Summary: | net-wireless/bluez-5.51 - error: Symlink points to BuildRoot: /usr/lib/systemd/user/dbus-org.bluez.obex.service -> /var/tmp/portage/net-wireless/bluez-5.51/image/usr/lib/systemd/user/obex.service | ||
---|---|---|---|
Product: | Portage Development | Reporter: | Alessandro Barbieri <lssndrbarbieri> |
Component: | Core | Assignee: | Portage team <dev-portage> |
Status: | CONFIRMED --- | ||
Severity: | normal | CC: | adr, pacho |
Priority: | Normal | ||
Version: | unspecified | ||
Hardware: | All | ||
OS: | All | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- | |
Attachments: |
info
bluez-5.51:20191002-145446.log |
Description
Alessandro Barbieri
2019-10-02 17:22:43 UTC
Created attachment 591610 [details]
info
emerge --info
Created attachment 591612 [details]
bluez-5.51:20191002-145446.log
buildlog
I this a failure of rpm, portage or bluez? When building normally, symlink is pointing to proper place: $ ls -l /usr/lib/systemd/user/dbus-org.bluez.obex.service lrwxrwxrwx 1 root root 34 sep 29 16:57 /usr/lib/systemd/user/dbus-org.bluez.obex.service -> /usr/lib/systemd/user/obex.service But I have no idea about how this rpm creation is being done, if it's being run at the proper place It looks similar to bug 689184 @dev-portage, how are we supposed to call dosym? Rewriting symlinks is portage's job, per section 13.4.1 of PMS.
The part of the ebuild that triggers the error is where it passes the absolute path as the first argument to dosym here:
> for i in $(find "${ED}"/usr/$(get_libdir)/bluez/test -maxdepth 1 -type f ! -name "*.*"); do
> dosym "${i}" /usr/bin/bluez-"${i##*/}"
> done
*bump* Hi all, this bug is still present in bluez-5.72.ebuild: localhost ~ # ls -al /usr/bin/bluez-test-nap lrwxr-xr-x 1 root root 76 Mar 26 16:48 /usr/bin/bluez-test-nap -> /var/tmp/portage/net-wireless/bluez-5.72/image/usr/lib64/bluez/test/test-nap Kind regards! |