Custom Query (1145 matches)

Reported by wdc, 15 years ago.

Thirdparty software from ansoft maintained by others has a silly hard-coded test for /usr/lib/libGL.so

It turns out that the libgl1-mesa-dev adds this file, and has other value so rather than cutting into third party softare we'd like to add this to debathena-thirdparty.

See also testers-7831

Reported by xavid, 15 years ago.

   debathena / livecd / xavid  19:54  ('I want to be the duke of Chicago.`)
       Jaunty-compatible live cd is pushed and ready for someone to prod on 
       debuild.  If someone other than me wants to try the
       less-trusted-version, http://debathena-livecd-build.xvm/cds/
   debathena / livecd / xavid  19:59  ('Sucks-to-be-you-tastic.`)
       I guess I'll send mail if no one's particularly up for prodding it now.
   debathena / livecd / andersk  16:53  (Anders Kaseorg)
       debathena-livecd-convert runs `update-initramfs -u` from inside the
       live CD chroot.  As it turns out, this results in machines installed
       using the live CD, such as ozok-the-destroyer.mit.edu, to be fuxx0red
       in various ways (e.g. compcache is enabled).
   debathena / livecd / broder  16:54  (Evan Broder)
       Can you put this on the existing Live CD bug, or open a new one?
   debathena / livecd / jdreed  16:54  (This zephyr does not necessarily reflect
       A new one, please.  The existing one is a bit too general
   debathena / livecd.d / jdreed  16:55  (This zephyr does not necessarily refle
       I blame the person who opened the ticket for not picking a better
       subject line
   debathena / livecd.d / broder  16:55  (Evan Broder)
       Haha
   debathena / livecd / xavid  16:56  (“Werewolves stash their car keys. You don
       Is that called in the postinst of new kernels?
   debathena / livecd / andersk  16:56  (Anders Kaseorg)
       Yes, but you also call it explicitly.
   debathena / livecd / xavid  16:57  ("Never bite a gift horse in the neck.")
       Right, but I only started doing that a few days ago, and I didn't think
       ozok had been reinstalled in that time.
   debathena / livecd / ezyang  16:57  (Edward Z. Yang)
       Ozok was reinstalled on April 12th
   debathena / livecd / xavid  16:58  ('Do rocks chew their cud?')
       (Previously, the live CD would just be totally broken if there didn't
       exist a more recent kernel to upgrade to at build time.)
   debathena / livecd.d / jdreed  17:00  (This zephyr does not necessarily refle
       Is the LiveCD still a manual build process?  Weren't we going to do
       something about that?
   debathena / livecd / xavid  17:00  ('Look!  It's really happy!  And by happy,
       I can replace the explicit one with a call to mkinitramfs, but getting
       the postinst to DTRT will be more difficult.
   debathena / livecd / andersk  17:00  (Anders Kaseorg)
       Well, how do the Ubuntu guys do it?
   debathena / livecd.d / xavid  17:01  ('Sucks-to-be-you-tastic.`)
       The Live CD is automated but still has to be run manually, mostly
       because we're not past the point where we need to manually test each
       new iso
   debathena / livecd / xavid  17:02  (“Can vampires suck blood from rocks?”)
       No idea?  My main reference for this has been
       https://help.ubuntu.com/community/LiveCDCustomization, which doesn't
       mention this issue.
   debathena / livecd / andersk  17:05  (Anders Kaseorg)
       Hmm, is the official CD’s casper/initrd.gz different from the
       /boot/initrd.img-2.6.27-11-generic in the filesystem image?
   debathena / livecd / xavid  17:05  ("Never bite a gift horse in the neck.")
       I assume so.
   debathena / livecd / xavid  17:08  ('Sucks-to-be-you-tastic.`)
       (Based on this new information; before I'd assumed they were the same.)
   debathena / livecd / andersk  17:10  (Anders Kaseorg)
       Have you looked at
       https://help.ubuntu.com/community/LiveCDCustomizationFromScratch ?
   debathena / livecd / xavid  17:18  (“I like horse-molesting monkeys.”)
       Hmm; that's worth looking at.
   debathena / livecd / xavid  17:25  (“Where does the scary glowing man eat the
       Though, it looks like in those instructions it also uses the
       initramfs generated by installing a kernel, and has the same initrd in
       both the actual /boot and the outer /casper, so that doesn't address
       this.
   debathena / livecd / xavid  17:25  (“Welcome back cybergirls! and you—honorar
       What are the symptoms of compcache being enabled?  Are you certain that
       normal Ubuntus installed from a Live CD don't have it enabled?
   debathena / livecd / ezyang  17:26  (Edward Z. Yang)
       The easiest way to check is lsmod | grep compcache
   debathena / livecd / ezyang  17:26  (Edward Z. Yang)
       One important side-effect is that ramzswap is registered as swap
   debathena / livecd / ezyang  17:27  (Edward Z. Yang)
       I know fo sho that it isn't default, since we've had multiple confirm
       that, while they have the compcache kernel module, it was not active
   debathena / livecd / xavid  17:27  ('Can you guys tell me at wrapup if I’m th
       Hmm... well, I'll investigate more when I'm at a VMware and less
       swamped.
   debathena / livecd / ezyang  17:28  (Edward Z. Yang)
       Sounds good.
   debathena / livecd / xavid  17:28  (“There’s nothing that says we can’t have 
       I certainly don't do anything that would affect that directly.
   debathena / livecd / andersk  17:34  (Anders Kaseorg)
       compcache gets enabled if you boot with an initrd that was generated  
       on a system with the casper package installed.
   debathena / livecd / andersk  17:35  (Anders Kaseorg)
       > One important side-effect is that ramzswap is registered as swap
       (This is, in fact, entirely the point, not a side effect.)
   debathena / livecd / ezyang  17:35  (Edward Z. Yang)
       sure :-)
   debathena / livecd / xavid  17:42  ("Angry people are always cooler when they
       Ok, that should be easy enough to work around, then.

Reported by broder, 15 years ago.

erlang-base conflicts with erlang-doc-html (<< ${Upstream-Version}), even though erlang-doc-html is a completely separate source package. It's a cute trick, so long as the two packages stay in sync.

Unfortunately, erlang-base was manually synced into Jaunty after the DebianImportFreeze?, and nobody thought to grab erlang-doc-html as well.

This makes the two uninstallable, which means that thirdparty also won't currently install on Jaunty.

I see two options:

• SRU for erlang-base removing the conflict

• SRU for erlang-doc-html syncing in the new upstream version.

I don't know which is more likely to happen in bounded time. In the mean time, we should make erlang-doc-html a recommendation of thirdparty-languages instead of a dependency.