weifeng liu created MESOS-6072:
----------------------------------
Summary: mesos-0.28.1 failed to launch wordpress with overlay
network
Key: MESOS-6072
URL: https://issues.apache.org/jira/browse/MESOS-6072
Project: Mesos
Issue Type: Bug
Components: docker
Affects Versions: 0.28.1
Environment: centos7 64bit
mesos-0.2.81
docker-1.11.0
Reporter: weifeng liu
Priority: Minor
centos 7, x86-64bit
mesos 0.2.81
docker 1.11.0
1. create a docker overlay network with subnet defined 10.10.0.0/16, overlay
network name is mynet
2. create mysql app on marathon, and the json used as below shows:
{
"id": "/db",
"cmd": null,
"cpus": 0.6,
"mem": 512,
"disk": 0,
"instances": 1,
"container": {
"type": "DOCKER",
"volumes": [],
"docker": {
"image": "mysql:5.7.14",
"network": "BRIDGE",
"portMappings": [
{
"containerPort": 3306,
"hostPort": 0,
"servicePort": 10001,
"protocol": "tcp",
"labels": {}
}
],
"privileged": false,
"parameters": [
{
"key": "net",
"value": "mynet"
},
{
"key": "net-alias",
"value": "db"
},
{
"key": "hostname",
"value": "db"
}
],
"forcePullImage": false
}
},
"env": {
"MYSQL_DATABASE": "wordpress",
"MYSQL_ROOT_PASSWORD": "password"
},
"portDefinitions": [
{
"port": 10001,
"protocol": "tcp",
"labels": {}
}
]
}
after a while, the mysql runs as expected!
3. create wordpress app on marathon, and the json is:
{
"id": "/server",
"cmd": null,
"cpus": 0.6,
"mem": 256,
"disk": 0,
"instances": 1,
"container": {
"type": "DOCKER",
"volumes": [],
"docker": {
"image": "wordpress",
"network": "BRIDGE",
"portMappings": [
{
"containerPort": 80,
"hostPort": 0,
"servicePort": 10004,
"protocol": "tcp",
"labels": {}
}
],
"privileged": false,
"parameters": [
{
"key": "net",
"value": "mynet"
},
{
"key": "net-alias",
"value": "server"
},
{
"key": "hostname",
"value": "server"
}
],
"forcePullImage": false
}
},
"env": {
"WORDPRESS_DB_HOST": "db:3306",
"WORDPRESS_DB_PASSWORD": "password"
},
"portDefinitions": [
{
"port": 10004,
"protocol": "tcp",
"labels": {}
}
]
}
once wordpress app started, the mysql app will be shutdown and then stared by
marathon, on the other hand, wordpress can not connected to db, and eventually
failed and started by marathon... this process will be repeated forever
from mysql log, it seems that mysql process is killed, and the error log is :
2016-08-23T06:35:18.271876Z 0 [Note] Event Scheduler: Loaded 0 events
2016-08-23T06:35:18.272052Z 0 [Note] mysqld: ready for connections.
Version: '5.7.14' socket: '/var/run/mysqld/mysqld.sock' port: 3306 MySQL Comm
unity Server (GPL)
06:36:11 UTC - mysqld got signal 6 ;
This could be because you hit a bug. It is also possible that this binary
or one of the libraries it was linked against is corrupt, improperly built,
or misconfigured. This error can also be caused by malfunctioning hardware.
Attempting to collect some information that could help diagnose the problem.
As this is a crash and something is definitely wrong, the information
collection process might fail.
key_buffer_size=8388608
read_buffer_size=131072
max_used_connections=1
max_threads=151
thread_count=1
connection_count=1
It is possible that mysqld could use up to
key_buffer_size + (read_buffer_size + sort_buffer_size)*max_threads = 68190 K b
ytes of memory
Hope that's ok; if not, decrease some variables in the equation.
Thread pointer: 0x7f9b8c000ae0
if the wordpress app is not started by marathon, but by docker command
directly, it will startup successfully without any error!
it's pretty wired, and it puzzled me for quite a long time.
the above bug can be reproduced with mysql:5.7.14 and above tags, mysql:5.7.13
and mysql:5.6 can run successfully.
Anyone can help me out?
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
