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
At least on some machines (notably Briggs&Stratton), the /tmp filesystem on the BMC doesn't have much space, so when we rsync over P8DTU.pnor before flashing it, we use up a fair amount of /tmp space. So much, that if I invoke op-test again to flash a new BOOTKERNEL, there's not enough room in /tmp to transfer the bootkernel over.
With ENOSPC transferring images, we don't really pick up that error elegantly and you have to resort to the op-test debug log to find out what went wrong.
We should possibly remove images (such as pnor and individual LIDs) from /tmp on the bmc after we've flashed them to help avoid these kinds of problems.
The text was updated successfully, but these errors were encountered:
At least on some machines (notably Briggs&Stratton), the /tmp filesystem on the BMC doesn't have much space, so when we rsync over P8DTU.pnor before flashing it, we use up a fair amount of /tmp space. So much, that if I invoke op-test again to flash a new BOOTKERNEL, there's not enough room in /tmp to transfer the bootkernel over.
With ENOSPC transferring images, we don't really pick up that error elegantly and you have to resort to the op-test debug log to find out what went wrong.
We should possibly remove images (such as pnor and individual LIDs) from /tmp on the bmc after we've flashed them to help avoid these kinds of problems.
The text was updated successfully, but these errors were encountered: