Gentoo Websites Logo
Go to: Gentoo Home Documentation Forums Lists Bugs Planet Store Wiki Get Gentoo!
Bug 124659 - [stable req] gossip-0.9 loudmouth-1.0.1
Summary: [stable req] gossip-0.9 loudmouth-1.0.1
Status: RESOLVED WONTFIX
Alias: None
Product: Gentoo Linux
Classification: Unclassified
Component: New packages (show other bugs)
Hardware: All Linux
: High normal (vote)
Assignee: Gentoo Linux Gnome Desktop Team
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2006-03-02 09:10 UTC by foser (RETIRED)
Modified: 2007-01-27 10:49 UTC (History)
3 users (show)

See Also:
Package list:
Runtime testing required: ---


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description foser (RETIRED) gentoo-dev 2006-03-02 09:10:31 UTC
please stabilize gossip-0.9  & loudmouth-1.0.1
Comment 1 John N. Laliberte (RETIRED) gentoo-dev 2006-03-23 18:33:04 UTC
@foser: who should be stabling this? :)
Comment 2 foser (RETIRED) gentoo-dev 2006-03-24 03:44:16 UTC
he who reads this
Comment 3 foser (RETIRED) gentoo-dev 2006-04-09 10:03:53 UTC
what is going on here? third time I'm adding arches.
Comment 4 Olivier Crete (RETIRED) gentoo-dev 2006-04-09 11:28:31 UTC
amd64 stable..
Comment 5 nixnut (RETIRED) gentoo-dev 2006-04-09 12:26:31 UTC
Stable on ppc
Comment 6 Jason Wever (RETIRED) gentoo-dev 2006-04-12 19:46:53 UTC
SPARC'd
Comment 7 Markus Rothe (RETIRED) gentoo-dev 2006-04-17 11:01:17 UTC
loudmouth-1.0.1: stable on ppc64
gossip-0.9 (and higher versions): added ~ppc64

keeping us CC'd until gossip stable. (30 days)
Comment 8 Matthias Langer 2006-04-23 18:37:32 UTC
i've done some testing (thanks to David Morgan for beeing a very helpfull test chat partner) with loudmouth-1.0.1 [-debug -doc +ssl] and gossip-0.9 on x86. while the program basically works, there is an annoying issue not present in gossip-0.8 with loudmouth-0.17.2-r2. if another user adds you to his/her contact list, gossip-0.9 doesn't ask you for confirmation. i'm therfore not sure if these packages should really be stabled on x86.

Portage 2203-svn (default-linux/x86/2006.0, gcc-3.4.5, glibc-2.3.5-r3, 2.6.16-gentoo-r3 i686)
=================================================================
System uname: 2.6.16-gentoo-r3 i686 AMD Athlon(tm) XP 2400+
Gentoo Base System version 1.6.14
dev-lang/python:     2.3.5-r2, 2.4.2
sys-apps/sandbox:    1.2.12
sys-devel/autoconf:  2.13, 2.59-r7
sys-devel/automake:  1.4_p6, 1.5, 1.6.3, 1.7.9-r1, 1.8.5-r3, 1.9.6-r1
sys-devel/binutils:  2.16.1
sys-devel/libtool:   1.5.22
virtual/os-headers:  2.6.11-r2
ACCEPT_KEYWORDS="x86"
AUTOCLEAN="yes"
CBUILD="i686-pc-linux-gnu"
CFLAGS="-O2 -march=athlon-xp -pipe"
CHOST="i686-pc-linux-gnu"
CONFIG_PROTECT="/etc /usr/kde/2/share/config /usr/kde/3.4/env /usr/kde/3.4/share/config /usr/kde/3.4/shutdown /usr/kde/3/share/config /usr/lib/X11/xkb /usr/share/config /usr/share/texmf/dvipdfm/config/ /usr/share/texmf/dvips/config/ /usr/share/texmf/tex/generic/config/ /usr/share/texmf/tex/platex/config/ /usr/share/texmf/xdvi/ /var/qmail/control"
CONFIG_PROTECT_MASK="/etc/eselect/compiler /etc/gconf /etc/terminfo /etc/env.d"
CXXFLAGS="-O2 -march=athlon-xp -pipe"
DISTDIR="/usr/portage/distfiles"
FEATURES="autoconfig collision-protect distlocks sandbox sfperms strict"
GENTOO_MIRRORS="http://gentoo.inode.at/ "
LANG="en_US.utf8"
LC_ALL="en_US.utf8"
LINGUAS="en de"
MAKEOPTS="-j2"
PKGDIR="/usr/portage/packages"
PORTAGE_TMPDIR="/var/tmp"
PORTDIR="/usr/portage"
PORTDIR_OVERLAY="/usr/local/portage"
SYNC="rsync://192.168.0.1/gentoo-portage"
USE="x86 3dnow 3dnowext X a52 aalib alsa apm audiofile avi berkdb bitmap-fonts bonobo bzip2 bzlib cairo cdr cli crypt css cups curl dbus divx4linux dri dts dv dvd dvdr dvdread emboss encode evo exif expat fam fame ffmpeg firefox flac foomaticdb fortran gd gdbm gif glut gmp gnome gphoto2 gpm gstreamer gtk gtk2 gtkhtml guile hal idn imagemagick imlib ipv6 isdnlog java jpeg junit lcms libg++ libwww mad mikmod mmx mmxext mng motif mp3 mpeg nautilus ncurses nls nptl nsplugin nvidia ogg oggvorbis openal opengl pam pcre pdflib perl plotutils png pppd python quicktime readline real reflection ruby sdl session slang speex spell spl sqlite sse ssl subtitles svga tcltk tcpd tetex theora tiff truetype truetype-fonts type1-fonts udev unicode usb vcd video_cards_nvidia vorbis win32codecs wma xine xml xml2 xmms xorg xv xvid zlib linguas_en linguas_de userland_GNU kernel_linux elibc_glibc"
Unset:  ASFLAGS, CTARGET, INSTALL_MASK, LDFLAGS
Comment 9 Mark Loeser (RETIRED) gentoo-dev 2006-05-03 14:51:57 UTC
Gnome: can you advise us on how to handle this?  In testing the newer versions, they seem somewhat broken.  The authorization request never gets to the person using gossip, so other people can't add them to their buddy list it seems.
Comment 10 foser (RETIRED) gentoo-dev 2006-05-05 08:57:49 UTC
I'm not quite sure what you mean. Who authorizes and who doesn't get added ? And what version are you using ?

I just got tested someone adding me with 0.9, no problem.

By now I prefer if you would stabilize 0.10.2 by the way.
Comment 11 Matthias Langer 2006-05-18 10:18:19 UTC
i've tried gossip-0.10.2 [ +dbus -debug +spell ] and loudmouth-1.0.1 [ -debug -doc +ssl ] on two x86 boxes. on BOX_1, i upgraded from 0.9 to 0.10.2, on BOX_2 gossip has not been installed before.

Results for BOX_1:
gossip starts, but i'm not able to connect to a server. the app just seems to ignore me completely when clicking on "connect". note that i've run etc-update, and also tried remerging gossip after "rm -r $(rlocate gossip | grep gconf)".

Results for BOX_2:
here gossip seems to work, at least partially. i was able to connect to a server, add someone to my contact list and do some conversation. however, i was appearing to be offline all the time for my chat partner (call him BUDDY) and gossip didn't ask me for authorisation after:

1.) i removed BUDDY from my contact list
2.) BUDDY removed me from his contact list
3.) BUDDY added me to his contact list again

BOX_1:
Portage 2.0.54-r2 (default-linux/x86/2006.0, gcc-3.4.5, glibc-2.3.6-r3, 2.6.16-gentoo-r7 i686)
=================================================================
System uname: 2.6.16-gentoo-r7 i686 AMD Athlon(tm) XP 2400+
Gentoo Base System version 1.6.14
dev-lang/python:     2.3.5-r2, 2.4.2
dev-python/pycrypto: [Not Present]
dev-util/ccache:     [Not Present]
dev-util/confcache:  [Not Present]
sys-apps/sandbox:    1.2.17
sys-devel/autoconf:  2.13, 2.59-r7
sys-devel/automake:  1.4_p6, 1.5, 1.6.3, 1.7.9-r1, 1.8.5-r3, 1.9.6-r1
sys-devel/binutils:  2.16.1
sys-devel/libtool:   1.5.22
virtual/os-headers:  2.6.11-r2
ACCEPT_KEYWORDS="x86"
AUTOCLEAN="yes"
CBUILD="i686-pc-linux-gnu"
CFLAGS="-march=athlon-xp -O2 -pipe"
CHOST="i686-pc-linux-gnu"
CONFIG_PROTECT="/etc /usr/kde/2/share/config /usr/kde/3.4/env /usr/kde/3.4/share/config /usr/kde/3.4/shutdown /usr/kde/3/share/config /usr/lib/X11/xkb /usr/share/config /usr/share/texmf/dvipdfm/config/ /usr/share/texmf/dvips/config/ /usr/share/texmf/tex/generic/config/ /usr/share/texmf/tex/platex/config/ /usr/share/texmf/xdvi/ /var/qmail/control"
CONFIG_PROTECT_MASK="/etc/eselect/compiler /etc/gconf /etc/terminfo /etc/env.d"
CXXFLAGS="-march=athlon-xp -O2 -pipe"
DISTDIR="/usr/portage/distfiles"
FEATURES="autoconfig collision-protect distlocks sandbox sfperms strict test"
GENTOO_MIRRORS="http://gentoo.ynet.sk/pub "
LANG="en_US.utf8"
LC_ALL="en_US.utf8"
LINGUAS="en de"
MAKEOPTS="-j2"
PKGDIR="/usr/portage/packages"
PORTAGE_TMPDIR="/var/tmp"
PORTDIR="/usr/portage"
PORTDIR_OVERLAY="/usr/local/portage"
SYNC="rsync://192.168.0.1/gentoo-portage"
USE="x86 3dnow 3dnowext X a52 aac aalib acpi alsa apm audiofile avi berkdb bitmap-fonts bonobo bzip2 cairo cdr cli crypt css cups curl dbus dri dts dvd dvdr dvdread eds emboss encode exif expat fam fbcon ffmpeg firefox flac foomaticdb fortran gd gdbm gif ginac glut gmp gnome gphoto2 gpm gstreamer gtk gtk2 gtkhtml guile hal icq idn imagemagick imlib ipv6 isdnlog java javascript jpeg jpeg2k junit lcms libg++ libwww mad matroska mikmod mime mmx mmxext mng motif mozsvg mp3 mpeg msn nautilus ncurses nls nptl nsplugin nvidia offensive ogg oggvorbis openal opengl pam pcre pdflib perl plotutils png posix pppd python quicktime readline real reflection ruby sdl session slang sockets speex spell spl sqlite sqlite3 sse ssl subtitles svg svga tcltk tcpd tetex theora tiff truetype truetype-fonts type1-fonts udev unicode usb vcd vorbis win32codecs wma xine xml xml2 xmms xorg xv xvid zlib video_cards_nvidia linguas_en linguas_de userland_GNU kernel_linux elibc_glibc"
Unset:  ASFLAGS, CTARGET, INSTALL_MASK, LDFLAGS, PORTAGE_RSYNC_EXTRA_OPTS, PORTAGE_RSYNC_OPTS

BOX_2:
Portage 2.0.54-r2 (default-linux/x86/2006.0, gcc-3.4.5, glibc-2.3.6-r3, 2.6.16.16 i686)
=================================================================
System uname: 2.6.16.16 i686 AMD Athlon(tm) XP 1900+
Gentoo Base System version 1.6.14
dev-lang/python:     2.3.5-r2, 2.4.2
dev-python/pycrypto: [Not Present]
dev-util/ccache:     [Not Present]
dev-util/confcache:  [Not Present]
sys-apps/sandbox:    1.2.17
sys-devel/autoconf:  2.13, 2.59-r7
sys-devel/automake:  1.4_p6, 1.5, 1.6.3, 1.7.9-r1, 1.8.5-r3, 1.9.6-r1
sys-devel/binutils:  2.16.1
sys-devel/libtool:   1.5.22
virtual/os-headers:  2.6.11-r2
ACCEPT_KEYWORDS="x86"
AUTOCLEAN="yes"
CBUILD="i686-pc-linux-gnu"
CFLAGS="-O2 -march=athlon-xp -pipe"
CHOST="i686-pc-linux-gnu"
CONFIG_PROTECT="/etc /usr/kde/2/share/config /usr/kde/3/share/config /usr/lib/X11/xkb /usr/share/config /var/qmail/control"
CONFIG_PROTECT_MASK="/etc/eselect/compiler /etc/gconf /etc/terminfo /etc/env.d"
CXXFLAGS="-O2 -march=athlon-xp -pipe"
DISTDIR="/usr/portage/distfiles"
FEATURES="autoconfig colission-protect distlocks sandbox sfperms strict"
GENTOO_MIRRORS="ftp://ftp.sh.cvut.cz/MIRRORS/gentoo/gentoo "
LC_ALL="en_US.UTF-8"
LINGUAS="en de"
MAKEOPTS="-j2"
PKGDIR="/usr/portage/packages"
PORTAGE_TMPDIR="/var/tmp"
PORTDIR="/usr/portage"
PORTDIR_OVERLAY="/usr/local/portage"
SYNC="rsync://rsync.europe.gentoo.org/gentoo-portage"
USE="x86 3dnow 3dnowext X acpi alsa apm audiofile avi berkdb bitmap-fonts browserplugin bzip2 bzlib cairo cli crypt curl dbus dri dvd emboss encode exif expat fam firefox foomaticdb fortran gd gdbm gif glut gmp gnome gphoto2 gpm gstreamer gtk gtk2 gtkhtml hal idn imlib ipv6 isdnlog java jpeg lcms libg++ libwww mad mikmod mime mmx mmxext mng motif mp3 mpeg ncurses nls nptl nsplugin nvidia ogg oggvorbis opengl pam pcre pdflib perl png pppd python quicktime readline real reflection ruby sdl session spell spl sqlite sse ssl tcpd tiff truetype truetype-fonts type1-fonts udev unicode usb vorbis win32codecs wma xine xml xml2 xorg xv xvid zlib linguas_en linguas_de userland_GNU kernel_linux elibc_glibc"
Unset:  ASFLAGS, CTARGET, INSTALL_MASK, LANG, LDFLAGS, PORTAGE_RSYNC_EXTRA_OPTS, PORTAGE_RSYNC_OPTS
Comment 12 Mark Loeser (RETIRED) gentoo-dev 2006-05-24 14:22:18 UTC
gnome: some feedback please? :)  The version suggested looks pretty iffy as well.

mlangc: could you test the current stable version and let me know if it is broken as well?  If it is, it may be best to just pull x86's stable marking from this package for the time being.
Comment 13 David Morgan 2006-05-24 14:33:19 UTC
I'm pretty sure we tested the current stable version and it worked...
Comment 14 Matthias Langer 2006-05-24 18:13:17 UTC
we did.
Comment 15 Matthias Langer 2006-05-24 19:44:12 UTC
i've verified again (with Halcy0n), that gossip-0.8 doesn't suffer from the problems  mentioned in comment #8 and comment #11 even when installed together with  loudmouth-1.0.1  [ +debug -doc +ssl ] (testing was done on BOX_1 from comment #11).

recapitulating:
  >=gossip-0.9 seems to have serious problems
  loudmouth-1.0.1 is ok
Comment 16 Mark Loeser (RETIRED) gentoo-dev 2006-05-26 11:02:57 UTC
I marked loudmouth-1.0.1 stable on x86, but gossip seems to have some real problems.  Please add us back when gossip is ready to go.

I'm adding Matthias to CC so he can try to help you.

Thanks
Comment 17 foser (RETIRED) gentoo-dev 2006-05-31 06:18:42 UTC
As I indicated before the reports are too vague to do anything with. I do not really care if you mark it stable or not.. 

My excuses to mark for not working with him, but I have had zero time lately to look into his reports and had no success reproducing them.
Comment 18 Mark Loeser (RETIRED) gentoo-dev 2006-05-31 15:54:03 UTC
(In reply to comment #17)
> As I indicated before the reports are too vague to do anything with. I do not
> really care if you mark it stable or not.. 

He tried to give you a more detailed explanation.  What information do you need?  Currently, I don't think it is in a state to mark it stable.  It went into an infinite loop and hung when I tried to use it to connect to my Google Talk account.
Comment 19 foser (RETIRED) gentoo-dev 2006-06-02 02:23:34 UTC
iirc I told Mark on IRC some things that could be important like jabber server, transports information/versions, loudmouth debug output, IM client on the other side, etc.

Anyway, here's another report I can't reproduce. I log in on my google account all the time, no lockups or anything. Just saying 'it doesn't work for me, so we cannot stabilize' helps noone. You say it goes into a loop, but give no evidence for it, there's not a single piece of tangible evidence here & no other reports to back up the claims. Am I supposed to ask for every little piece of info from people that are supposed to be seasoned testers ? And in a final act of upstanding tester-ethics you report a problem and then remove yourself from CC. Do you really expect me to take these reports seriously like this ?

I'm not saying gossip is problem free, it's too early in development for that. But it does the basics quite well and 0.9 (0.10/11 by now) do a lot better than any older version, so as I see it the x86 arch testing team just condemns 'stable' users to the least reliable version in the tree.
Comment 20 Matthias Langer 2006-06-02 03:43:01 UTC
> ..., so as I see it the x86 arch testing team just condemns
> 'stable' users to the least reliable version in the tree.

I'm sure that gossip-0.10.x works fine for you, otherwise you would not have made a stable request. But that gossip works on 'box A' and 'box B' does by no means imply that it also works on 'box C' and 'box D'. I've spent several hours with testing various gossip versions on two x86 machines just for this bug (i nether use gossip nor jabber normally), documented clearly enough what went wrong und found that the current stable version is the one that works best for me. There is nothing more i can do; If you think my reports are too vague, why don't you ask for clarification ?
Comment 21 foser (RETIRED) gentoo-dev 2006-06-02 04:14:55 UTC
Well, I agree I haven't made that too clear (comment #10 is about clarification however), the one time in IRC you contacted me I gave some pointers as to what could explain the problem. You just caught me at a bad moment there, because I had no time to dive into it there and then.

Oh btw, gossip by design doesn't show contacts that aren't authed on both sides, so that could explain your missing contacts.
Comment 22 Matthias Langer 2006-06-02 05:56:01 UTC
(In reply to comment #21)
I've also understood comment #10 as clarification request; that's one of the reasons i've posted comment #11.

As far as i remember, you told me in IRC to compile gossip with debuging messages. Thus i recompiled gossip with [ +debug ], but that didn't reveal anything, as i didn't see a single 'debuging message'. If you meant something else by 'compiling with debuging messages' let me now ....
Comment 23 Matthias Langer 2006-06-02 06:01:44 UTC
maybe i should have added that i've also compiled loudmouth with debug ...
Comment 24 Mark Loeser (RETIRED) gentoo-dev 2006-06-02 06:09:53 UTC
(In reply to comment #19)
> iirc I told Mark on IRC some things that could be important like jabber server,
> transports information/versions, loudmouth debug output, IM client on the other
> side, etc.

I'll assume you mean Matthias, because I don't remember talking to you about this :)


> Anyway, here's another report I can't reproduce. I log in on my google account
> all the time, no lockups or anything. Just saying 'it doesn't work for me, so
> we cannot stabilize' helps noone. You say it goes into a loop, but give no
> evidence for it, there's not a single piece of tangible evidence here & no
> other reports to back up the claims. Am I supposed to ask for every little
> piece of info from people that are supposed to be seasoned testers ? And in a
> final act of upstanding tester-ethics you report a problem and then remove
> yourself from CC. Do you really expect me to take these reports seriously like
> this ?

You already said that you aren't interested in getting it marked stable.  Matthias is trying to help you with other problems he found already, but you just keep saying the reports are too vague and I haven't seen you asking for clarification on anything.  My comment was merely to say, I found another way it is broken.  Perhaps it was something you already knew of, since both Matthias and I seem to be running into the same problems.  Just because it works for you doesn't mean it doesn't work for the rest of us.


> I'm not saying gossip is problem free, it's too early in development for that.
> But it does the basics quite well and 0.9 (0.10/11 by now) do a lot better than
> any older version, so as I see it the x86 arch testing team just condemns
> 'stable' users to the least reliable version in the tree.

If it is that early in its development, then perhaps it shouldn't have been marked stable in the first place.  I'm going to add x86@ back so other people from the team can chime in with their comments on that.
Comment 25 Matthias Langer 2006-06-02 06:37:36 UTC
removing *this from CC as i don't want to get mails twice
Comment 26 Mark Loeser (RETIRED) gentoo-dev 2006-06-03 14:36:12 UTC
Well, for the time being, the x86 team doesn't feel this is stable enough to be marked as such.  When you want the next version to go stable, please CC us on the bug and we'll give the new version a go.  mlangc also ran into the same problem that I did when trying to connect to gtalk.

We have moved gossip-0.8 back to ~x86 in the meantime, so you can drop older versions if you want without worrying about us.

Thanks
Comment 27 foser (RETIRED) gentoo-dev 2006-06-04 03:22:44 UTC
This is getting more ridiculous by the day. 24 hours ago you still wanted to work it out, by now the mood has swung and you decide for some reason -without a shred of real evidence- that it doesn't work for you and as such is 'not ready' and then even move an old stable version to ~arch (which is testing for things to go stable, not a place to put stuff that is not ready). Gossip has been in the tree for years now and no bugreports have have been filed lately about people unable to connect to anything, still you feel it's not ready. In the mean time 3 other arches have it marked stable. What doesn't count up here ? 

If the x86 team can't come up with proof of any sort, I'll have to do it's stable marking myself again. I rather trust gossips real userbase than 2 random self-assigned 'testers' who do not use the application in the first place and file no bugreports in the second place to test for me.

I think the x86 is overstepping boundaries here, adding devrel for an unbiased look at the situation.
Comment 28 Alec Warner (RETIRED) archtester gentoo-dev Security 2006-06-04 08:01:12 UTC
(In reply to comment #27)
> This is getting more ridiculous by the day. 24 hours ago you still wanted to
> work it out, by now the mood has swung and you decide for some reason -without
> a shred of real evidence- that it doesn't work for you and as such is 'not
> ready' and then even move an old stable version to ~arch (which is testing for
> things to go stable, not a place to put stuff that is not ready). Gossip has
> been in the tree for years now and no bugreports have have been filed lately
> about people unable to connect to anything, still you feel it's not ready. In
> the mean time 3 other arches have it marked stable. What doesn't count up here
> ? 
> 
> If the x86 team can't come up with proof of any sort, I'll have to do it's
> stable marking myself again. I rather trust gossips real userbase than 2 random
> self-assigned 'testers' who do not use the application in the first place and
> file no bugreports in the second place to test for me.
> 
> I think the x86 is overstepping boundaries here, adding devrel for an unbiased
> look at the situation.
> 

Er, you said above you could care less if it went stable.  Quoting

"As I indicated before the reports are too vague to do anything with. I do not
really care if you mark it stable or not.."

if you have someone that it works for (besides yourself) I'd be glad to speak to them.  Maybe we are doing something wrong?
Comment 29 Mark Loeser (RETIRED) gentoo-dev 2006-06-04 08:04:57 UTC
(In reply to comment #27)
> This is getting more ridiculous by the day. 24 hours ago you still wanted to
> work it out, by now the mood has swung and you decide for some reason -without
> a shred of real evidence- that it doesn't work for you and as such is 'not
> ready' and then even move an old stable version to ~arch (which is testing for
> things to go stable, not a place to put stuff that is not ready). Gossip has
> been in the tree for years now and no bugreports have have been filed lately
> about people unable to connect to anything, still you feel it's not ready. In
> the mean time 3 other arches have it marked stable. What doesn't count up here
> ? 

We tried to give you ways to reproduce the bugs both of us were running into, but you said...

Comment #17
As I indicated before the reports are too vague to do anything with. I do not
really care if you mark it stable or not.. 


We had 3 people (myself included) verify the problems we tried to tell you how to reproduce.  You told us our reports were vague, and didn't tell us the information you needed (atleast, you never told us in the bug so that we could all try and get you more information).


> If the x86 team can't come up with proof of any sort, I'll have to do it's
> stable marking myself again. I rather trust gossips real userbase than 2 random
> self-assigned 'testers' who do not use the application in the first place and
> file no bugreports in the second place to test for me.

If an application is to be declared "stable", it should just work for anyone that tries to use it.  As it stands, it does not.  3 of us reproduced the same exact problems, so it seems to me that the problem is not with any of us.


> I think the x86 is overstepping boundaries here, adding devrel for an unbiased
> look at the situation.

::sighs:: ...We tried to work with you.  This bug has been open for ages.  It wasn't going anywhere, so we did what we thought was best for our users and so you could remove any versions you wanted without x86 being a hinderance to you.  

Is it really necessary to escalate this...its all really blowing a small problem out of proportion.  If you think we are wrong, you could have come to me and tried to work this out, but I guess trying to work together is out of the question...which is really a sad state of affairs.
Comment 30 solar (RETIRED) gentoo-dev 2006-06-04 08:17:58 UTC
(In reply to comment #27)

> If the x86 team can't come up with proof of any sort, I'll have to do it's
> stable marking myself again.

Don't do that. That would cause alot of drama for no good reason.
Comment 31 Bryan Østergaard (RETIRED) gentoo-dev 2006-06-04 12:33:43 UTC
Reading this bug it seems to me that the x86 has some genuine concerns about about gossip and that those issues deserves a closer look.

Foser, if you need any information from the x86 team about any possible issues with gossip / loudmouth I think it'd be best to state exactly what information you need on this bug. Same thing goes for x86 team if you need any info from foser.

I hope everybody can work together that way and that we can avoid all the unnecessary discussions about who said what on irc or some other random medium.
Comment 32 foser (RETIRED) gentoo-dev 2006-06-06 05:46:46 UTC
(In reply to comment #31)
> Reading this bug it seems to me that the x86 has some genuine concerns about
> about gossip and that those issues deserves a closer look.

I do not deny that, but x86 keeps removing itself from the bug without adding any evidence thats near useful. To me the original (0.9) release has been ~arch long enough (and stable on 3 non-x86 arches) to make me pretty sure there are no real problems with it, I'd say gossip 0.10 is even better, but getting that stable would mean another multi-year plan at this rate.

> Foser, if you need any information from the x86 team about any possible issues
> with gossip / loudmouth I think it'd be best to state exactly what information
> you need on this bug. Same thing goes for x86 team if you need any info from
> foser.

I can't reproduce the problem, I don't know what to look for. There are so many points of failure I can't triage it by asking a few simple questions, especially because obviously both testers aren't normally users of gossip and not familiar with it's behaviour, which explains to me at least one 'problem' described (see comment #11 & #21 ). Also getting something stable is not a priority to me. When I am short on time a stable marking bug is not the first thing I dive into, to explain comment #17 .

On IRC I indicated to Matthias what kind of output could be helpful (loudmouth/gossip debug output)., server to connect to, etc.. Only in comment #18 a server is mentioned, 0.9/0.8 weren't fit to work 100% with googletalk because of the ways it interacts with clients, so another 'problem' solved. Use a 'real'/oldskool jabberserver to test, not the new kid on the block.

As far as the devrel involvement goes, I added the CC because it adds a fresh perspective and half the problem here is the lacking communication between the x86 team and me as maintainer (hence, devrel).

Afaic the x86 team is making a mistake by taking questionable testing without any reproducable evidence over the gossip userbase we get our usual reports from and taking serious steps by destabilizing 0.8 (!) months after it had been made stable without a single bugreport to indicate otherwise. To me the x86 team -like we expect from users- should be able to prove there is really a problem, before I go spend time on solving anything, but instead the x86 team is acting (moving 0.8 to ~x86 again) on something I cannot even reproduce and then removing itself from the bug. It's the world upside down, the fact that I have to prove it works, instead of them proving it doesn't (comment #28). That is overstepping boundaries.

If you want to be taken seriously as an arch testing team, make some procedures , fill out a normal bugreport form and take the judgement of the maintainer seriously. It is obvious to me that you guys do not follow jabber and/or gossip development and I think that predicts at least 75% of the 'problems' stated here: unfamiliarity with the application in question. No doubt in my mind 0.9/0.10 are fit to be stable, like 0.8 was, but only better. Why do I feel like I'm fighting cluelessness here ?
Comment 33 Bryan Østergaard (RETIRED) gentoo-dev 2006-06-06 07:28:40 UTC
(In reply to comment #32)
> (In reply to comment #31)
> > Reading this bug it seems to me that the x86 has some genuine concerns about
> > about gossip and that those issues deserves a closer look.
> 
> I do not deny that, but x86 keeps removing itself from the bug without adding
> any evidence thats near useful. To me the original (0.9) release has been ~arch
> long enough (and stable on 3 non-x86 arches) to make me pretty sure there are
> no real problems with it, I'd say gossip 0.10 is even better, but getting that
> stable would mean another multi-year plan at this rate.
My comment wasn't meant at you alone and I agree that removing the x86 team isn't going to solve anything. If the x86 wants anything fixed they should obviously stay cc'ed on the bug.
> 
> > Foser, if you need any information from the x86 team about any possible issues
> > with gossip / loudmouth I think it'd be best to state exactly what information
> > you need on this bug. Same thing goes for x86 team if you need any info from
> > foser.
> 
> I can't reproduce the problem, I don't know what to look for. There are so many
> points of failure I can't triage it by asking a few simple questions,
> especially because obviously both testers aren't normally users of gossip and
> not familiar with it's behaviour, which explains to me at least one 'problem'
> described (see comment #11 & #21 ). Also getting something stable is not a
> priority to me. When I am short on time a stable marking bug is not the first
> thing I dive into, to explain comment #17 .
Not being able to reproduce a bug doesn't mean there's no bug though. That's also why the x86 team needs to provide all the relevant information they can.

> 
> On IRC I indicated to Matthias what kind of output could be helpful
> (loudmouth/gossip debug output)., server to connect to, etc.. Only in comment
> #18 a server is mentioned, 0.9/0.8 weren't fit to work 100% with googletalk
> because of the ways it interacts with clients, so another 'problem' solved. Use
> a 'real'/oldskool jabberserver to test, not the new kid on the block.
> 
Would adding einfo's to the ebuild help solve this? I assume normal users wouldn't know it doesn't work with googletalk and other new kids on the block.

> As far as the devrel involvement goes, I added the CC because it adds a fresh
> perspective and half the problem here is the lacking communication between the
> x86 team and me as maintainer (hence, devrel).
That's exactly what I'm trying to do and I believe this could be solved with better communication.
> 
> Afaic the x86 team is making a mistake by taking questionable testing without
> any reproducable evidence over the gossip userbase we get our usual reports
> from and taking serious steps by destabilizing 0.8 (!) months after it had been
> made stable without a single bugreport to indicate otherwise. To me the x86
> team -like we expect from users- should be able to prove there is really a
> problem, before I go spend time on solving anything, but instead the x86 team
> is acting (moving 0.8 to ~x86 again) on something I cannot even reproduce and
> then removing itself from the bug. It's the world upside down, the fact that I
> have to prove it works, instead of them proving it doesn't (comment #28). That
> is overstepping boundaries.
I think that several devs / arch testers on the x86 team not being able to connect is sufficient evidence there's a problem. I don't know whether that problem can be solved using einfo's as I suggested or something more is needed. Assuming that the devs / arch testers in question don't know anything about the application puts them in pretty much the same situation as a large part of our userbase imo. Basic functionality should work out of the box and any obvious problems should be pointed out somehow imo.

> 
> If you want to be taken seriously as an arch testing team, make some procedures
> , fill out a normal bugreport form and take the judgement of the maintainer
> seriously. It is obvious to me that you guys do not follow jabber and/or gossip
> development and I think that predicts at least 75% of the 'problems' stated
> here: unfamiliarity with the application in question. No doubt in my mind
> 0.9/0.10 are fit to be stable, like 0.8 was, but only better. Why do I feel
> like I'm fighting cluelessness here ?
I'm repeating myself here but I don't expect users to follow jabber or gossip development either and I don't think it's a reasonable requirement. I don't know anything about jabber or gossip but as a user I'd expect to be able to emerge gossip and connect to servers (taking into account any warnings / hints portage has thrown at me).
Comment 34 Matthias Langer 2006-06-06 16:31:13 UTC
(In reply to comment #32)
> On IRC I indicated to Matthias what kind of output could be helpful
> (loudmouth/gossip debug output)., server to connect to, etc..

I really don't want to escalte this discussion, but the only thing you told me on IRC is to compile loudmouth/gossip with debugging output (just looked at the chat logs).
Comment 35 Mark Loeser (RETIRED) gentoo-dev 2006-06-06 20:53:53 UTC
kloeri really said almost everything I would have...except... :)

The reason we removed ourselves was because the bug wasn't going anywhere, and mlangc was going to try and work with you.  You never really told us on the bug what you required as far as information to debug it.  It looks like Matthias wasn't really sure what you needed either.  Could be communication problems, but it doesn't change the fact that the application is very buggy for us.

Sorry, but I still don't see any reason to change the actions we have taken with regards to gossip's keywording.  I just searched, and I can't find any bug reports, or anything on the forums about the keyword being pulled, so I'm assuming not many people even used it because it was so buggy for us.  If you tell us on the bug exactly the information you require, I'm sure our ATs would be happy to help, but that doesn't mean I'm going to keep x86@ on the bug forever, because it just ends up cluttering all of our other bugs.
Comment 36 foser (RETIRED) gentoo-dev 2006-06-08 09:05:08 UTC
Googletalk wasn't widely available at the time of those releases, should we now add warnings back in time ? This is not back-to-the-future. I think that is the only 'problem' here really. Use a real jabber server (not one in beta stage), learn how gossip behaves, it's not IM-app 'x' you are used to.

And if I know a bunch of users left and right and ask them if they have any trouble and they do not, it means to me the testers are doing something wrong. If i get no bugreports that even remotely resemble the testers problems, I think the testers are doing something wrong. The evidence is still lacking from their side, so what is there for me to say ? Nice they can't connect (whatever that means), but I can't possibly change that without any info. If you're a self-assigned 'tester', do your job and get some info. And yes, LM/Gossip debug output would help, but I still got none.

The fact that normal gossip users have no problems and 2 testers who don't use the app do gives me the idea that they do not know what they're doing. But I give up, let the test-police do their unsubstantiated overcontrol. No procedures, no nothing, just unproven claims.
Comment 37 Chris Gianelloni (RETIRED) gentoo-dev 2006-06-08 10:16:42 UTC
So I've been watching this and keeping my mouth shut, but I'm really sick of reading this drivel.

Can we not just mark the older version back to stable, *after* a warning notice is put in the ebuild about it not working with GoogleTalk and be done with it?

I mean, is there really any more reason to keep up this crap?  I mean, I agree that the application should work out of the box.  If it does not work, then yes, it should be noted why it doesn't work.  Yes, you *should* do "back-to-the-future" and add notices to *stable* packages when they don't work with public services.

If 0.8 works fine with "normal" jabber services, then it should have its KEYWORDS put back to x86 and a nice notice should be made.  If 0.9 (or newer) isn't expected to work with a particular service, then put a warning in the ebuild saying as much and let the testing happen on the services it does support so this can be closed.  The constant childish bickering and thinly-veiled (and poorly executed ;P) 1984 references don't belong here.

So really people, can we just put our egos aside for a second and work on the *bug* at hand?

This has presented by the letter "4" and the number "orange"... Thank you.
Comment 38 Joshua Jackson (RETIRED) gentoo-dev 2006-06-08 10:22:27 UTC
(In reply to comment #36)
> Nice they can't connect (whatever
> that means), 

One has to wonder about the inability to understand that being unable to connect to a server is a foreign concept here. Might just want to try connecting to a server when you don't have a internet connection, and you would probably get the same kind of result and understand what the concept of not connecting is about. Sarcasm entirely meant in the above.

> The fact that normal gossip users have no problems and 2 testers who don't use
> the app do gives me the idea that they do not know what they're doing. But I
> give up, let the test-police do their unsubstantiated overcontrol. No
> procedures, no nothing, just unproven claims.
> 

Foser, you have been less then helpful on this bug as well. I don't particularly care for the fact that you are putting this entirely onto the testers back. As you have said you are the maintainer and have a better understanding of the application and how it works internally then the rest of us. With that knowledge there should be some inkling as to what could be wrong. You have not bothered to even suggest anything other then the debug compiling to matthias. I will get with him on this and try and arrange something.

You've also mentioned using another server then googletalk as its newer then these releases are...yet you have not suggested any servers from your vast repository of knowledge of servers that are old school enough. If you want to talk about communication, then lets talk about communication on both sides, without pointing the finger at each other and involving devrel in something that has absolutely nothing to do with them. It was escalated far beyond what it should of been.
Comment 39 Matthias Langer 2006-06-08 10:50:08 UTC
(In reply to comment #36)
> And yes, LM/Gossip debug output would help, but I still got none.
hmm, if you answer comment #22 i can see what i can do ...
Comment 40 Joshua Jackson (RETIRED) gentoo-dev 2006-06-26 10:30:35 UTC
I've personally tested it at this point and had the same exact issue. At this point, as there hasn't be any further communication by both parties. Even with a willingness to do something with some guidance from the maintainer, I don't feel that this will be going anywhere anytime soon. I also can't recommend that this goes stable, so as far as that goes, I think it should be shelved for the moment til concerns are addressed. Mlangc, as always thank you for your willingness to test out any and all packages for the x86 team.

Comment 41 foser (RETIRED) gentoo-dev 2006-07-08 06:43:19 UTC
so much for arch teams.. abstract away responsibility
Comment 42 Hans de Graaff gentoo-dev Security 2007-01-27 09:42:24 UTC
I just had to recompile gossip due to dbus upgrades, and realized that I was still using a testing version on my otherwise mostly stable laptop. I've been using gossip 0.16 with loudmouth 1.0.1 there for 4 months without any issue, the package has been rock-stable for me (using my ISP's jabber server jabber.xs4all.nl). I was about to open a bug to stabilize it when I found this bug, so I figured I'd add a comment here. 

Foser: do you think it makes sense to open a new stablize bug for this or to re-open this one? I think it's a shame that we don't have a stable version of Gossip in portage.
Comment 43 Gilles Dartiguelongue (RETIRED) gentoo-dev 2007-01-27 10:24:02 UTC
considering 0.22 is the latest version using loudmouth, I guess we could stabilize this version directly. Next version of gossip will use telepathy so we might be slower to stabilize future versions for a while.

Could the bug be reopened and the summary changed ?
Comment 44 Petteri Räty (RETIRED) gentoo-dev 2007-01-27 10:49:33 UTC
(In reply to comment #43)
> considering 0.22 is the latest version using loudmouth, I guess we could
> stabilize this version directly. Next version of gossip will use telepathy so
> we might be slower to stabilize future versions for a while.
> 
> Could the bug be reopened and the summary changed ?
> 

This bug was originally about gossip-0.9 and carries so much noise that IMHO is is better to open a new bug if you want 0.22 stable, but 0.22 has not been in the tree for the full month yet so it can't be marked stable before February 13th.