-4b7c-4f9e-ab09-95060d9377b6 created
/bpftrace/include/clang_workarounds.h:14:10: fatal error: 'linux/types.h' file not found
exit status 1
When running the same bptrace program from the node itself, it works as expected.
The problem occur only when using
kubectl trace
from the master branch and not when installingkubectl-trace
withkrew
.Steps to reproduce the problem:
kubectl-trace
from source:go get -u github.com/iovisor/kubectl-trace/cmd/kubectl-trace
which kubectl-trace
points to$GOPATH/bin/kubectl-trace
Where
$NS
is set to the namespace,$SA
the service account,$NODE
the hostname of the node.At the moment I am seeing the following output:
When running the same
bptrace
program from the node itself, it works as expected.Additional info can be found at this Slack thread.
cc @fntlnz
The text was updated successfully, but these errors were encountered: