How to monitor a Zimbra Collaboration Environment using pflogsumm, Telegraf, InfluxDB and Grafana

*Note: This project is a Community contribution, not tested neither supported officialy by Zimbra. Use it at your own risk.

---

Getting started

This dashboard contains multiples sections with the goal to monitor a full Zimbra Collaboration Server or Servers, we have some sections to monitor the Linux and machine overall performance, and one dedicated section just to monitor Zimbra Collaboration. Special thanks to Lex Rivera for his Linux System dashboard

Download the checkzimbraversion.sh scripts from the GitHub repository and save it on the next path:

/opt/zimbra/common/bin/checkzimbraversion.sh
chmod +x /opt/zimbra/common/bin/checkzimbraversion.sh

More information available in: https://github.com/jorgedlcruz/zimbra-grafana

Zimbra Collaboration Performance

• ZCS Version (Only 8.7 and above)
• Received Megabytes
• Total Emails/received
• Deferred
• Held
• Incoming
• Maildrop

Linux and machine performance:

• CPUs (defaults to all)
• Disks (per-disk IOPS)
• Network interfaces (packets, bandwidth, errors/drops)
• Mountpoints (space / inodes)

Coming next

This is just a v0.3 of this Dashboard, the next step will be to use the Zimbra SOAP API to obtain some extra information from the Zimbra Collaboration Environment, like:

• Number of Active Users
• Number of Inactive Users
• Number of Domains
• Number of Users with ActiveSync
• etc.

Collector Config

Sample /etc/telegraf.d/zimbra.conf with inputs for Zimbra Processes, Zimbra Scripts, and Linux System Monitoring:

# Read metrics about cpu usage
[[inputs.cpu]]
  percpu = true
  totalcpu = true
  fielddrop = ["time_*"]

# Read metrics about disk usage by mount point
[[inputs.disk]]
  ignore_fs = ["tmpfs", "devtmpfs"]

[[inputs.diskio]]

[[inputs.kernel]]

[[inputs.mem]]

[[inputs.processes]]

[[inputs.swap]]

[[inputs.system]]

[[inputs.procstat]]
  exe = "memcached"
  prefix = "memcached"

[[inputs.procstat]]
  exe = "java"
  prefix = "java"

[[inputs.procstat]]
  exe = "mysqld"
  prefix = "mysqld"

[[inputs.procstat]]
  exe = "slapd"
  prefix = "slapd"

[[inputs.procstat]]
  exe = "nginx"
  prefix = "nginx"

[[inputs.net]]

[[inputs.exec]]
  commands = ["/opt/zimbra/common/bin/checkzimbraversion.sh"]
  name_override = "zimbra_stats"
  data_format = "value"
  data_type = "string"

# # OpenLDAP cn=Monitor plugin
# # As zimbra user run the next to obatin the password zmlocalconfig -s zimbra_ldap_password ldap_master_url
 [[inputs.openldap]]
   host = "YOURZIMBRASERVERHOSTNAME"
   port = 389
   insecure_skip_verify = true
   bind_dn = "uid=zimbra,cn=admins,cn=zimbra"
   bind_password = "YOURZIMBRALDAPPASSWORD"
   reverse_metric_names = true

 [[inputs.postfix]]
    queue_directory = "/opt/zimbra/data/postfix/spool"
    interval = "1s"

• Download the grafana-zimbra-collaboration-dashboard.json JSON file and import it into your Grafana
• Change your data inside the Grafana if needed and enjoy :)

---

Coming next

This is just a v0.3 of this Dashboard, the next step will be to use the Zimbra SOAP API to obtain some extra information from the Zimbra Collaboration Environment, like:

• Number of Active Users
• Number of Inactive Users
• Number of Domains
• Number of Users with ActiveSync
• etc.

In next versions we will parse directly the logs and put the attempts of logins, and successful logins on a map.