Scripts and job logs
DETAILS: Tier: Free, Premium, Ultimate Offering: GitLab.com, Self-managed, GitLab Dedicated
You can use special syntax in script
sections to:
- Split long commands into multiline commands.
- Use color codes to make job logs easier to review.
- Create custom collapsible sections to simplify job log output.
script
Use special characters with Sometimes, script
commands must be wrapped in single or double quotes.
For example, commands that contain a colon (:
) must be wrapped in single quotes ('
).
The YAML parser needs to interpret the text as a string rather than
a "key: value" pair.
For example, this script uses a colon:
job:
script:
- curl --request POST --header 'Content-Type: application/json' "https://gitlab/api/v4/projects"
To be considered valid YAML, you must wrap the entire command in single quotes. If
the command already uses single quotes, you should change them to double quotes ("
)
if possible:
job:
script:
- 'curl --request POST --header "Content-Type: application/json" "https://gitlab/api/v4/projects"'
You can verify the syntax is valid with the CI Lint tool.
Be careful when using these characters as well:
-
{
,}
,[
,]
,,
,&
,*
,#
,?
,|
,-
,<
,>
,=
,!
,%
,@
,`
.
Ignore non-zero exit codes
When script commands return an exit code other than zero, the job fails and further commands do not execute.
Store the exit code in a variable to avoid this behavior:
job:
script:
- false || exit_code=$?
- if [ $exit_code -ne 0 ]; then echo "Previous command failed"; fi;
before_script
or after_script
for all jobs
Set a default You can use before_script
and after_script
with default
:
- Use
before_script
withdefault
to define a default array of commands that should run before thescript
commands in all jobs. - Use
after_script
with default to define a default array of commands that should run after any job completes or is canceled.
You can overwrite a default by defining a different one in a job. To ignore the default
use before_script: []
or after_script: []
:
default:
before_script:
- echo "Execute this `before_script` in all jobs by default."
after_script:
- echo "Execute this `after_script` in all jobs by default."
job1:
script:
- echo "These script commands execute after the default `before_script`,"
- echo "and before the default `after_script`."
job2:
before_script:
- echo "Execute this script instead of the default `before_script`."
script:
- echo "This script executes after the job's `before_script`,"
- echo "but the job does not use the default `after_script`."
after_script: []
after_script
commands if a job is cancelled
Skip
- Introduced in GitLab 17.0 with a flag named
ci_canceling_status
. Enabled by default. Requires GitLab Runner version 16.11.1.- Generally available in GitLab 17.3. Feature flag
ci_canceling_status
removed.
after_script
commands run if a job is canceled while the before_script
or script
section of that job are running.
The job's status in the UI is canceling
while the after_script
are executing,
and changes to canceled
after the after_script
commands complete. The $CI_JOB_STATUS
predefined variable has a value of canceled
while the after_script
commands are running.
To prevent after_script
commands running after canceling a job, configure the after_script
section to:
- Check the
$CI_JOB_STATUS
predefined variable at the start of theafter_script
section. - End execution early if the value is
canceled
.
For example:
job1:
script:
- my-script.sh
after_script:
- if [ "$CI_JOB_STATUS" == "canceled" ]; then exit 0; fi
- my-after-script.sh
Split long commands
You can split long commands into multiline commands to improve readability with
|
(literal) and >
(folded) YAML multiline block scalar indicators.
WARNING:
If multiple commands are combined into one command string, only the last command's
failure or success is reported.
Failures from earlier commands are ignored due to a bug.
To work around this, run each command as a separate script
item, or add an exit 1
command to each command string.
You can use the |
(literal) YAML multiline block scalar indicator to write
commands over multiple lines in the script
section of a job description.
Each line is treated as a separate command.
Only the first command is repeated in the job log, but additional
commands are still executed:
job:
script:
- |
echo "First command line."
echo "Second command line."
echo "Third command line."
The example above renders in the job log as:
$ echo First command line # collapsed multiline command
First command line
Second command line.
Third command line.
The >
(folded) YAML multiline block scalar indicator treats empty lines between
sections as the start of a new command:
job:
script:
- >
echo "First command line
is split over two lines."
echo "Second command line."
This behaves similarly to multiline commands without the >
or |
block
scalar indicators:
job:
script:
- echo "First command line
is split over two lines."
echo "Second command line."
Both examples above render in the job log as:
$ echo First command line is split over two lines. # collapsed multiline command
First command line is split over two lines.
Second command line.
When you omit the >
or |
block scalar indicators, GitLab concatenates non-empty
lines to form the command. Make sure the lines can run when concatenated.
Shell here documents work with the
|
and >
operators as well. The example below transliterates lower case letters
to upper case:
job:
script:
- |
tr a-z A-Z << END_TEXT
one two three
four five six
END_TEXT
Results in:
$ tr a-z A-Z << END_TEXT # collapsed multiline command
ONE TWO THREE
FOUR FIVE SIX
Add color codes to script output
Script output can be colored using ANSI escape codes, or by running commands or programs that output ANSI escape codes.
For example, using Bash with color codes:
job:
script:
- echo -e "\e[31mThis text is red,\e[0m but this text isn't\e[31m however this text is red again."
You can define the color codes in Shell environment variables, or even CI/CD variables, which makes the commands easier to read and reusable.
For example, using the same example as above and environment variables defined in a before_script
:
job:
before_script:
- TXT_RED="\e[31m" && TXT_CLEAR="\e[0m"
script:
- echo -e "${TXT_RED}This text is red,${TXT_CLEAR} but this part isn't${TXT_RED} however this part is again."
- echo "This text is not colored"
Or with PowerShell color codes:
job:
before_script:
- $esc="$([char]27)"; $TXT_RED="$esc[31m"; $TXT_CLEAR="$esc[0m"
script:
- Write-Host $TXT_RED"This text is red,"$TXT_CLEAR" but this text isn't"$TXT_RED" however this text is red again."
- Write-Host "This text is not colored"
Expand and collapse job log sections
- Support for output of multi-line command bash shell output Introduced in GitLab 16.5 behind the GitLab Runner feature flag,
FF_SCRIPT_SECTIONS
.
Job logs are divided into sections that can be collapsed or expanded. Each section displays the duration.
In the following example:
- Three sections have been collapsed and can be expanded.
- Three sections are expanded and can be collapsed.
Custom collapsible sections
You can create collapsible sections in job logs by manually outputting special codes that GitLab uses to delimit collapsible sections:
- Section start marker:
\e[0Ksection_start:UNIX_TIMESTAMP:SECTION_NAME\r\e[0K
+TEXT_OF_SECTION_HEADER
- Section end marker:
\e[0Ksection_end:UNIX_TIMESTAMP:SECTION_NAME\r\e[0K
You must add these codes to the script section of the CI configuration.
For example, using echo
:
job1:
script:
- echo -e "\e[0Ksection_start:`date +%s`:my_first_section\r\e[0KHeader of the 1st collapsible section"
- echo 'this line should be hidden when collapsed'
- echo -e "\e[0Ksection_end:`date +%s`:my_first_section\r\e[0K"
The escape syntax may differ depending on the shell that your runner uses.
For example if it is using Zsh, you may need to escape the special characters
with \\e
or \\r
.
In the example above:
-
date +%s
: Command that produces the Unix timestamp (for example1560896352
). -
my_first_section
: The name given to the section. The name can only be composed of letters, numbers, and the_
,.
, or-
characters. -
\r\e[0K
: Escape sequence that prevents the section markers from displaying in the rendered (colored) job log. They are displayed when viewing the raw job log, accessed in the upper-right corner of the job log by selecting Show complete raw ({doc-text}).-
\r
: carriage return (returns the cursor to the start of the line). -
\e[0K
: ANSI escape code to clear the line from the cursor position to the end of the line. (\e[K
alone does not work; the0
must be included).
-
Sample raw job log:
\e[0Ksection_start:1560896352:my_first_section\r\e[0KHeader of the 1st collapsible section
this line should be hidden when collapsed
\e[0Ksection_end:1560896353:my_first_section\r\e[0K
Sample job console log:
Use a script to improve display of collapsible sections
To remove the echo
statements that create the section markers from the job output,
you can move the job contents to a script file and invoke it from the job:
-
Create a script that can handle the section headers. For example:
# function for starting the section function section_start () { local section_title="${1}" local section_description="${2:-$section_title}" echo -e "section_start:`date +%s`:${section_title}[collapsed=true]\r\e[0K${section_description}" } # Function for ending the section function section_end () { local section_title="${1}" echo -e "section_end:`date +%s`:${section_title}\r\e[0K" } # Create sections section_start "my_first_section" "Header of the 1st collapsible section" echo "this line should be hidden when collapsed" section_end "my_first_section" # Repeat as required
-
Add the script to the
.gitlab-ci.yml
file:job: script: - source script.sh
Pre-collapse sections
You can make the job log automatically collapse collapsible sections by adding the collapsed
option to the section start.
Add [collapsed=true]
after the section name and before the \r
. The section end marker
remains unchanged:
- Section start marker with
[collapsed=true]
:\e[0Ksection_start:UNIX_TIMESTAMP:SECTION_NAME[collapsed=true]\r\e[0K
+TEXT_OF_SECTION_HEADER
- Section end marker (unchanged):
\e[0Ksection_end:UNIX_TIMESTAMP:SECTION_NAME\r\e[0K
Add the updated section start text to the CI configuration. For example,
using echo
:
job1:
script:
- echo -e "\e[0Ksection_start:`date +%s`:my_first_section[collapsed=true]\r\e[0KHeader of the 1st collapsible section"
- echo 'this line should be hidden automatically after loading the job log'
- echo -e "\e[0Ksection_end:`date +%s`:my_first_section\r\e[0K"
Troubleshooting
Syntax is incorrect
in scripts that use :
If you use a colon (:
) in a script, GitLab might output:
Syntax is incorrect
script config should be a string or a nested array of strings up to 10 levels deep
For example, if you use "PRIVATE-TOKEN: ${PRIVATE_TOKEN}"
as part of a cURL command:
pages-job:
stage: deploy
script:
- curl --header 'PRIVATE-TOKEN: ${PRIVATE_TOKEN}' "https://gitlab.example.com/api/v4/projects"
environment: production
The YAML parser thinks the :
defines a YAML keyword, and outputs the
Syntax is incorrect
error.
To use commands that contain a colon, you should wrap the whole command
in single quotes. You might need to change existing single quotes ('
) into double quotes ("
):
pages-job:
stage: deploy
script:
- 'curl --header "PRIVATE-TOKEN: ${PRIVATE_TOKEN}" "https://gitlab.example.com/api/v4/projects"'
environment: production
&&
in a script
Job does not fail when using If you use &&
to combine two commands together in a single script line, the job
might return as successful, even if one of the commands failed. For example:
job-does-not-fail:
script:
- invalid-command xyz && invalid-command abc
- echo $?
- echo "The job should have failed already, but this is executed unexpectedly."
The &&
operator returns an exit code of 0
even though the two commands failed,
and the job continues to run. To force the script to exit when either command fails,
enclose the entire line in parentheses:
job-fails:
script:
- (invalid-command xyz && invalid-command abc)
- echo "The job failed already, and this is not executed."
Multiline commands not preserved by folded YAML multiline block scalar
If you use the - >
folded YAML multiline block scalar to split long commands,
additional indentation causes the lines to be processed as individual commands.
For example:
script:
- >
RESULT=$(curl --silent
--header
"Authorization: Bearer $CI_JOB_TOKEN"
"${CI_API_V4_URL}/job"
)
This fails as the indentation causes the line breaks to be preserved:
$ RESULT=$(curl --silent # collapsed multi-line command
curl: no URL specified!
curl: try 'curl --help' or 'curl --manual' for more information
/bin/bash: line 149: --header: command not found
/bin/bash: line 150: https://gitlab.example.com/api/v4/job: No such file or directory
Resolve this by either:
-
Removing the extra indentation:
script: - > RESULT=$(curl --silent --header "Authorization: Bearer $CI_JOB_TOKEN" "${CI_API_V4_URL}/job" )
-
Modifying the script so the extra line breaks are handled, for example using shell line continuation:
script: - > RESULT=$(curl --silent \ --header \ "Authorization: Bearer $CI_JOB_TOKEN" \ "${CI_API_V4_URL}/job")
Job log output is not formatted as expected or contains unexpected characters
Sometimes the formatting in the job log displays incorrectly with tools that rely
on the TERM
environment variable for coloring or formatting. For example, with the mypy
command:
GitLab Runner runs the container's shell in non-interactive mode, so the shell's TERM
environment variable is set to dumb
. To fix the formatting for these tools, you can:
- Add an additional script line to set
TERM=ansi
in the shell's environment before running the command. - Add a
TERM
CI/CD variable with a value ofansi
.
after_script
section execution stops early and incorrect $CI_JOB_STATUS
values
In GitLab Runner 16.9.0 to 16.11.0:
- The
after_script
section execution sometimes stops too early. - The status of the
$CI_JOB_STATUS
predefined variable is incorrectly set asfailed
while the job is canceling.