Question

我刚注意到我的kubelet日志上出现了很多错误，如下所示：

#!/bin/bash
command -v aws >/dev/null 2>&1 || { echo >&2 "This command requires aws but it's not installed or not in the path.  Aborting."; exit 1; }
bold=$(tput bold)
normal=$(tput sgr0)

if [ "$#" -ne 2 -a "$#" -ne 1 ]; then
    echo "usage: $0 <code file path> [lambda function name if not same]"
    exit 1
fi

if ! [ -e $1 ]; then
    echo "File $1 does not exist"
    exit 1
fi

ZIPFILE="lamzip.zip"
LAMBDA_FUNC="lambda_function.py"

if [ "$#" -ne 2 ]; then
    AWS_LAMBDA_NAME=$(basename $1)
    AWS_LAMBDA_NAME="${AWS_LAMBDA_NAME%.*}"
else
    AWS_LAMBDA_NAME=$2
fi

echo "Will upload your code file to aws lamda ${bold}$AWS_LAMBDA_NAME${normal}"

cp -f $1 /tmp/$LAMBDA_FUNC
rm -f /tmp/$ZIPFILE
zip -j /tmp/$ZIPFILE  /tmp/$LAMBDA_FUNC

CMD="aws lambda update-function-code --function-name $AWS_LAMBDA_NAME --zip-file fileb:///tmp/$ZIPFILE --publish"
echo executing: $CMD
$CMD

exit 0

我使用heapster来监控集群，似乎对'stats'的调用得到了答案，但由于未知原因，它被视为错误500.

我在裸机服务器上运行集群。

kubernetes版本：v1.1.3
  码头工人：1.9.1
  os：debian 8.1

Answer 1

对我来说似乎是一个错误。我已提交https://github.com/kubernetes/kubernetes/issues/18868

为什么kubelet无法从docker获取统计信息（错误500）？

1 个答案: