Created attachment 4368 [details] screenshot following are the steps to reproduce issue. 1. plugin usb which contains live image burned 2. configure device BIOS to firstly boot from USB if necessary 3. boot the device and select option "Install" from boot menu 4. proceed through default install process 5. Remove USB, and reboot into new installed system. Expected : after completion of yocto image installation on system should be reboot after removed USB and pressed Enter. Actual : ERROR: target efi install failed required system rebooting by manually. Note: I have attached screenshot. This issue is came on NUC7 and corei7-64_MTURBOT64.
Anuj believe that this issue was due to this commit in OE-core: https://git.yoctoproject.org/cgit/cgit.cgi/poky/commit/?id=b6fbb3f3d4d6367b0fd7921078f67057551c7ede This takes out reboot, poweroff from initramfs because VIRTUAL-RUNTIME_init_manager isn't set to busybox. after done following changes as suggested by Anuj /meta/recipes-core/busybox/busybox/defconfig CONFIG_HALT=y CONFIG_POWEROFF=y CONFIG_REBOOT=y with this changes image is building successfully. Image successfully installed and booted on NUC7 and Mturtbot. In this new image doesn’t have boot and install issue.
So the problem is that busybox in initramfs doesn't have reboot because it's no longer built unless init manager is set to be busybox. I can't find a way to fix this properly. Does anyone have any suggestions? Why was the original change needed? Would setting up alternatives for reboot work for that use case?
(In reply to comment #2) > So the problem is that busybox in initramfs doesn't have reboot because it's > no longer built unless init manager is set to be busybox. > > I can't find a way to fix this properly. > > Does anyone have any suggestions? > > Why was the original change needed? Would setting up alternatives for reboot > work for that use case? I made that patch to avoid conflicts between systemd and busybox's reboot/halt/poweroff. I didn't take into consideration of initramfs which uses no init manager and requires reboot. And YES, setting up alternatives will work for that use case. When I made the patch, I chose moving the CONFIG_XXX into init.cfg over setting up alternatives, although both ways would solve the problem. As we've past feature freeze, I think what we can do is to move reboot out of init.cfg and also set update-alternatves for reboot in systemd. Any other suggestion?
Thanks Chen. This is now merged: https://git.yoctoproject.org/cgit/cgit.cgi/poky/commit/?id=ef4a417ab04779bd9fc8a6540a20cca7d14e98cf https://git.yoctoproject.org/cgit/cgit.cgi/poky/commit/?id=8421aede4fdd5132eb953a59099670f9ab1f7f01 I forgot to add the bug ID in commit message :(
This bug is observed in yocto- 2.5.2 relaese. image is downloaded from following URL https://autobuilder.yocto.io/pub/releases/yocto-2.5.2.rc1/machines/genericx86-64/core-image-sato-sdk-genericx86-64.hddimg following are the steps to reproduce issue. 1. plugin usb which contains live image burned 2. configure device BIOS to firstly boot from USB if necessary 3. boot the device and select option "Install" from boot menu 4. proceed through default install process 5. Remove USB, and reboot into new installed system. Expected : after completion of yocto image installation on system should be reboot after removed USB and pressed Enter. Actual : ERROR: target efi install failed required system rebooting by manually. Note: I have attached screenshot. This issue is came on NUC7 and corei7-64_MTURBOT64.
Created attachment 4409 [details] Screenshoot logs screenshot is attached.
is this something I need to back port or does someone with this board need to help find a fix?
See Armin's question.
(In reply to comment #7) > is this something I need to back port or does someone with this board need > to help find a fix? Yes, I think someone had sent a backport request too some time back. This isn't board specific. We need these commits: https://git.yoctoproject.org/cgit/cgit.cgi/poky/commit/?id=360bba5cc90f7404c517d4d5aabde8cf186917c5 https://git.yoctoproject.org/cgit/cgit.cgi/poky/commit/?id=8421aede4fdd5132eb953a59099670f9ab1f7f01 https://git.yoctoproject.org/cgit/cgit.cgi/poky/commit/?id=ef4a417ab04779bd9fc8a6540a20cca7d14e98cf
(In reply to comment #9) > (In reply to comment #7) > > is this something I need to back port or does someone with this board need > > to help find a fix? > > Yes, I think someone had sent a backport request too some time back. This > isn't board specific. We need these commits: > > https://git.yoctoproject.org/cgit/cgit.cgi/poky/commit/ > ?id=360bba5cc90f7404c517d4d5aabde8cf186917c5 > > https://git.yoctoproject.org/cgit/cgit.cgi/poky/commit/ > ?id=8421aede4fdd5132eb953a59099670f9ab1f7f01 > > https://git.yoctoproject.org/cgit/cgit.cgi/poky/commit/ > ?id=ef4a417ab04779bd9fc8a6540a20cca7d14e98cf We have all three are in Thud since Nov 2018. http://git.yoctoproject.org/cgit/cgit.cgi/poky/commit/meta/recipes-core/systemd?h=thud&id=8421aede4fdd5132eb953a59099670f9ab1f7f01 http://git.yoctoproject.org/cgit/cgit.cgi/poky/commit/meta/recipes-core/systemd?h=thud&id=360bba5cc90f7404c517d4d5aabde8cf186917c5 http://git.yoctoproject.org/cgit/cgit.cgi/poky/commit/meta/recipes-core/busybox?h=thud&id=ef4a417ab04779bd9fc8a6540a20cca7d14e98cf
Anything else?
I am closing this for now. Please re-open if it appears again.
.