You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I'm trying to modify a ubuntu raspberry pi base image.
It rather frequently happens that not all devices appear causing packer to fail the build.
My assumption is that there is a race condition of losetup and reading the dev devices in code again causing only some partitions to be available at the time of reading.
This fixes issue solo-io#162 and solo-io#165. I refactored the timeout handling,
with solo-io#169 by increasing the existing timeout and removing the additional 2 sec
sleep, because the 2 sec sleep was not working in all cases.
Unfortunately, the increase alone is not sufficient, so I'm re-adding
the 2 sec sleep again.
This fixes issue #162 and #165. I refactored the timeout handling,
with #169 by increasing the existing timeout and removing the additional 2 sec
sleep, because the 2 sec sleep was not working in all cases.
Unfortunately, the increase alone is not sufficient, so I'm re-adding
the 2 sec sleep again.
I'm trying to modify a ubuntu raspberry pi base image.
It rather frequently happens that not all devices appear causing packer to fail the build.
My assumption is that there is a race condition of losetup and reading the dev devices in code again causing only some partitions to be available at the time of reading.
By looking at https://github.com/solo-io/packer-plugin-arm-image/blob/master/pkg/builder/step_map_image.go
I think adding a short sleep could stabilize on that situation. Another option would be to really read the number of partitions and retry several times until a timeout value.
What do you think ?
The text was updated successfully, but these errors were encountered: