# split with multiple columns in debug **URL:** \ **Category:** Ansible Project\ **Created:** [February 27, 2023, 4:46pm UTC](https://forum.ansible.com/t/split-with-multiple-columns-in-debug/37701 "2023-02-27T16:46:54Z")\ **Posts on this page:** 11\ **Page:** 1
**Author:** ![Veera](https://sea1.discourse-cdn.com/ansible/user_avatar/forum.ansible.com/veera/32/4107_2.png) [@Veera](https://forum.ansible.com/u/Veera)\ **Post date:** [February 27, 2023, 4:46pm UTC](https://forum.ansible.com/t/split-with-multiple-columns-in-debug/37701/1 "2023-02-27T16:46:54Z")
Hi, When I try to split the output of uptime to capture the "load average from the below line $ uptime 18:37:01 up 5 days, 4:37, 2 users, load average: 0.02, 0.05, 0.00 using the below playbook ---
**Author:** ![Walter\_Rowe](https://sea1.discourse-cdn.com/ansible/user_avatar/forum.ansible.com/walter_rowe/32/4506_2.png) [@Walter\_Rowe](https://forum.ansible.com/u/Walter_Rowe)\ **Post date:** [February 27, 2023, 5:27pm UTC](https://forum.ansible.com/t/split-with-multiple-columns-in-debug/37701/2 "2023-02-27T17:27:34Z")
> load\_avg: “{{ up\_out.stdout\_lines[0].split()[7:] }}” Just use [7:] to get from position 7 on in your list. ---
**Author:** ![shearn](https://avatars.discourse-cdn.com/v4/letter/s/ba8739/32.png) [@shearn](https://forum.ansible.com/u/shearn)\ **Post date:** [February 27, 2023, 6:00pm UTC](https://forum.ansible.com/t/split-with-multiple-columns-in-debug/37701/3 "2023-02-27T18:00:32Z")
Veera, Be careful splitting uptime on white space because when a system is up less than 24 hours, it will show 12:58:10 up 6:51, 0 users, load average: 1.08, 0.99, 1.05 which will have one less white space and your desired output will be position 6. I would recommend splitting the output on comma. The fourth value will always be load average. Regards, Stan ---
**Author:** ![Walter\_Rowe](https://sea1.discourse-cdn.com/ansible/user_avatar/forum.ansible.com/walter_rowe/32/4506_2.png) [@Walter\_Rowe](https://forum.ansible.com/u/Walter_Rowe)\ **Post date:** [February 27, 2023, 6:09pm UTC](https://forum.ansible.com/t/split-with-multiple-columns-in-debug/37701/4 "2023-02-27T18:09:06Z")
An even more elegant solution would be to use ‘[-3:]’ to get the last three items regardless of how many there are. ---
**Author:** ![Walter\_Rowe](https://sea1.discourse-cdn.com/ansible/user_avatar/forum.ansible.com/walter_rowe/32/4506_2.png) [@Walter\_Rowe](https://forum.ansible.com/u/Walter_Rowe)\ **Post date:** [February 27, 2023, 6:12pm UTC](https://forum.ansible.com/t/split-with-multiple-columns-in-debug/37701/5 "2023-02-27T18:12:37Z")
**% cat split.yml** ---
**Author:** ![Walter\_Rowe](https://sea1.discourse-cdn.com/ansible/user_avatar/forum.ansible.com/walter_rowe/32/4506_2.png) [@Walter\_Rowe](https://forum.ansible.com/u/Walter_Rowe)\ **Post date:** [February 27, 2023, 6:18pm UTC](https://forum.ansible.com/t/split-with-multiple-columns-in-debug/37701/6 "2023-02-27T18:18:48Z")
If you want the commas removed you can do that too. - “{{ text\_to\_split1.replace(‘,’,‘’).split()[-3:] }}” - “{{ text\_to\_split2.replace(‘,’,‘’).split()[-3:] }}” ok: [localhost] =\> { “msg”: [ [ “0.02”, “0.05”, “0.00” ], [ “1.08”, “0.99”, “1.05” ] ] } Walter ---
**Author:** ![Veera](https://sea1.discourse-cdn.com/ansible/user_avatar/forum.ansible.com/veera/32/4107_2.png) [@Veera](https://forum.ansible.com/u/Veera)\ **Post date:** [February 28, 2023, 5:07am UTC](https://forum.ansible.com/t/split-with-multiple-columns-in-debug/37701/7 "2023-02-28T05:07:45Z")
Thanks a lot for the valuable input… ---
**Author:** ![Veera](https://sea1.discourse-cdn.com/ansible/user_avatar/forum.ansible.com/veera/32/4107_2.png) [@Veera](https://forum.ansible.com/u/Veera)\ **Post date:** [February 28, 2023, 5:24am UTC](https://forum.ansible.com/t/split-with-multiple-columns-in-debug/37701/8 "2023-02-28T05:24:52Z")
Thanks Walter… the options helped me a lot. ---
**Author:** ![vbotka](https://sea1.discourse-cdn.com/ansible/user_avatar/forum.ansible.com/vbotka/32/359_2.png) [@vbotka](https://forum.ansible.com/u/vbotka)\ **Post date:** [February 28, 2023, 7:11am UTC](https://forum.ansible.com/t/split-with-multiple-columns-in-debug/37701/9 "2023-02-28T07:11:23Z")
Given the registered variable \*up\_out\*   - command: uptime     register: up\_out Use the filter community.general.jc to parse the stdout. The utility \*jc\* "converts the output of many commands, file-types, and strings to JSON or YAML". For example,   uptime: "{{ up\_out.stdout|community.general.jc('uptime') }}" gives   uptime:     load\_15m: 1.21     load\_1m: 1.84     load\_5m: 1.41     time: 07:57:41     time\_hour: 7     time\_minute: 57     time\_second: 41     uptime: 11 days, 18:12     uptime\_days: 11     uptime\_hours: 18     uptime\_minutes: 12     uptime\_total\_seconds: 1015920     users: 2 The first three attributes come from "load average". Quoting from man uptime: "The averages are taken over the three time intervals." Now, the usage is trivial. For example,   load\_average: \>     {{ uptime.load\_1m }},     {{ uptime.load\_5m}},     {{ uptime.load\_15m }} gives   load\_average: |-     1.84, 1.41, 1.21 ---
**Author:** ![Walter\_Rowe](https://sea1.discourse-cdn.com/ansible/user_avatar/forum.ansible.com/walter_rowe/32/4506_2.png) [@Walter\_Rowe](https://forum.ansible.com/u/Walter_Rowe)\ **Post date:** [February 28, 2023, 12:53pm UTC](https://forum.ansible.com/t/split-with-multiple-columns-in-debug/37701/10 "2023-02-28T12:53:24Z")
Excellent find Vladimir! I just shared this with my team. I was unaware of ‘jc’. Walter ---
**Author:** ![Veera](https://sea1.discourse-cdn.com/ansible/user_avatar/forum.ansible.com/veera/32/4107_2.png) [@Veera](https://forum.ansible.com/u/Veera)\ **Post date:** [March 1, 2023, 6:38am UTC](https://forum.ansible.com/t/split-with-multiple-columns-in-debug/37701/11 "2023-03-01T06:38:11Z")
Thanks Vladimir !!!