Description

This article explains the meaning of the alarm, ' Host x RTC Battery failure' and how to overcome it.

Symptoms

Active Alarm: 

user@host> show chassis alarms
1 alarms currently active
Alarm time               Class  Description
2020-02-11 02:19:56 EST  Major  Host 1 RTC Battery failure

Solution

Root cause of the RTC Battery Failure

  1. A glitch in the upgrade process
  2. RE issue
  3. RTC Battery worn out

RTC battery --can be a CMOS battery, or a Litium battery -- powers the internal real-time clock hardware on a Routing Engine. The purpose of the real-time clock hardware is keep the date and time of the RE while it was powered-off, or removed from a system. It provides initial date and time information for the RE to use to set the initial system clock when the RE first boot up. The "RTC Battery failure" alarm indicates that the battery does not provide power to the real-time clock hardware. The battery failure results in the Real-Time Clock (RTC) getting reset. Resulting in the incorrect timestamp when the system boot up initially. Once the system is fully operational, the system update the system clock using NTP to set the date and time accurately.

While the "RTC Battery failure" is classified as a MAJOR alarm, it does not affect overall system functionalities.

QFX/ PTX device:

On QFX and PTX series devices running prior to Junos 17.4 versions, these messages could be a false alarm and may be safely ignored. In that case, one workaround would be to configure a filter to filter out this error message from logging into Syslog. Refer to KB9382 - How to prevent certain syslog messages from being written to the log file [juniper.net] .

On devices running versions after 17.4, follow the same troubleshooting procedure as mentioned for MX devices below.

MX device:

Re-attempt Upgrade Process:

In case this error has popped up post upgrade/downgrade, re-perform the same upgrade downgrade process by jumping to the previous version and then changing to the new version. There have been previous cases where the issue resolved after re-attempting the upgrade process even after a successful previous upgrade. Doing so, the alarm is likely to disappear due to its transient nature.

Try to stay on the latest recommended release when considering upgrades as suggested in KB21476 - Junos Software Versions - Suggested Releases to Consider and Evaluate [juniper.net] .

Battery Change:

The issue could lie with the battery as well. It is better to change the battery and check if the issue resolves. The battery should lie on top right side of the routing engine (sometimes could be built-in). Note that this battery is not hot-swappable, so you should perform the replacement under a maintenance window. Once the battery is replaced, the alarm will ideally disappear. Even if the alarm persists after changing the battery with a new one, you can ignore the alarm until the next Junos upgrade (since you know that the battery is new) when the alarm will clear.

Hardware Replacement:

In cases where an upgrade cannot be performed and replacing the battery is not favorable to your situation, you can change the RE with a new one. Please contact JTAC for an RMA of the RE. In certain cases, where RE is built-in in nature, the entire chassis/CB will have to be replaced.


ACX devices
For ACX devices or any other vmhost-based systems, you can check the RTC battery status using the below set of commands. 

start shell user root   >> login as root user
vhclient -s   >> login to the linux host
cat /proc/driver/rtc.  >> Check the details for RTC driver details. This would include the battery status as well

This would help in identifying if the battery is good. If yes, please try a vmhost reboot of the RE to clear the alarm.
Note :  For dual-RE systems, please ensure you do not reboot the current primary RE without switching over the RE mastership to the back-up RE 

Sample output from an ACX device : 

cat /proc/driver/rtc

rtc_time    : 14:08:19
rtc_date    : 2022-12-14
alrm_time    : 00:00:00
alrm_date    : 2015-01-02
alarm_IRQ    : no
alrm_pending  : no
update IRQ enabled   : no
periodic IRQ enabled  : no
periodic IRQ frequency : 1024
max user IRQ frequency : 64
24hr      : yes
periodic_IRQ  : no
update_IRQ   : no
HPET_emulated  : yes
BCD       : yes
DST_enable   : no
periodic_freq  : 1024
batt_status   : okay
 

Modification History

2023-04-26: Added solution for ACX devices.

Related Information