'Twas brillig, and Anssi Hannula at 04/04/12 01:36 did gyre and gimble: > 04.04.2012 03:09, Anssi Hannula kirjoitti: >> 03.04.2012 16:41, Colin Guthrie kirjoitti: >>> Hi, >>> >>> With a memory of the infamous "initrd killer" problem that is still >>> lurking in the background somewhere, something related to theme updates >>> is messing things up: >>> >>> e.g. see the before and after listings for my /boot >>> >>> >>> [colin@jimmy code (master)]$ ll /boot >>> total 39784 >> [...] >>> lrwxrwxrwx 1 root root 31 Apr 3 09:22 initrd.img -> >>> initrd-3.3.1-desktop-1.mga2.img >> [...] >> >>> >>> After: >>> >>> [colin@jimmy code (master)]$ ll /boot >> [...] >>> -rw-r--r-- 1 root root 7556648 Apr 3 14:33 initrd.img >> [...] >> >> Reproduced: >> >> root 9723 0.0 0.0 25760 1372 pts/11 S+ 03:02 0:00 | >> \_ /bin/sh /usr/share/bootsplash/scripts/switch-themes >> Mageia-Default >> root 9745 0.2 0.2 93496 20724 pts/11 S+ 03:02 0:00 | >> \_ /usr/bin/perl /usr/sbin/bootloader-config --action >> update-splash >> root 30133 0.0 0.0 25760 1312 pts/11 S+ 03:06 0:00 | >> \_ /bin/sh >> /usr/share/bootsplash/scripts/make-boot-splash /boot/initrd.img 800 >> root 30143 0.0 0.0 25760 1340 pts/11 S+ 03:06 0:00 | >> \_ /bin/sh >> /usr/share/bootsplash/scripts/make-boot-splash-raw /boot/initrd.img >> Mageia-Default >> >> => bork >> >> So it seems like some recent change in DrakX which causes it to call >> make-boot-splash even for /boot/initrd.img... > > Actually... it would be my recentish change/fix in > /usr/share/bootsplash/scripts/make-boot-splash-raw in which $foo.img is > now replaced with > # gzip > $foo.img.tmp > # mv -f $foo.img.tmp $foo.img > while it previously was a direct > # gzip > $foo.img > > I've now added a "readlink -f" in the beginning of the script so that it > will operate on the symlink target directly, in bootsplash-3.3.8. Thanks > for noticing this.
No worries. As a followup question, do you think it'll end up regenerating some initrd's several times? e.g. I have two symlinks, initrd.img and initrd-desktop.img both pointing to the same initrd. Will this ultimately get regenerated three times, two of which are obviously redundant? Col -- Colin Guthrie colin(at)mageia.org http://colin.guthr.ie/ Day Job: Tribalogic Limited http://www.tribalogic.net/ Open Source: Mageia Contributor http://www.mageia.org/ PulseAudio Hacker http://www.pulseaudio.org/ Trac Hacker http://trac.edgewall.org/