Zabbix is ChinaNetCloud's main monitoring system, which is heavily customized for their use. It monitors over 200 metrics per server across large-scale, globally distributed systems. Zabbix is considered one of the most advanced monitoring systems in the world, capable of monitoring billions of data points per month across servers, databases, applications and other infrastructure. It provides flexible dashboards, graphs, maps and alerts to give operations teams deep visibility into performance, issues and resource usage.
4. • World's Most Advanced Monitoring System
• Designed for Large-Scale Systems
• Heavily customized for our use
• 200+ points per server
• CPU, RAM, I/O, Network, Security
• Databases, Web Servers, Code, Java
• Globally distributed
• Proxies in several countries
• 25 million points/day
Zabbix Monitoring System
5. • Linux OS
• CPU, RAM, I/O
• Network
• Disks & RAID
• Load Balancers
• Web Servers
• PHP & FPM
• MySQL
• MongoDB
• Oracle & PostgreSQL
• MemCache & Redis
• Varnish & Squid
• RabbitMQ
• Switches & Firewalls
• SAN
Monitor Everything
6. Basic Monitoring Points
Apache
Apache Busy Workers
Apache Bytes per Request
Apache Bytes per Second
Apache CPULoad
Apache Idle Workers
Apache Request Per Seconds
Apache ScoreBoard
Apache Total Accesses
Apache Total kBytes
Availability
FS - Free disk space on /
FS - Free disk space on /home
FS - Free disk space on /opt
FS - Free disk space on /tmp
FS - Free disk space on /usr
FS - Free disk space on /var
FS - Free disk space on / in %
FS - Free disk space on /var in %
FS - Free disk space on /home in %
FS - Free disk space on /usr in %
FS - Free disk space on /opt in %
FS - Free disk space on /tmp in %
FS - Total disk space on /
FS - Total disk space on /home
FS - Total disk space on /opt
FS - Total disk space on /tmp
FS - Total disk space on /usr
FS - Used disk space on /
FS - Used disk space on /home
FS - Used disk space on /opt
FS - Used disk space on /tmp
FS - Used disk space on /usr
FS - Used disk space on /var
Memory buffers
Memory cached
Memory free
Memory shared
Memory total physical
Swap Free space
Swap Total space
CPU
CPU - idle time (avg1)
CPU - system time (avg1)
CPU - Load avg 15min
CPU - Load avg 1min
CPU - Load avg 5min
CPU - user time (avg1)
OS
Host boot time
Host information
Host local time
Host name
Host status
Host uptime (in sec)
Maximum number of opened files
Number of users connected
Performance
CPU - idle time (avg1)
CPU - system time (avg1)
CPU - Load avg 15min
CPU - Load avg 1min
CPU - Load avg 5min
CPU - user time (avg1)
Net - Incoming traffic on interface eth0
Net - Incoming traffic on interface lo
Net - Outgoing traffic on interface eth0
Net - Outgoing traffic on interface lo
Processes
Number of processes
Number of running processes
Number of running processes zabbix_agentd
Number of running processes inetd
Number of running processes sshd
Number of running processes syslogd
Version of zabbix_agent(d) running
sugarcrm
Download speed for scenario 'frontpage'
Download speed for step 'front-page connection' of
scenario 'frontpage'
Failed step of scenario 'frontpage'
Response code for step 'front-page connection' of
scenario 'frontpage'
Response time for step 'front-page connection' of scenario
'frontpage'
Vmstat
Amount of idle memory
Amount of memory used as buffers
Amount of memory used as cache
Amount of virtual memory used
CPU Time spent idle
CPU Time spent running kernel code
CPU Time spent running non-kernel code
CPU Time spent waiting for IO
IO Blocks in (bi)
IO Blocks out (bo)
Memory Swap In (si)
Memory Swap Out (so)
Number of context switches per second
Number of interrupts per second
Processes Blocked
Processes Runnable
Filesystem
FS - Free disk space on /
FS - Free disk space on /home
FS - Free disk space on /opt
FS - Free disk space on /tmp
FS - Free disk space on /usr
FS - Free disk space on /var
FS - Free disk space on / in %
FS - Free disk space on /var in %
FS - Free disk space on /home in %
FS - Free disk space on /usr in %
FS - Free disk space on /opt in %
FS - Free disk space on /tmp in %
FS - Total disk space on /
FS - Total disk space on /home
FS - Total disk space on /opt
FS - Total disk space on /tmp
FS - Total disk space on /usr
FS - Used disk space on /
FS - Used disk space on /home
FS - Used disk space on /opt
FS - Used disk space on /tmp
FS - Used disk space on /usr
FS - Used disk space on /var
General
Host boot time
Host information
Host local time
Host name
Host status
Host uptime (in sec)
Integrity
Checksum of /etc/passwd
Checksum of /etc/services
Checksum of /usr/bin/ssh
Checksum of /usr/sbin/sshd
Checksum of /etc/ntp.conf
Checksum of /etc/syslog-ng/syslog-ng.conf
Checksum of /etc/ssh/ssh_config
Checksum of /etc/sysctl.conf
Checksum of /etc/fstab
Checksum of /etc/sysconfig/network
Checksum of /etc/sysconfig/network-scripts/ifcfg-eth0
Checksum of /etc/hosts
Checksum of /etc/resolv.conf
Log files
Size of /var/log/maillog
Size of /var/log/secure
Size of /var/log/messages
Size of /var/log/kern.log
Size of /var/log/kern
Size of /var/log/wtmp
Size of /var/log/btmp
Memory
Memory buffers
Memory cached
Memory free
Memory shared
Memory total physical
Swap In Pages
Swap Out Pages
Network
Net - Incoming traffic on interface eth0
Net - Incoming traffic on interface lo
Net - Outgoing traffic on interface eth0
Net - Outgoing traffic on interface lo
27. Thank you – ChinaNetCloud
谢谢参观– 云络科技
Address: X2 Space 1-601, 1238 Xietu Road,
Xuhui District, Shanghai, China 200032
上海市徐汇区斜土路1238号X2创意空间1-601室
28. Narrative
• Zabbix is our core monitoring system, over 200
points per server. We run one of the world’s largest
monitoring systems with billions of data points per
month.
• Zabbix description, world’s most popular large-scale
• Our system is distributed
• Our system is customized on many levels
• We have full deep monitoring, including MySQL, Java
JMX, HAProxy, MongoDB, and much more.
• Key is our ticket & wiki integration
• Also automated tickets, automated setup, etc.
Editor's Notes
Hello and welcome to ChinaNetCloud. I am Steve Mushero and I’ll be your tour guide today as we walk through out offices in Shanghai.
ChinaNetCloud was founded in 2008 by two American technology entrepreneurs from Silicon Valley who also had many years of experience in China. The company’s founding goal was and always has been to run the world’s Internet servers, to provide world-class services, and to really help our customers succeed, in games, e-commerce, apps, media, and much more. Here you can see the front of our office and entryway, with our partners’ posters – we are a Linux shop, with thousands of Linux servers, on Red Hat, Centos, and Ubuntu, and with most databases on MySQL with some Oracle. We are China’s largest MySQL operator with deep experience up to terabytes of data and 25,000 queries per second per server.
ChinaNetCloud was founded in 2008 by two American technology entrepreneurs from Silicon Valley who also had many years of experience in China. The company’s founding goal was and always has been to run the world’s Internet servers, to provide world-class services, and to really help our customers succeed, in games, e-commerce, apps, media, and much more. Here you can see the front of our office and entryway, with our partners’ posters – we are a Linux shop, with thousands of Linux servers, on Red Hat, Centos, and Ubuntu, and with most databases on MySQL with some Oracle. We are China’s largest MySQL operator with deep experience up to terabytes of data and 25,000 queries per second per server.
ChinaNetCloud was founded in 2008 by two American technology entrepreneurs from Silicon Valley who also had many years of experience in China. The company’s founding goal was and always has been to run the world’s Internet servers, to provide world-class services, and to really help our customers succeed, in games, e-commerce, apps, media, and much more. Here you can see the front of our office and entryway, with our partners’ posters – we are a Linux shop, with thousands of Linux servers, on Red Hat, Centos, and Ubuntu, and with most databases on MySQL with some Oracle. We are China’s largest MySQL operator with deep experience up to terabytes of data and 25,000 queries per second per server.
ChinaNetCloud was founded in 2008 by two American technology entrepreneurs from Silicon Valley who also had many years of experience in China. The company’s founding goal was and always has been to run the world’s Internet servers, to provide world-class services, and to really help our customers succeed, in games, e-commerce, apps, media, and much more. Here you can see the front of our office and entryway, with our partners’ posters – we are a Linux shop, with thousands of Linux servers, on Red Hat, Centos, and Ubuntu, and with most databases on MySQL with some Oracle. We are China’s largest MySQL operator with deep experience up to terabytes of data and 25,000 queries per second per server.
ChinaNetCloud was founded in 2008 by two American technology entrepreneurs from Silicon Valley who also had many years of experience in China. The company’s founding goal was and always has been to run the world’s Internet servers, to provide world-class services, and to really help our customers succeed, in games, e-commerce, apps, media, and much more. Here you can see the front of our office and entryway, with our partners’ posters – we are a Linux shop, with thousands of Linux servers, on Red Hat, Centos, and Ubuntu, and with most databases on MySQL with some Oracle. We are China’s largest MySQL operator with deep experience up to terabytes of data and 25,000 queries per second per server.
ChinaNetCloud was founded in 2008 by two American technology entrepreneurs from Silicon Valley who also had many years of experience in China. The company’s founding goal was and always has been to run the world’s Internet servers, to provide world-class services, and to really help our customers succeed, in games, e-commerce, apps, media, and much more. Here you can see the front of our office and entryway, with our partners’ posters – we are a Linux shop, with thousands of Linux servers, on Red Hat, Centos, and Ubuntu, and with most databases on MySQL with some Oracle. We are China’s largest MySQL operator with deep experience up to terabytes of data and 25,000 queries per second per server.
ChinaNetCloud was founded in 2008 by two American technology entrepreneurs from Silicon Valley who also had many years of experience in China. The company’s founding goal was and always has been to run the world’s Internet servers, to provide world-class services, and to really help our customers succeed, in games, e-commerce, apps, media, and much more. Here you can see the front of our office and entryway, with our partners’ posters – we are a Linux shop, with thousands of Linux servers, on Red Hat, Centos, and Ubuntu, and with most databases on MySQL with some Oracle. We are China’s largest MySQL operator with deep experience up to terabytes of data and 25,000 queries per second per server.
We are an operations company, in that our every day focus is on Internet operations, building and managing the servers and systems that power the Internet at all levels. In our office, the first main area you’ll see is our Operations Area, dedicated to the teams that handle 7x24 server alerts, requests, and maintenance.
We are an operations company, in that our every day focus is on Internet operations, building and managing the servers and systems that power the Internet at all levels. In our office, the first main area you’ll see is our Operations Area, dedicated to the teams that handle 7x24 server alerts, requests, and maintenance.
We are an operations company, in that our every day focus is on Internet operations, building and managing the servers and systems that power the Internet at all levels. In our office, the first main area you’ll see is our Operations Area, dedicated to the teams that handle 7x24 server alerts, requests, and maintenance.
We are an operations company, in that our every day focus is on Internet operations, building and managing the servers and systems that power the Internet at all levels. In our office, the first main area you’ll see is our Operations Area, dedicated to the teams that handle 7x24 server alerts, requests, and maintenance.
We are an operations company, in that our every day focus is on Internet operations, building and managing the servers and systems that power the Internet at all levels. In our office, the first main area you’ll see is our Operations Area, dedicated to the teams that handle 7x24 server alerts, requests, and maintenance.
We are an operations company, in that our every day focus is on Internet operations, building and managing the servers and systems that power the Internet at all levels. In our office, the first main area you’ll see is our Operations Area, dedicated to the teams that handle 7x24 server alerts, requests, and maintenance.
We are an operations company, in that our every day focus is on Internet operations, building and managing the servers and systems that power the Internet at all levels. In our office, the first main area you’ll see is our Operations Area, dedicated to the teams that handle 7x24 server alerts, requests, and maintenance.
We are an operations company, in that our every day focus is on Internet operations, building and managing the servers and systems that power the Internet at all levels. In our office, the first main area you’ll see is our Operations Area, dedicated to the teams that handle 7x24 server alerts, requests, and maintenance.
We are an operations company, in that our every day focus is on Internet operations, building and managing the servers and systems that power the Internet at all levels. In our office, the first main area you’ll see is our Operations Area, dedicated to the teams that handle 7x24 server alerts, requests, and maintenance.
We are an operations company, in that our every day focus is on Internet operations, building and managing the servers and systems that power the Internet at all levels. In our office, the first main area you’ll see is our Operations Area, dedicated to the teams that handle 7x24 server alerts, requests, and maintenance.
We are an operations company, in that our every day focus is on Internet operations, building and managing the servers and systems that power the Internet at all levels. In our office, the first main area you’ll see is our Operations Area, dedicated to the teams that handle 7x24 server alerts, requests, and maintenance.
Key to our 7x24 reliability and system performance is one of the world’s largest monitoring systems, taking in billions of data points per month, about 200 per server at very detailed levels. With lots of advanced technology, the system predicts problems and provides detailed and customized instructions for engineers to follow quickly when there are problems. Here you can see part of our Alert Response and tracking area, with customer alerts, maintenance notices, team assignments, request dashboards, and top server loads. Plus our support coordinators and managers working with customers on various issues.
A closer look at the front-line engineers shows more monitoring, including numerous status screens, web views, load managers, request consoles, etc. You may see various colors such as shades of red or green, and at very high levels, rare emergencies or outages create special rapid responses and team coordination to quickly restore a customer’s system and solve problems rapidly.
We pride ourselves in our engineering teams and their expertise on all parts of the Internet stack, from hardware and data centers, to Linux and high-availability, to SQL and NOSQL databases, to caches, app servers, web servers, load balancers, and firewalls. Plus all the world’s and China’s public cloud systems. These engineers architect, design, engineer, and build new systems every day, using our fourth generation 500 point custom design and auto build system.
The rest of the core ChinaNetCloud Operations Floor is filed with a mix of engineers on our reliability, performance, and security engineering teams, each of which focuses on best-practices, world-class configurations, improved architectures, and processes that help ensure our customers’ reliability, performance, and security.
The last look at our Operations Area where infrastructure management teams work to manage and improve our monitoring, logging, procedure, troubleshooting, backup, audit, reporting, portal, ticket, repository, build, and 20 other systems we use every day to support our customers.
ChinaNetCloud is a very visible and well-known company, with articles and press in many places including quotes in the Wall Street Journal, lots of Chinese press, and more. We also often speak at industry events on the Internet, cloud computing, system operations, startups, and more.
And that’s it – this has been your tour of ChinaNetCloud’s offices in Shanghai. I hope you enjoyed this little visit and understand a bit more about what we do and how we do it. We look forward to your next visit in person.
For ChinaNetCloud, I’m Steve Mushero.