News

Top ten Junos troubleshooting tips

As the single operating system that powers Juniper’s broad portfolio of physical and virtual networking and security products, Junos OS is used by hundreds of thousands of organisations across the world to run their networking operations in a reliable, secure and flexible way.

But what if your Junos device is experiencing unexpected or unexplained issues? Before you hit the panic button or seek help from other experts, try the handy tips below. Most of them are aimed at SME and Enterprise/Campus networks, but the commands are applicable to all Junos devices, from the very smallest to the very largest!

  1. show system uptime: My experience is that in normally stable enterprise networks, unexpected issues are often power related. Did someone knock a power cable whilst installing equipment in the rack next to your switches? “show system uptime” will show if your Junos device has recently power-cycled for any reason.

  2. show virtual-chassis status: Juniper Virtual Chassis technology is widely deployed in Enterprise networks on the EX and QFX series. This command help verify that all switches in your Virtual Chassis are present and operating correctly. If any devices are shown as “Not Present”, further investigation is required.

  3. show log messages | last 10: Junos devices have a Syslog-based logging system. This command shows the most recent ten messages and can give an indication of any recent issues. You can of course look further back in the log, or pipe though the “match” command to look for keywords.

  4. show chassis environment: Is your Junos device cool and happy? “show chassis environment” gives various temperature readings though your device, along with information on power and fans. If your fans show as “spinning at high speed”, you probably have insufficient cooling around your device, or maybe your Comms Room is running too hot.

  5. show chassis alarms: Is the Alarm LED lit on your Junos device? This command will show any hardware-based alarms, including power supply or fan issues, etc.

  6. show system alarms: This helps identify any software issues. Have you remembered to save your Junos Rescue configuration? Has your UTM license expired? This command is the best place to start.

  7. show interfaces descriptions | match down: It is best practice to add a description to your important interfaces (uplinks, WAN links, etc.). If you’ve done this, this command highlights any interface that is important enough to have a description, but is shown as “down”. Any such entries may require further investigation.

  8. show lacp interfaces: If you’re doing “link bundling” or “LAGs” using LACP (Link Aggregation Control Protocol), this command will help identify any redundant links in your bundle which are down when they should be up.

  9. show ethernet-switching table: If you’re troubleshooting a connectivity issue in an L2 environment, it helps to know if your switch has learned any MAC addresses on a given port; and if so, which VLAN they’re assigned to.

  10. show interfaces <if-name>: The simple “show interfaces” command returns a wealth of useful information, including:

  • “last flapped” – if an important interface recently went up or down unexpectedly, this should be investigated further.

  • “input rate” “output rate” (Packets Per Second)– If your WAN interface has a healthy PPS output rate, but zero input rate, this may indicate an issue with your WAN provider.

     

Alex Walker is Technical Consultant of SDN/NFV at Infradata UK, where he heads up our Automation practice, and is Development Team Lead for our nuQulus Automation Platform.

November 8 2017

Share this page:

This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Find out more here.