2018/04/26 23:32:58 [crit] 1457#0: *259351 open() “/usr/share/nginx/html/50x.html” failed (24: Too many open files), client: 112.17.247.117, server: ***, request: “GET /*** HTTP/1.1”, upstream: “***”, host: “***”
worker_rlimit_nofile 65535;
2018/04/26 23:32:58 [crit] 1457#0: *259351 open() “/usr/share/nginx/html/50x.html” failed (24: Too many open files), client: 112.17.247.117, server: ***, request: “GET /*** HTTP/1.1”, upstream: “***”, host: “***”
worker_rlimit_nofile 65535;
while read line读取文件时,如果文件最后一行之后没有换行符\n,则read读取最后一行时遇到文件结束符EOF,循环终止,虽然此时$line内存有最后一行,但程序已经没有机会再处理此行,因此可以通过以下代码来解决此问题:
while read line || [[ -n ${line} ]]; do
…
done
这样当文件没有结束时不会测试-n $line,当遇到文件结束时,仍然可以通过测试$line是否有内容来进行继续处理。
http://ju.outofmemory.cn/entry/86710
#!/bin/sh
# Updates etc at: https://github.com/woxxy/MySQL-backup-to-Amazon-S3
# Under a MIT license
# change these variables to what you need
MYSQLROOT=XXX
MYSQLPASS=XXXX
S3BUCKET=xx-mysql-backup
# DATABASE='--all-databases'
DATABASE=${1-default}
FILENAME=${DATABASE}
HOST=master
DUMPOPTION='--quick --single-transaction'
# the following line prefixes the backups with the defined directory. it must be blank or end with a /
S3PATH=mysql-backup/${DATABASE}
# when running via cron, the PATHs MIGHT be different. If you have a custom/manual MYSQL install, you should set this manually like MYSQLDUMPPATH=/usr/local/mysql/bin/
MYSQLDUMPPATH=
#tmp path.
TMP_PATH=/tmp/
DATESTAMP=$(date +".%Y-%m-%d.%H:%M:%S")
DELDATESTAMP=$(date -d"15 day ago 2017-04-16" +".%Y-%m-%d.%H:%M:%S")
DAY=$(date +"%d")
DAYOFWEEK=$(date +"%A")
PERIOD=${2-day}
if [ ${PERIOD} = "auto" ]; then
if [ ${DAY} = "01" ]; then
PERIOD=month
elif [ ${DAYOFWEEK} = "Sunday" ]; then
PERIOD=week
else
PERIOD=day
fi
fi
echo "Selected period: $PERIOD."
echo "Starting backing up the database to a .gz file..."
# dump all databases
${MYSQLDUMPPATH}mysqldump -h${HOST} ${DUMPOPTION} --user=${MYSQLROOT} --password=${MYSQLPASS} ${DATABASE} | gzip > ${TMP_PATH}${FILENAME}.gz
echo "Done backing up the database to a file."
# echo "Starting compression..."
# tar czf ${TMP_PATH}${FILENAME}${DATESTAMP}.tar.gz ${TMP_PATH}${FILENAME}.sql
mv ${TMP_PATH}${FILENAME}.gz ${TMP_PATH}${FILENAME}${DATESTAMP}.gz
# echo "Done compressing the backup file."
# upload all databases
echo "Uploading the new backup..."
s3cmd put -f --check-md5 -s --continue-put ${TMP_PATH}${FILENAME}${DATESTAMP}.gz s3://${S3BUCKET}/${S3PATH}${PERIOD}/
echo "New backup uploaded."
if [ $? -ne 0 ]
then
echo "Re uploading the backup file..."
s3cmd put -f --check-md5 -s --continue-put ${TMP_PATH}${FILENAME}${DATESTAMP}.gz s3://${S3BUCKET}/${S3PATH}${PERIOD}/
echo "Re upload backup file done."
fi
echo "Moving the backup from past $PERIOD to another folder..."
s3cmd mv --recursive s3://${S3BUCKET}/${S3PATH}${PERIOD}/${FILENAME}${DELDATESTAMP} s3://${S3BUCKET}/${S3PATH}previous_${PERIOD}/
echo "Past backup moved."
# we want at least two backups, two months, two weeks, and two days
echo "Removing old backup (15 ${PERIOD}s ago)..."
s3cmd del --recursive s3://${S3BUCKET}/${S3PATH}previous_${PERIOD}/
echo "Old backup removed."
# remove databases dump
# rm ${TMP_PATH}${FILENAME}.sql
echo "Removing the gz files..."
rm ${TMP_PATH}${FILENAME}${DATESTAMP}.gz
echo "Files removed."
echo "All done."
Job for mariadb.service failed because the control process exited with error code. See “systemctl status mariadb.service” and “journalctl -xe” for details.
MySQL 8.0启动报错,也是这个问题(已经chown了)
mysqld: File ‘./binlog.index’ not found (OS errno 13 – Permission denied)
chcon -R -t mysqld_db_t /mnt/mysql
参考页面:https://www.thegeekstuff.com/2016/05/move-mysql-directory/
实现应用内购买结算:
https://developer.android.google.cn/google/play/billing/billing_integrate.html?hl=zh-cn
字段参考:
https://developer.android.google.cn/google/play/billing/billing_reference.html?hl=zh-cn#getBuyIntent
# PHP服务端验证
$inappPurchaseData = Request::input('INAPP_PURCHASE_DATA');
$inappDataSignature = Request::input('INAPP_DATA_SIGNATURE');
$googlePublicKey = 'xxxx'; # google后台能获取到的
$publicKey = "-----BEGIN PUBLIC KEY-----" . PHP_EOL . chunk_split($googlePublicKey, 64, PHP_EOL) .
"-----END PUBLIC KEY-----";
$publicKeyHandle = openssl_get_publickey($publicKey);
$verify = openssl_verify($inappPurchaseData, base64_decode($inappDataSignature),
$publicKeyHandle, OPENSSL_ALGO_SHA1);
# $verify == 1 则验证通过, 0 表示验证失败
$remote_addr 可信时才能信任 X-Forwarded-For、$proxy_add_x_forwarded_for
nginx配置代理转发,部分配置如下
location / {
# index index.html index.htm index.php;
# try_files $uri $uri/ /index.php$request_uri;
proxy_set_header Host $host;
#proxy_set_header X-Real-IP $remote_addr;
#proxy_set_header X-Real-Port $remote_port;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_pass https://10.29.185.61:8063;
}
$proxy_add_x_forwarded_forthe “X-Forwarded-For” client request header field with the $remote_addr variable appended to it, separated by a comma. If the “X-Forwarded-For” field is not present in the client request header, the $proxy_add_x_forwarded_for variable is equal to the $remote_addr variable.
“ X-Forward-For”客户端请求头字段,后面附加 $remote _ addr 变量,中间用逗号分隔。如果客户端请求头中没有“ X-Forward-For”字段,则 $proxy_add_x_forwarded_for 变量等于 $remote_addr 变量。
GET_LOCK(str,timeout)
Tries to obtain a lock with a name given by the string str, using a timeout of timeout seconds. A negative timeout value means infinite timeout. The lock is exclusive. While held by one session, other sessions cannot obtain a lock of the same name.
参考:https://dev.mysql.com/doc/refman/5.7/en/miscellaneous-functions.html#function_get-lock
参考:https://blog.csdn.net/tangtong1/article/details/51792617
Install (centos7)
yum install python-setuptools && easy_install pip pip install git+https://github.com/shadowsocks/shadowsocks.git@master
Usage
ssserver -p 443 -k password -m aes-256-cfb
To run in the background:
sudo ssserver -p 443 -k password -m aes-256-cfb --user nobody -d start
To stop:
sudo ssserver -d stop
To check the log:
sudo less /var/log/shadowsocks.log
环境:php7.1、centos7、redis3.2.11
故障现象:php命令行环境下blpop 的时候是不是出现:blpop Error while reading line from the server.
blpop的等待时间是设置为了15秒
—-
Redis经常出现Error while reading line from the server.
cat /proc/sys/net/ipv4/tcp_timestamps
发现返回1
将其设置为0
echo 0 > /proc/sys/net/ipv4/tcp_timestamps
参考:http://blog.csdn.net/jueshengtianya/article/details/50440696
net.ipv4.tcp_timestamps=0
说明:时间戳可以避免序列号的卷绕。一个1Gbps的链路肯定会遇到以前用过的序列号。时间戳能够让内核接受这种“异常”的数据包。这里需要将其关掉。
值为0(禁用时间戳)
值为1(启用时间戳)
—-
然而进行以上操作以后还是会出现