Application logs filling disk in v2.0

Products

• Rally On-Premise

Issue/Introduction

In Rally on-premises 2.0, the log files are filling the disk and not being rotated out.

This article may also be used to clear out log files for other reasons, such as high growth.

 

Environment

Release : 2.0

Cause

This is a known issue in 2.0 but has been resolved in 2.1

Resolution

The resolution is to upgrade to version 2.1.0 of Rally on-premises.

 

As a workaround, it is possible to delete the files manually

SSH into the application server

Run the following command

$ docker ps | grep _alm

Sample Output:

17f009a362a1        <HOST_IP_ADDRESS>:9874/rallysoftware-ac2go-alm:23677-2.0.1                 "run /_run/start_ser…"   8 days ago          Up 8 days           7001/tcp                                      repl10aeecc4_alm.1.pg4ykc4066094y5dv44xnugfr

 

Make a note of the container ID in the far left

Use that container ID to run the following command:

$ docker exec -it 17f009a362a1 bash

Next change directory to the application logs directory with the following command:

$ cd /mnt/logs/alm

 

There are several classes of logs that can be deleted:

• access.log
• application.log
• gc.log
• ldap.log
• request-start.log
• request.log

 

If the docker containers will not start because the drive is completely full, it may be necessary to access the logs as they sit on the disk.  To access those logs, please follow this article:

https://ca-broadcomcsm.wolkenservicedesk.com/wolken/esd/knowledgebase_search?articleId=129526

 

Log files are compressed by date using year-month-date notation.

Only delete as many log files as you are comfortable purging.  It is recommended to maintain 30 days of logs if the size of the logs is not overly large and causing issues.

 

Log Deletion Examples:

To delete all application.log files from 2019, use a command as follows:

$ rm application.log.2019-*

 

To delete all access.log files from March of 2020, use a command as follows:

$ rm access.log.2020-03-*