Pulseaudio update issue...
Hello, First off, not trying to bash or otherwise start a war about pulseaudio. Just checking if anyone else is experiencing issues with the latest F12 update (0.9.21-2) of pulseaudio? For me changing between tracks in rhythmbox include a sound 'pop' at the same time as the volume goes up/down. Also when doing tab completion in gnome-terminal the music playing nearly pauses for the tab 'ding' sound. I have reverted to the original version (0.9.19-2) and it seems to work, however the volume control applet can't start as it is looking for newer pulseaudio libs, even though I reverted it... I've looked through the bugs already on bugzilla but I'm not sure if its been posted yet or not... Anyone having issues like that? -- fedora-devel-list mailing list fedora-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-list
Re: Pulseaudio update issue... [ UPDATE 2 ]
So it seems it is related to thunderbird. I have the preference set to play a sound when new mail arrives. After it has, sounds is messed up... Bug with thunderbird I presume? -- fedora-devel-list mailing list fedora-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-list
Re: UPDATE Pulseaudio update issue...
Ok so totally bizarre, I re-updated via yum... It caused rhythmbox to freeze as the connection to the server died. I killed and restarted it, and the track changes were now sound seamless, as is tab completion in gnome-terminal again... I'm really not sure what the issue was, I've rebooted a few times today with noticing that an update had occurred, and then after the downgrade etc... So whatever it was is no longer an issue... odd. -- fedora-devel-list mailing list fedora-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-list
Re: packages requiring me to reboot...
So again today, I see some updates two of which require a full system reboot. nfs-utils and ibus-rawcode. My system seriously needs to be shut down for those to be properly updated? This is what I don't get. nfs-utils never got a system reboot before, it doesn't get one on RHEL/Centos boxes... What requires a reboot here? Again, I don't want the tone of this email to come off as anger, rude or whatever, mainly I'm wondering why so many packages require a reboot, why isn't nfs-utils just restarting any services it has or that depend on it if needed? Is that not reliable? -- fedora-devel-list mailing list fedora-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-list
Re: packages requiring me to reboot...
On 12/16/2009 09:51 AM, Seth Vidal wrote: On Wed, 16 Dec 2009, Peter Jones wrote: On 12/16/2009 11:43 AM, Seth Vidal wrote: you're an experienced user? You're comfortable knowing what does and what does not require a reboot? Then why are you using PK? Disable pk and do the updates directly via yum. Bam - no more requests to reboot. I get what you're saying, and it's kindof a fair point, but there's also some utility to having the system automatically, proactively notify you of updates. And you can do that. Just don't have pk DO the update. There are lots of ways to get notifications of updates not using PK in the system. And again, we're not talking about for the default everyday user. we're talking about the experienced user who is comfortable knowing what does and does not need a reboot. All I'm saying is - we've not taken away any option, the experienced user can do what they want. yeah, I totally get what you mean. I just feel like there are more and more reboot requests because that is easier. Obviously I know when/if I need to reboot based on what I'm running. However I'm questioning the number of packages requesting a reboot I guess. Maybe this is a feature that needs to be addressed in the rpm layer or something so that upgrades can have multiple effects with regards to needing a reboot. I'm not sure how PK gets the request to reboot from a package, but I'm wondering about it. For example, why aren't some of the packages simply a 'log out of X', or 'restart app', or ??? PK could provide that information. However, as it stands, if firefox is updated, I would (under the current way this seems to work) fully expect it to ask for my system to reboot, instead of closing FF and starting it again. I'm not sure if it does or not, but it seems like a package basically has complex upgrade issues, so we reboot. Are there other tags packages can have other than reboot? Should there be? etc etc.. I am an advanced user, and manage a handful of servers and workstations, so yes I don't have to reboot. I'm just wondering about the reboot 'feature' usage patterns I'm seeing. -- Nathanael -- fedora-devel-list mailing list fedora-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-list
Re: packages requiring me to reboot...
On 12/16/2009 10:11 AM, Seth Vidal wrote: On Wed, 16 Dec 2009, Nathanael D. Noblet wrote: Maybe this is a feature that needs to be addressed in the rpm layer or something so that upgrades can have multiple effects with regards to needing a reboot. I'm not sure how PK gets the request to reboot from a package, but I'm wondering about it. It doesn't get it from the pkg. It uses the updateinfo.xml metadata that is generated by our update processing system that is called 'bodhi'. You can see this data using the yum-security plugin. Cool. seems like a package basically has complex upgrade issues, so we reboot. Are there other tags packages can have other than reboot? Should there be? etc etc.. No. The reason for this is that PKs target audience is not someone like me, and as such no need to provide different messages per package? I am an advanced user, and manage a handful of servers and workstations, so yes I don't have to reboot. I'm just wondering about the reboot 'feature' usage patterns I'm seeing. And again. PK is not designed for you. The 'reboot often' solution is not FOR you. I said this earlier on another subject but you shouldn't be shocked that camels are slow swimmers. So basically, PK is designed for the non-experienced users, as such everything it does is dumbed down, and experienced users should just ignore it, using other tools to keep their system up to date. So one last question then, in the case of nfs-utils, (ignoring for now any nfs specific restart/condrestart issues). The packaging guidlines will continue to require that a post update script does what is sensible for an update, and not just depend on the admin rebooting their server? -- Nathanael -- fedora-devel-list mailing list fedora-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-list
Re: packages requiring me to reboot...
On 12/16/2009 10:28 AM, Seth Vidal wrote: On Wed, 16 Dec 2009, Nathanael D. Noblet wrote: seems like a package basically has complex upgrade issues, so we reboot. Are there other tags packages can have other than reboot? Should there be? etc etc.. No. The reason for this is that PKs target audience is not someone like me, and as such no need to provide different messages per package? No, the reason for this is there is not more to go on, yet. I would love to require more detailed info on the update including if it is an important/trivial/security/packaging/upstream-update or what not fix. Hands are needed to help advance this. Care to lend one? Yes. I'm attempting to become more involved. I've submitted my first package, and am going through the review process. That doesn't help in this particular case, but I am not complaining for the sake of complaining. I want to help. I fully realize what I use daily for work is the result of many people like you who build this stuff. Thus my desire to become part of it. What can I do here? the post scripts do what is sensible, on many occasions restarting the daemon will not ensure that the new sw is in use and in other occasions there is no graceful way to restart. so your options are: 1. don't restart but ask the user to 2. restart and drop whatever connections are active. neither are great. For sure. -- fedora-devel-list mailing list fedora-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-list
Re: packages requiring me to reboot...
On 12/16/2009 10:38 AM, Seth Vidal wrote: On Wed, 16 Dec 2009, Nathanael D. Noblet wrote: Hands are needed to help advance this. Care to lend one? Yes. I'm attempting to become more involved. I've submitted my first package, and am going through the review process. That doesn't help in this particular case, but I am not complaining for the sake of complaining. I want to help. I fully realize what I use daily for work is the result of many people like you who build this stuff. Thus my desire to become part of it. What can I do here? How much python do you know? We need some time spent on the updateinfo.xml and what information we provide there and tying this in with what info is required from packagers submitting updates to their pkgs. Unfortunately very very little. I can program in C/C++ and PHP. I've used python years ago. However I can learn if I have to. Another good angle to approach is to talk to the folks in fedora-qa and see where they can use a hand. Will do. -- fedora-devel-list mailing list fedora-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-list
packages requiring me to reboot...
Hello, I feel like there are an increasing number of packages requiring a system reboot. I'm wondering why. The following updates were installed today, and required a full system reboot. I can't seem to find any package in the list that I can conceivably see requiring a reboot, is it that PK doesn't have the concept of X logout vs reboot? Is it a bug in the packaging or PK or is there anything I can do/file to improve the situation? Dec 15 09:07:21 Updated: glib2-2.22.3-1.fc12.x86_64 Dec 15 09:07:23 Updated: mysql-libs-5.1.40-1.fc12.x86_64 Dec 15 09:07:23 Updated: gpm-libs-1.20.6-9.fc12.x86_64 Dec 15 09:07:25 Updated: mysql-5.1.40-1.fc12.x86_64 Dec 15 09:07:28 Updated: PyQt4-4.6.2-5.fc12.x86_64 Dec 15 09:07:32 Updated: mysql-server-5.1.40-1.fc12.x86_64 Dec 15 09:07:34 Updated: gpm-1.20.6-9.fc12.x86_64 Dec 15 09:07:37 Updated: 1:tk-8.5.7-3.fc12.x86_64 Dec 15 09:07:37 Updated: mpfr-2.4.1-5.fc12.x86_64 Dec 15 09:07:39 Updated: foomatic-4.0.3-5.fc12.x86_64 Dec 15 09:07:40 Updated: mysql-embedded-5.1.40-1.fc12.x86_64 Dec 15 09:07:41 Updated: glib2-2.22.3-1.fc12.i686 Dec 15 09:07:45 Updated: gtk2-2.18.4-3.fc12.x86_64 Dec 15 09:07:58 Updated: 1:gdm-2.28.1-25.fc12.x86_64 Dec 15 09:07:58 Updated: ibus-libs-1.2.0.20091204-2.fc12.x86_64 Dec 15 09:07:59 Updated: imsettings-libs-0.107.4-4.fc12.x86_64 Dec 15 09:08:02 Updated: glib2-devel-2.22.3-1.fc12.x86_64 Dec 15 09:08:07 Updated: yelp-2.28.1-1.fc12.x86_64 Dec 15 09:08:10 Updated: f-spot-0.6.1.5-1.fc12.x86_64 Dec 15 09:08:13 Updated: 1:xscreensaver-base-5.10-4.fc12.x86_64 Dec 15 09:08:13 Updated: 1:xscreensaver-gl-base-5.10-4.fc12.x86_64 Dec 15 09:08:16 Updated: gtk2-devel-2.18.4-3.fc12.x86_64 Dec 15 09:08:25 Updated: totem-2.28.4-1.fc12.x86_64 Dec 15 09:08:25 Updated: totem-nautilus-2.28.4-1.fc12.x86_64 Dec 15 09:08:27 Updated: 1:xscreensaver-gl-extras-5.10-4.fc12.x86_64 Dec 15 09:08:29 Updated: 1:xscreensaver-extras-5.10-4.fc12.x86_64 Dec 15 09:08:34 Updated: imsettings-0.107.4-4.fc12.x86_64 Dec 15 09:08:34 Updated: 1:gdm-user-switch-applet-2.28.1-25.fc12.x86_64 Dec 15 09:08:35 Updated: 1:gdm-plugin-fingerprint-2.28.1-25.fc12.x86_64 Dec 15 09:08:35 Updated: totem-mozplugin-2.28.4-1.fc12.x86_64 Dec 15 09:08:37 Updated: python-reportlab-2.3-1.fc12.x86_64 Dec 15 09:08:38 Updated: jna-3.2.4-1.fc12.x86_64 Dec 15 09:08:38 Updated: memcached-1.4.4-1.fc12.x86_64 Dec 15 09:08:38 Updated: less-436-3.fc12.x86_64 Dec 15 09:08:40 Updated: cscope-15.6-6.fc12.x86_64 Dec 15 09:08:40 Updated: xorg-x11-drv-mouse-1.5.0-1.fc12.x86_64 Dec 15 09:08:40 Updated: f-spot-screensaver-0.6.1.5-1.fc12.x86_64 Dec 15 09:08:46 Updated: gtk2-devel-docs-2.18.4-3.fc12.x86_64 Dec 15 09:08:46 Updated: gpm-devel-1.20.6-9.fc12.x86_64 Dec 15 09:08:46 Updated: liveusb-creator-3.9-1.fc12.noarch Dec 15 09:08:48 Updated: mysql-devel-5.1.40-1.fc12.x86_64 Dec 15 09:08:53 Updated: etoys-4.0.2339-1.fc12.noarch Dec 15 09:08:59 Updated: ibus-1.2.0.20091204-2.fc12.x86_64 Dec 15 09:08:59 Updated: ibus-gtk-1.2.0.20091204-2.fc12.x86_64 Wouldn't it be sufficient to logout? Is it a bug? -- Nathanael -- fedora-devel-list mailing list fedora-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-list
Re: packages requiring me to reboot...
On 12/15/2009 09:54 AM, Seth Vidal wrote: Does gdm entirely restart when you logout? I don't believe so. I suspect you get the same result by killing X then going back to that runlevel but for many many many users a reboot is going to be less error-prone. Isn't there gdm-restart for that purpose? I don't really know, but I'm just confused as to why a program that lets me login requires a reboot... I *really* don't want to sound whiny or anything like that, or be one of those that compare us to windows... but one of my favorite things from years ago was that I only had to reboot with a new kernel. Now I feel like I reboot every update. I mean, even the ibus stuff was stating I needed a reboot. As far as I know that is used for alternative language input, which I don't use, fair enough it doesn't know that. But what about it needs a reboot? I'm also curious why gdm is still running once I've logged in. I see the user-switch stuff but I'm just wondering. I mean rebooting isn't the end of the world but man it sure happens a lot now -- fedora-devel-list mailing list fedora-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-list
Re: kernel update highly recommended
On 12/09/2009 11:14 AM, Kyle McMartin wrote: Hi folks, I'd highly recommend if you're running 2.6.31 or 2.6.32, that you update to the latest kernel in the koji builds here: http://koji.fedoraproject.org/koji/taskinfo?taskID=1864871 http://koji.fedoraproject.org/koji/taskinfo?taskID=1864876 They fix a rather severe security problem with ext4 caused by insufficient permission checking by the ext4 ioctl code, allowing a malicious local user to corrupt files. Note, the ioctl isn't currently used by userspace, so if you build your own kernels, you can just nuke the entire EXT4_IOC_MOVE_EXT ioctl case. NOTE: This is only a problem if you're using EXT4, if you aren't, you're safe. I'll get these pushed out to stable asap, but I wanted to let folks know just in case rawhide doesn't compose before the downtime. This a rawhide only issue or F12 as well? -- fedora-devel-list mailing list fedora-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-list
Re: Review request...
On 11/18/2009 06:23 PM, Kevin Kofler wrote: Michael Schwendt wrote: Many packagers don't know that maintaining a proper spec %changelog for relevant spec file changes and %release bumps are considered important during review already. Others add meaningless/dummy %changelog entries even in Fedora cvs. But that's a people issue that needs to be solved, not papered over in the name of being nice. Sadly, incompetence as in unfamiliarity with guidelines which are assumed to be prerequisites for proper packaging is growing in our ranks (both from new packagers and from people who ought to know better), something needs to be done about it. For sure. Personally I've been using Redhat/Fedora for years now, but this is my first package submission to fedora. I've wanted to get involved for awhile now. I had read the guidelines, and honestly want to provide a properly configured package. I think (and this very well could have been a language issue) that knowing what the issue I'm missing in my current spec would help immensely. I mean I know there are packaging guidelines, and there is a lot of information there, so it is plausible for someone new not to see sub documentation or notice that their spec isn't in compliance. Having the exact issue pointed out helps with the learning. Is there a 'ReviewingReviews' guideline? Would that even help? Anyway, I hope to get some feedback on the actual review too, but I mainly started this thread because I wasn't sure what I wasn't doing. I was asked for a scratch build which I *thought* I had provided a link to. I provided links to spec files and srpms. However was continually being asked for that same thing, and the requests and my responses were obviously not being understood by either party. I posted to make sure I wasn't missing something obvious, some guideline of 'Here's how to post your spec file, srpm and scratch build', if I wasn't doing it correctly. Sincerely trying to provide the best package I can, Nathanael -- fedora-devel-list mailing list fedora-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-list
Re: RFC: Btrfs snapshots feature for F13
On 11/17/2009 12:05 PM, Josef Bacik wrote: Sure, this isn't a perfect solution, it's just a nice to have feature if you care for it. It's nice to take a complete snapshot of your system right before you update just in case something goes horribly wrong and you lose say configuration files or some such. If you modified other things and have to rollback, you can always just mount the newer snapshot when you boot into the old snapshot and copy the new data that you want back. This isn't for the faint of heart, I envisioned it really for people who want to play the rawhide game with less exposure to its instability. Thanks, I've long wanted to help with fedora rawhide a bit more but only really have it on the one computer I work on. So this would make it possible for me to do that. I think that's great. I have a few questions. Suppose I do an update, and it breaks X or some other important piece for me. I reboot into my previous snapshot. From there I continue to work. Do I have to remove the 'future' snapshot I came back from to continue working in that snapshot? Or is that just best practice. If I move forward, I'd have to move all the work I did in that snapshot to the head/snapshot that just got fixed with an update... Just thinking out loud here. -- fedora-devel-list mailing list fedora-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-list
Re: rpmlint warnings...
On 11/15/2009 02:30 PM, Dominik 'Rathann' Mierzejewski wrote:
On Sunday, 15 November 2009 at 21:59, Nathanael Noblet wrote:
Hello,
So I recently posted my first package and the review. While I waited I
started cleaning up more issues I found after I realized you could run rpmlint
on the actual rpm and not just the spec file. I'd like the review to go as
quickly as possible so I'm just trying to get all those warnings cleaned up.
My package has a number of sub packages for various backend drivers. These
subpackages basically contain a .so file for the most part however I'm getting
rpmlint messages as follows
libdspam.x86_64: W: devel-file-in-non-devel-package /usr/lib64/libdspam.so
how is libdspam.so determined to be a devel file?
Shared objects (libraries) residing in %{_libdir} usually have names like
libfoo.so.X.Y.Z where X.Y.Z is their ABI version number. -devel subpackages
contain libfoo.so which is usually a link to libfoo.X.Y.Z and is used for
linking against libfoo (-lfoo in linker command line).
libdspam.so.7.0.0: ELF 64-bit LSB shared object, x86-64, version 1 (SYSV),
dynamically linked, stripped
Is what I get back from file. What is it that I'm missing?
Judging by the above, your libdspam.so should in fact be named
libdspam.so.7.0.0.
[g...@iridium ~]$ ls -l /usr/lib64/libdspam.*
-rw-r--r--. 1 root root 175812 2009-11-15 13:54 /usr/lib64/libdspam.a
-rwxr-xr-x. 1 root root954 2009-11-15 13:54 /usr/lib64/libdspam.la
lrwxrwxrwx. 1 root root 17 2009-11-15 13:59 /usr/lib64/libdspam.so
- libdspam.so.7.0.0
lrwxrwxrwx. 1 root root 17 2009-11-15 13:59 /usr/lib64/libdspam.so.7
- libdspam.so.7.0.0
-rwxr-xr-x. 1 root root 111000 2009-11-15 13:54 /usr/lib64/libdspam.so.7.0.0
[g...@iridium ~]$ ldd /usr/bin/dspam_2sql
linux-vdso.so.1 = (0x7fffccfda000)
== libdspam.so.7 = /usr/lib64/libdspam.so.7 (0x7f7f4d89e000)
libm.so.6 = /lib64/libm.so.6 (0x00335a40)
libdl.so.2 = /lib64/libdl.so.2 (0x00335a80)
libldap-2.4.so.2 = /usr/lib64/libldap-2.4.so.2 (0x7f7f4d659000)
liblber-2.4.so.2 = /usr/lib64/liblber-2.4.so.2 (0x00336200)
libpthread.so.0 = /lib64/libpthread.so.0 (0x00335ac0)
libc.so.6 = /lib64/libc.so.6 (0x00335a00)
/lib64/ld-linux-x86-64.so.2 (0x003359c0)
libresolv.so.2 = /lib64/libresolv.so.2 (0x00335d00)
libsasl2.so.2 = /usr/lib64/libsasl2.so.2 (0x7f7f4d43d000)
libssl.so.10 = /usr/lib64/libssl.so.10 (0x00336800)
libcrypto.so.10 = /usr/lib64/libcrypto.so.10 (0x00336600)
libcrypt.so.1 = /lib64/libcrypt.so.1 (0x7f7f4d205000)
libgssapi_krb5.so.2 = /lib64/libgssapi_krb5.so.2 (0x00336700)
libkrb5.so.3 = /lib64/libkrb5.so.3 (0x003366c0)
libcom_err.so.2 = /lib64/libcom_err.so.2 (0x003365c0)
libk5crypto.so.3 = /lib64/libk5crypto.so.3 (0x00336740)
libz.so.1 = /lib64/libz.so.1 (0x00335b00)
libfreebl3.so = /usr/lib64/libfreebl3.so (0x7f7f4cfa4000)
libkrb5support.so.0 = /lib64/libkrb5support.so.0 (0x00336680)
libkeyutils.so.1 = /lib64/libkeyutils.so.1 (0x00336780)
libselinux.so.1 = /lib64/libselinux.so.1 (0x00335bc0)
It seems to me then that libdspam.so.7.0.0 is the actual file, and I
have libdspam.so and libdspam.so.7 as symlinks. Based off the ldd of
some of the binaries I can see that it is linked to x.so.VER for most
libraries...
So does that mean that my libdspam.so.7.0.0 and libdspam.so.7 are in the
one package and then libdspam.a/la/so are part of -devel ?
Would that be the correct assumption?
Thanks for the tips so far.
--
Nathanael D. Noblet
--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list
Re: rpmlint warnings...
On 11/15/2009 06:52 PM, Matt Domsch wrote:
On Sun, Nov 15, 2009 at 01:59:57PM -0700, Nathanael Noblet wrote:
Hello,
So I recently posted my first package and the review. While I waited I
started cleaning up more issues I found after I realized you could run rpmlint
on the actual rpm and not just the spec file. I'd like the review to go as
quickly as possible so I'm just trying to get all those warnings cleaned up.
My package has a number of sub packages for various backend drivers. These
subpackages basically contain a .so file for the most part however I'm getting
rpmlint messages as follows
libdspam.x86_64: W: devel-file-in-non-devel-package /usr/lib64/libdspam.so
You should also see if you can build the app such that it installs its
private plugins in a subdir of %{_libdir} rather than
clutter up %{_libdir} with libraries that nothing else can use.
Yeah, there is a lib, and then dlopened backend drivers. So we have.
%{_libdir}/libdspam.so.7 (libdspam package)
%{_libdir}/dspam/libdspam-mysql_drv.so (libdspam-mysql)
For each backend driver.
--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list
Re: rpmlint warnings...
On 11/15/2009 03:05 PM, Michael Schwendt wrote:
On Sun, 15 Nov 2009 22:30:44 +0100, Dominik wrote:
On Sunday, 15 November 2009 at 21:59, Nathanael Noblet wrote:
Hello,
So I recently posted my first package and the review. While I waited I
started cleaning up more issues I found after I realized you could run rpmlint
on the actual rpm and not just the spec file. I'd like the review to go as
quickly as possible so I'm just trying to get all those warnings cleaned up.
My package has a number of sub packages for various backend drivers. These
subpackages basically contain a .so file for the most part however I'm getting
rpmlint messages as follows
libdspam.x86_64: W: devel-file-in-non-devel-package /usr/lib64/libdspam.so
how is libdspam.so determined to be a devel file?
Shared objects (libraries) residing in %{_libdir} usually have names like
libfoo.so.X.Y.Z where X.Y.Z is their ABI version number. -devel subpackages
contain libfoo.so which is usually a link to libfoo.X.Y.Z and is used for
linking against libfoo (-lfoo in linker command line).
libdspam.so.7.0.0: ELF 64-bit LSB shared object, x86-64, version 1 (SYSV),
dynamically linked, stripped
Is what I get back from file. What is it that I'm missing?
Judging by the above, your libdspam.so should in fact be named
libdspam.so.7.0.0.
It is. file prints the file name, not the SONAME.
Often, projects, which dlopen plugin/module shared libraries at run-time,
store the versioned libraries as well as the .so symlinks. Then they dlopen
the libraries via the .so symlinks.
Find out how and with what file names those backend libraries are loaded.
If the .so symlinks are not needed, don't package them. And if the
backend libraries have SONAMEs defined, check for [potential] conflicts with
system libraries.
So I've grepped through the source a bit and the library loads a storage
driver from the config file. So in the case of mysql the dspam.conf file has
StorageDriver /usr/lib64/dspam/libhash_drv.so
and passes that full path to dlopen. The sub driver package has the
following files.
/usr/lib64/dspam/libmysql_drv.so
/usr/lib64/dspam/libmysql_drv.so.7
/usr/lib64/dspam/libmysql_drv.so.7.0.0
It will open the .so which is a symlink to the real file of so.7.0.0. I
don't think it would be feasible to have the config file load the
so.7.0.0 because any update to to the SONAME would cause a config change
when it may not be necessary.
So in this case do I package them all? So does that mean the .so is okay
to be in the non devel package?
--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list
Review Request...
Hello, I just submitted my first package for review. I'm not sure how to mark it as need sponsor. I've created my FAS account and signed the cla. I've started a scratch build to see how it works on the other arches. I've compiled locally on f12 for x86_64. Do I need to do anything else to mark it as need sponsor? Otherwise I assume I just wait for feedback via the bugzilla entry correct? Thanks, -- Nathanael Noblet -- fedora-devel-list mailing list fedora-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-list
abrt / kernel oops issue
Hello, I've been running F12/rawhide from a preupgrade from F11 for a couple weeks now. I've just recently noticed the abrt feature. I started submitting the bugs it found in the kerneloops. Which has me wondering couple things. #1 - I have many many kerneloops, each stacktrace/log snippet seems fairly identical. I assume you would prefer not to get 20 identical abrt submitted bugs? Or should I submit them all? #2 - When looking at the trace it has the following line(s): Nov 10 09:15:57 iridium kernel: WARNING: at lib/list_debug.c:30 __list_add+0x68/0x81() (Tainted: GW ) ... Nov 10 09:15:57 iridium kernel: Pid: 2197, comm: Xorg Tainted: G W 2.6.31.5-127.fc12.x86_64 #1 I'm wondering why it is saying it is tainted.. or maybe Tainted: G mean Good?? I don't have any closed source modules loaded as far as I know. Unless Virtualbox is closed source but I didn't think it was. Virtualbox isn't running when this happens, though the module seems to be loaded. #3 Validity of the bugs it is finding... It calls the following a kerneloops... Nov 11 16:56:41 iridium gnome-session[2259]: WARNING: Could not parse desktop file /etc/xdg/autostart/network-manager-netbook.desktop: Key file does not have key 'Name' Nov 11 16:56:41 iridium gnome-session[2259]: WARNING: could not read /etc/xdg/autostart/network-manager-netbook.desktop Nov 11 16:56:42 iridium kernel: executing set pll Nov 11 16:56:42 iridium kernel: executing set crtc timing Nov 11 16:56:42 iridium kernel: [drm] TMDS-15: set mode 25 Nov 11 16:56:42 iridium kernel: executing set pll Nov 11 16:56:42 iridium kernel: executing set crtc timing Nov 11 16:56:42 iridium kernel: [drm] TMDS-11: set mode 2f Nov 11 16:56:42 iridium kernel: fuse init (API version 7.12) Nov 11 16:56:43 iridium pulseaudio[2406]: pid.c: Daemon already running. Nov 11 16:56:46 iridium restorecond: Unable to watch (/home//public_html/*) No such file or directory do I submit this anyway? Thanks, just rying to do my part without burying you guys in senseless bug reports. -- fedora-devel-list mailing list fedora-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-list
Re: abrt / kernel oops issue
On 11/12/2009 10:50 AM, Dan Williams wrote: If you have an oops or BUG of any sort, I think that sets the taint flag for further oops reports, because after the first one you can't really trust that the stacktrace or internal kernel structures aren't corrupted. Most of the time they aren't, but you simply can't trust that. So I'd expect the first one to be untainted, and then subsequent oops reports to have the taint flag set. I see.. I'll have to reboot and see if the first one is not being marked as tainted. Of course if you start loading random kernel modules that didn't come with the kernel itself, you can also taint the kernel. If you have staging drivers loaded, you'll have the taint_crap flag set because staging drivers are crap. I'm not manually loading anything, and the only module I have loaded that isn't part of the kernel rpm package itself is Virtualbox. Which I can remove or not load to see if it makes a difference. I have to also note that I didn't even know I was having kernel oops until abrt popped up. -- fedora-devel-list mailing list fedora-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-list
Evolution Data Server...
Hello, So this isn't a strictly development question, but based on the answer it very well could be. I don't use evolution, but the evolution-data-server is running. Is it used for anything else? If not, perhaps it would be good to not run it as part of the gnome session when the users default mail client isn't evolution. If it is used for other purposes then whatever. Otherwise I can file a bug report if desired... -- Nathanael noblet -- fedora-devel-list mailing list fedora-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-list
Re: Evolution Data Server...
On 11/02/2009 09:52 AM, Adam Williamson wrote: On Mon, 2009-11-02 at 09:36 -0700, Nathanael D. Noblet wrote: Hello, So this isn't a strictly development question, but based on the answer it very well could be. I don't use evolution, but the evolution-data-server is running. Is it used for anything else? If not, perhaps it would be good to not run it as part of the gnome session when the users default mail client isn't evolution. If it is used for other purposes then whatever. Otherwise I can file a bug report if desired... Yes, several other things use it. It's something of an unfortunate name; e-d-s is really a generic PIM information server. It's a sensible model: it lets multiple applications access and modify the information in question while they are all active. KDE, which did not used to use this model, had a problem where if anything other than KMail wanted to use contact data - say you wanted to synchronize it with another device via OpenSync - you had to close KMail first, or messiness could ensue (the sync would fail, or in a bad case KMail could fall over; I think in a really really bad case you could even lose or duplicate data). KDE is switching to the model of having a server for this information with Akonadi. GNOME's server for this information is e-d-s. The most common non-Evolution user of e-d-s data is the clock applet on the panel; it notifies you of impending appointments, and it does this by looking them up via e-d-s. But there are several others too. Good to know! ;) Thanks for the info. -- fedora-devel-list mailing list fedora-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-list
Re: Possible packages...
On 09/17/2009 10:22 AM, Adam Williamson wrote: On Thu, 2009-09-17 at 09:52 +0200, Rudolf Kastl wrote: PS3MediaServer. A Java program to talk to a PS3 with DLNA. I'm guessing this one would have problems because it requires ffmpeg or mplayer/mencoder... Plus as a java program its probably a bit more complex to create a proper spec file for. I've made the other kind often enough, but java ones not so much... There's a sort of 'agreed-upon-right-way-of-doing-this' candidate for this particular need, which is a nice modern GTK+ app and based on gstreamer...but I can't quite pull the name out of long-term storage at present. Someone will probably know what I mean, though. The one most people use (as the one I'm talking about is still a bit alpha) is mediatomb, which is also in Fedora already. Unless this provides something significant the other options don't, it may not be the best place to start, since it looks a bit complex. ps3mediaservers biggest improvement/enhancement is the ability to transcode video files on the fly. Since I wrote the message quoted by Rudolf, I remembered the name of the app I was trying to think of: Rygel - http://live.gnome.org/Rygel aside from that, the 'market leader' is mediatomb, which I think we have in Fedora or RPM Fusion already. It has been able to do transcoding for a long time, and there's a big knowledge base out there on how to use it. I'm not entirely sure adding another packaged ps3-intended-UPNP-server would be a net win anywhere. I had mediatomb installed, very much disliked it. It may have been my ability to configure it as well. However it wasn't as easy as ps3mediaserver. Granted I don't know if the ps3 one will work with all media players, I think it only encodes to what the ps3 can handle if I'm not mistaken. (unless ps3mediaserver's implementation of on-the-fly transcoding lets you fast-forward and rewind. that'd be good. mediatomb can't do that.) It does. -- fedora-devel-list mailing list fedora-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-list
Re: Possible packages...
On 09/16/2009 08:20 AM, Nathanael Noblet wrote: I didn't because it still had quite a few patches that needed to go upstream. I'll take a look at version 2.3... Yeah the lib-patches is still full of patches... -- fedora-devel-list mailing list fedora-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-list
Possible package...
Hello everyone, I've been looking at becoming a packager. I've yet to find software I need that isn't in fedora yet. I just found one I think. http://code.google.com/p/wkhtmltopdf/ It is based off of qt4 and webkit. Small source files, cmake build system. I can't find it in the repos anywhere. Would this be a good first package? I'll be using this for a couple of projects to convert html to pdf with much less hassle than ever before. Thoughts? -- Nathanael d. noblet -- fedora-devel-list mailing list fedora-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-list
Potential PackageKit improvement???
Hi, before I file anything about this, just wanted to ask. Today I received a KDE update, I don't run the kde desktop. It wants to reboot my whole computer... Perhaps it could be smarter to know whether it needs a reboot, a logout / login, particularly if I'm running gnome vs KDE... Is this a legitimate request? Should I file a bug / feature request? -- Nathanael d. Noblet -- fedora-devel-list mailing list fedora-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-list
Re: Package Kit messages...
On 07/31/2009 09:50 AM, Michael Cronenworth wrote: Richard Hughes on 07/31/2009 10:43 AM wrote: Not really. If you're running an old version of gimp, you can restart [snip] Fedora 10 and 11 support only 2,3 Unfortunately there's a bug somewhere then. I've been meaning to file another bug report on this. The PackageKit applet tooltip will state You must logout and you click on Logout and it presents you with a Shutdown/Restart Gnome dialog instead of the Logout dialog. Which is what I was trying to communicate... Should I file a bug then? -- fedora-devel-list mailing list fedora-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-list
Re: Package Kit messages...
On 07/31/2009 04:28 PM, Michael Cronenworth wrote: On 07/31/2009 05:27 PM, Nathanael D. Noblet wrote: Which is what I was trying to communicate... Should I file a bug then? Yes, please. CC me, too, or link me. And this is specifically PackageKit, and not some break out from it like an applet or something?... -- fedora-devel-list mailing list fedora-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-list
Package Kit messages...
Hello, So I recently updated F11 and was told I needed to log off for the changes to take effect. When I click the yield type sign and select log off, I get the dialog for shutdown, restart, hibernate, suspend... Just a small suggestion, maybe I'm off base, but 'Log off' to me is like switch user, its a logout from X. It may be more useful to have the message state 'Restart' than log out... Am I off base? If I'm not, what component would I file against, simply PackageKit ? -- Nathanael d. Noblet -- fedora-devel-list mailing list fedora-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-list
Re: Possible packages...
On 07/09/2009 02:31 PM, Jarod Wilson wrote: On Monday 06 July 2009 10:49:31 Eric Sandeen wrote: Nathanael Noblet wrote: On Jul 5, 2009, at 9:33 PM, Eric Sandeen wrote: .. Well their python run script checks for its dependancies, and if not met will do a svn checkout of the right copy, however, they don't keep copies of the libraries within their own repository. So if you fulfill all its dependancies that shouldn't be an issue. Ah, ok - maybe that was it. Currently, it looks like it still requires its own builds of a few things. Stuff (apparently) not in Fedora: -pydirector Yeah I don't think this is in fedora -PyKerberos (might just be named something slightly different) I thought this was I could be wrong though. Stuff in Fedora, but simply not used for whatever reason: -vobject (we have python-vobject) -pyflakes I thought they were used if found... I'd have to look at the run file to see if they ignore the system versions.. Stuff in Fedora, but still heavily patched for CalendarServer: -Twisted (the web2 portion, specifically) -xattr (requires Bob Ippolito's implementation) I've got it using the fedora version of xattr I think, and didn't notice the patches to Twisted... That said, I have it up and running on an F11 host at home right now, satisfying everything else w/Fedora packages. Yeah same here. -- fedora-devel-list mailing list fedora-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-list
Re: Mass-Package Orphanage
On 07/03/2009 09:23 AM, Jason L Tibbitts III wrote: NJ == Nigel Jonesd...@nigelj.com writes: NJ Horde: All of these packages (horde, imp, ingo, jeta, kronolith, turba) are PHP-based webapps, which should strike fear into most maintainers. I happen to be co-maintainer so I guess I've been promoted, although honestly I haven't had time to put much effort into these packages either. So, I'm pretty sure I can't be the only person in Fedora who cares about these packages. They need at least two maintainers, preferably three. Please feel free to pile on. I've recently decided to try to help out by being a package maintainer... I haven't submitted a package, I don't have any new packages I need. I'm wondering if it is possible to be a co-maintainer with someone willing? I've got lots of experience creating packages for our own server farm and workstations, but none with fedora approved packaging... What's the policy here? I wouldn't mind helping out with the horde stack, but am guessing I would need some sort of sponsor, double check commits before they happen type thing... ? -- fedora-devel-list mailing list fedora-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-list
Re: Mass-Package Orphanage
On 07/03/2009 10:52 AM, Jason L Tibbitts III wrote: NDN == Nathanael D Nobletnathan...@gnat.ca writes: NDN I'm wondering if it is possible to be a co-maintainer with NDN someone willing? In general, all you need is a sponsor. The usual route to that is via the submission of new packages, but it's not the only way. I happen to be a sponsor, so that's not an issue. What is an issue is that fact that the horde suite is somewhat delicate and security sensitive, and not really the best set of packages for a first-time packager. This is somewhat offset by the fact that the packages already exist and just need maintenance. Yeah, I figured as much. There are a few other packages I wouldn't mind helping out with all in all but without submitting a package for review it seems I'm coming at this a bit sideways... I think that if you're interested, you should start by requesting watchcommits and watchbugzilla on at least the Fedora branches of those packages. (I have no desire to maintain the EPEL5 branch, so someone else needs to step in to take care of that one.) The URLs are: Ok, well I could definitely use the EPEL5 packages as I up to now have been installing them on CentOS 5 servers manually. https://admin.fedoraproject.org/pkgdb/packages/name/horde https://admin.fedoraproject.org/pkgdb/packages/name/imp https://admin.fedoraproject.org/pkgdb/packages/name/ingo https://admin.fedoraproject.org/pkgdb/packages/name/jeta https://admin.fedoraproject.org/pkgdb/packages/name/kronolith https://admin.fedoraproject.org/pkgdb/packages/name/turba Later we can progress to sponsorship and commit access. I still hope we can find at least one other person to assist in maintaining these packages. Otherwise I fear I will just end up orphaning them again. Sounds good.. -- fedora-devel-list mailing list fedora-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-list
Pulseaudio question...
At the risk of bringing up a touchy subject, I have a couple pulseaudio questions. Just wondering if the behaviour I am experiencing is a bug or intended behaviour. Fedora 11 recently installed fully up to date as of now. I've opened a movie in totem, the main volume level is somewhere in the range of 70%, however totem is at about 4%. The movie is too quiet so I've upped the volume in totem to 19%. The main volume is now 84%. All fine and good. However if I modify the main volume at all (up or down), the main volume resets to whatever the main volume was before I changed totem's volume. If I in this case bring the main volume up to 84% again, totem is at the 19% I set it to... Is this expected behaviour? -- Nathanael d. Noblet T: 403.875.4613 -- fedora-devel-list mailing list fedora-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-list
Re: Upgrade to F11, now yum python module missing
On 06/10/2009 11:55 AM, Seth Vidal wrote: 1. export PYTHONPATH=/usr/lib/python2.5/site-packages 2. yum update yum I had the same issue, and I can confirm that this does work, if done as root. Obviously if you do line 1, then sudo yum update yum, the env is different, or at least it must be because it failed to work that way for me, so I had to su first. -- Nathanael d. Noblet T: 403.875.4613 -- fedora-devel-list mailing list fedora-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-list
Bittorrent speeds...
Hello, Is it just me or is the torrent a bit slow. I typically download TV shows at 1.5MB/s (MB, not Mb). 150-300MB in 4-5 minutes. This torrent is sitting around 20-40KB/s... Anything I can do about that? Is the tracker swamped or ??? -- Nathanael d. Noblet T: 403.875.4613 -- fedora-devel-list mailing list fedora-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-list
Re: Bittorrent speeds...
Conrad Meyer wrote: I was getting speeds in excess of a few megabytes a second on the Fedora 11 torrents, I'm not sure why you're having trouble. Yeah, I'm slightly confused, I downloaded a 5.5GB TV Series alongside it at 1.5MB. It never increased after that completed... -- Nathanael d. Noblet T: 403.875.4613 -- fedora-devel-list mailing list fedora-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-list
Re: Bittorrent speeds...
Matt Domsch wrote: On Tue, Jun 09, 2009 at 09:45:27PM -0600, Nathanael D. Noblet wrote: Hello, Is it just me or is the torrent a bit slow. I typically download TV shows at 1.5MB/s (MB, not Mb). 150-300MB in 4-5 minutes. This torrent is sitting around 20-40KB/s... Anything I can do about that? Is the tracker swamped or ??? Which torrents in particular? I just got the x86_64 LiveCD at about 1.3MB/sec; ~475 seeders of that. X86_64 DVD Install. I dropped the max connections and have slowly bumped it up from there. Its at 350 now, so I'm now getting 500K/s. Not really sure what was going on, when I looked at the peer/seeder list, there were lots and lots of connections, but 0K up/down. I figured if I dropped the number of connections maybe deluge would perhaps keep the ones sending stuff a bit better... Not really sure what's going on but its much better than the 8-20K I was getting. -- Nathanael d. Noblet T: 403.875.4613 -- fedora-devel-list mailing list fedora-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/fedora-devel-list
Re: Proposal (and yes, I'm willing to do stuff!): Must Use More Macros
Ray Strode wrote:
Hi,
On Fri, Jun 5, 2009 at 1:31 PM, Adam Williamsonawill...@redhat.com wrote:
I've been dipping my toes into packaging things for Fedora lately, and
one thing that feels a bit awkward is that the packaging guidelines are
full of boilerplate like:
[...]
Heck, there's an entire page full of these:
https://fedoraproject.org/wiki/Packaging/ScriptletSnippets
[...]
It seems to me it'd make sense to convert all these kinds of snippets
into macros. Am I right, or is there a reason against doing this?
It would be awesome to get rid of the boilerplate. Honestly though,
I'd rather the solution was just works than replace giant glob of
muck with %{glob_of_muck}
For instance, if a file gets dropped under /usr/share/icons/something
rpm should run gtk-update-icon-cache /usr/share/icons/something
automatically.
the gtk2 package should be able to drop a file in /usr/lib/rpm/redhat
that makes that happen. likewise, desktop-file-utils should be able
to drop a file there to make update-desktop-database get run and so
on.
I don't know how hard it would be to fix rpm to allow for that though.
Just off the top of my head, this doesn't feel like something rpm should
be in charge of. To me it seems more likely that we need something in a
base/core rpm that installs an inotify script for system dirs that does
what it should when something is dropped into it...?
--
Nathanael d. Noblet
T: 403.875.4613
--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list
