| Summary: | core-image-sato with systemd only has psplash sysvinit script left - decide what to do with psplash | ||
|---|---|---|---|
| Product: | [Build System, Metadata & Runtime] OE-Core | Reporter: | Ross Burton <ross.burton> |
| Component: | core | Assignee: | Unassigned <unassigned> |
| Status: | RESOLVED FIXED | QA Contact: | |
| Severity: | enhancement | ||
| Priority: | Medium | CC: | bluelightning, holger, joshuagloe, liezhi.yang, meta.mr.watcher, meta.watcher, Qi.Chen, radu.moisan, randy.macleod, richard.purdie, ross.burton, sgw |
| Version: | 1.4 | ||
| Target Milestone: | Future | ||
| Hardware: | All | ||
| OS: | Multiple | ||
| Whiteboard: | |||
| OS type for building Yocto: | --- | Type of Regression: | --- |
| Verified: | Documentation change: | No (bug/feature does not impact docs) | |
|
Description
Ross Burton
2013-04-25 15:32:35 UTC
Radu, can you take part 1 of this, the scoping? To get the list of recipes that need towk we can list all recipes that inherit update-rcd but don't inherit systemd. Then we can see what we can port from meta-systemd and what needs rewriting. yes, will compile a list asap I found the following, although I didn't go into each recipe in particular to check is systemd enabling makes sense. We need to enable systemd in (as grep'd for update-rc.d): acpid alsa-state apmd keymaps usbinit v86d ?? portmap init-ifupdown tcf-agent xserver-nodm-init modutils-initscripts ?? oprofileui-server Found some recipe that inherit systemd but not update-rc.d avahi wpa-supplicant util-linux Some recipes that I already have on poky-contrib (need rebasing/adjusting) xserver-nodm-init lighttpd cronie dropbear openssh dhcp keymaps http://cgit.openembedded.org/openembedded-core/commit/?id=5d90c5ebdb899b2951c97a94ff57867c1e491c15 Has this change been tested with online update? What happens if someone is doing opkg upgrade through a SSH connection and dropbear will be updated? Also http://cgit.openembedded.org/openembedded-core/commit/meta/recipes-core/dropbear?id=16b57e352f5844f301cc6c7ea4f87bf750c11d67 doesn't seem to be done inside the dropbearkeys.service Two packages in OE still have the problem, v86d and alsa-state. //Chen Qi The /etc/init.d/psplash.sh problem is related to Bug#4339 and should be solved there. psplash is a blocker issue. //Chen Qi FYI, there are now only two bbappends in meta-systemd left that have anything substantial in them - bluez4 and dhcp. We've already taken most things from the dhcp bbappend, but what's left is the dhclient service file (it may be questionable as to whether that's really needed though). http://cgit.openembedded.org/meta-openembedded/tree/meta-systemd/oe-core/recipes-connectivity/bluez http://cgit.openembedded.org/meta-openembedded/tree/meta-systemd/oe-core/recipes-connectivity/dhcp Chen, have you analysed core-image-sato recently? (In reply to comment #9) > Chen, have you analysed core-image-sato recently? Yes. The situation hasn't changed for a long time. The only script that's in /etc/init.d is psplash.sh. Someone sent out related patches about adding psplash services for systemd, but it seems that's not accepted. And then, if I recall correctly, someone adds plymouth in meta-oe, but it's not in oe-core. My question is, should we insist on psplash for sato images in OE, or we use plymouth when systemd is the init manager? Regards, Chen Qi |