I am a free GitLab user trying to explore GitLab CI/CD. I tried building Yocto image using "bitbake core-image-minimal" with GitLab CI pipeline. Every time I am getting diskspace error. How the diskspace in pipeline can be increased or diskspace can be increased for only premium or pro customers? Please find the error attached below
NOTE: recipe libice-1_1.0.10-r0: task do_configure: Started
WARNING: The free space of /builds/ganeshredcobra/rcc_rpi/yocto/poky/build-qemuarm/tmp (/dev/sda1) is running low (0.994GB left)
ERROR: No new tasks can be executed since the disk space monitor action is "STOPTASKS"!
WARNING: The free space of /builds/ganeshredcobra/rcc_rpi/yocto/poky/build-qemuarm/downloads (/dev/sda1) is running low (0.994GB left)
ERROR: No new tasks can be executed since the disk space monitor action is "STOPTASKS"!
WARNING: The free space of /builds/ganeshredcobra/rcc_rpi/yocto/poky/build-qemuarm/sstate-cache (/dev/sda1) is running low (0.994GB left)
ERROR: No new tasks can be executed since the disk space monitor action is "STOPTASKS"!
NOTE: recipe libice-1_1.0.10-r0: task do_configure: Succeeded
WARNING: The free space of /tmp (overlay) is running low (0.089GB left)
ERROR: No new tasks can be executed since the disk space monitor action is "STOPTASKS"!
WARNING: The free space of /builds/ganeshredcobra/rcc_rpi/yocto/poky/build-qemuarm/tmp (/dev/sda1) is running low (0.089GB left)
ERROR: Immediately abort since the disk space monitor action is "ABORT"!
WARNING: The free space of /builds/ganeshredcobra/rcc_rpi/yocto/poky/build-qemuarm/downloads (/dev/sda1) is running low (0.089GB left)
WARNING: The free space of /builds/ganeshredcobra/rcc_rpi/yocto/poky/build-qemuarm/sstate-cache (/dev/sda1) is running low (0.089GB left)
ERROR: Immediately abort since the disk space monitor action is "ABORT"!
ERROR: Immediately abort since the disk space monitor action is "ABORT"!
NOTE: Sending SIGTERM to remaining 1 tasks
NOTE: Sending SIGTERM to remaining 1 tasks
NOTE: Sending SIGTERM to remaining 1 tasks
NOTE: Tasks Summary: Attempted 944 tasks of which 0 didn't need to be rerun and all succeeded.
Summary: There were 9 WARNING messages shown.
Summary: There were 7 ERROR messages shown, returning a non-zero exit code.
You can change the hardware used for your job on shared SaaS runners by using tags on your job(s).
For linux runners by default (at time of writing), saas-linux-small-amd64 runners are used, which have 25GB of storage. On the free tier, you can use medium sized instances by using the tag saas-linux-medium-amd64, which increases the disk space to 50GB. Larger instances are available for users with subscription plans, as documented in the link.
You can add the tags like so:
myjob:
tags:
- saas-linux-medium-amd64
# ...
If you need more disk space, you'll either need to host your own runner or get a gitlab subscription.
Be aware, larger runners consume your available CI minutes at a higher rate.
Also note that you also appear to be running out of tempfs space ("The free space of /tmp (overlay) is running low"), which is different than the available space in the regular filesystem where your job workspace resides (with the full 25GB or 50GB).
You can set your TEMPDIR environment variable to change where temporary files/directories are created. By default this is /tmp. Maybe you would do something like this if you wanted to avoid using tmpfs which usually has limited space:
myjob:
before_script:
- mkdir ./temp
- export TEMPDIR="$(pwd)/temp"