[slrn] Omitting References Header
Blinky the Shark
has broken threading. In each of those two posts the References header
was missing.
WTF?
Has anyone else seen this? I broke my resolve to only stay with stable
versions of slrn, a couple months ago, and tried one or two of the
"custom" slrns floating about. One of the issues discovered there was a
problem or two with long headers. Might this be related to one of
these issues -- but instead of showing an error it's just dropping the
header silently and therefore, yes, breaking threading?
--
Blinky
Killing all posts from Google Groups
The Usenet Improvement Project - http://improve-usenet.org
Troy Piggins
* & my replies are inline below :
> In another group, today I've had a couple reports that one of my posts
> has broken threading. In each of those two posts the References header
> was missing.
>
> WTF?
>
> Has anyone else seen this? I broke my resolve to only stay with stable
> versions of slrn, a couple months ago, and tried one or two of the
> "custom" slrns floating about. One of the issues discovered there was a
> problem or two with long headers. Might this be related to one of
> these issues -- but instead of showing an error it's just dropping the
> header silently and therefore, yes, breaking threading?
I haven't noticed it, but I've moved on from the version you are
using.
If the post you are referring to is the one in a.f.c-a, the post
you followed up to had a pretty long References: header, your
post completely omitted a Ref: header, but then the subsequent
one did quote it fine. Leads one to think that the length of the
Ref: header does indeed have something to do with it.
You're not going to like my suggestion, but I'll provide it
anyway :) Get the latest source and compile it :) John E Davis
is actively working on it with a view to releasing 0.9.9
"soon". It's not far off the official release I believe.
That, or wait a little while and get the stable release 0.9.9
when it comes out - reckon there'll be binaries provided for it
to take away the hassles of compiling yourself.
--
Troy Piggins | http://piggo.com/~troy/slrn ___ | __ __
* Currently using slrn pre0.9.9-60 |___ | |/ |/ |
* Try it: ___| | | | |
$ svn co https://slrn.svn.sourceforge.net/svnroot/slrn/trunk slrn
Blinky the Shark
> * Blinky the Shark is quoted
> * & my replies are inline below :
>> In another group, today I've had a couple reports that one of my posts
>> has broken threading. In each of those two posts the References header
>> was missing.
>>
>> WTF?
>>
>> Has anyone else seen this? I broke my resolve to only stay with stable
>> versions of slrn, a couple months ago, and tried one or two of the
>> "custom" slrns floating about. One of the issues discovered there was a
>> problem or two with long headers. Might this be related to one of
>> these issues -- but instead of showing an error it's just dropping the
>> header silently and therefore, yes, breaking threading?
>
> I haven't noticed it, but I've moved on from the version you are
> using.
>
> If the post you are referring to is the one in a.f.c-a, the post
Yes, and there were two of them there. A couple afcans pointed them out
to me.
> you followed up to had a pretty long References: header, your
> post completely omitted a Ref: header, but then the subsequent
> one did quote it fine. Leads one to think that the length of the
> Ref: header does indeed have something to do with it.
That's all I could think of. That's far worse than the "folding" errors
I was getting before I did something suggested in here to fix that.
> You're not going to like my suggestion, but I'll provide it
I know what's coming. :)
> anyway :) Get the latest source and compile it :) John E Davis
> is actively working on it with a view to releasing 0.9.9
> "soon". It's not far off the official release I believe.
No, I've done enough of that shit. I'll stick with real releases.
> That, or wait a little while and get the stable release 0.9.9
> when it comes out - reckon there'll be binaries provided for it
> to take away the hassles of compiling yourself.
I believe I'll wait a while to see if it's really stable. Shame. I've
never had to feel that way about slrn in the past.
Thanks for the input, Peter.
I'll just look up at my headers on each of my posts and make sure
there's a reference header there. If not, I won't post that article.
Troy Piggins
Hah!
>> anyway :) Get the latest source and compile it :) John E Davis
>> is actively working on it with a view to releasing 0.9.9
>> "soon". It's not far off the official release I believe.
>
> No, I've done enough of that shit. I'll stick with real releases.
I knew that'd be your response... :)
>> That, or wait a little while and get the stable release 0.9.9
>> when it comes out - reckon there'll be binaries provided for it
>> to take away the hassles of compiling yourself.
>
> I believe I'll wait a while to see if it's really stable. Shame. I've
> never had to feel that way about slrn in the past.
>
> Thanks for the input, Peter.
:) We are not the same person!
> I'll just look up at my headers on each of my posts and make sure
> there's a reference header there. If not, I won't post that article.
I'll keep an eye on mine also, although I suspect I don't get
involved in as long a threads as you do, you little
power-usenetter you!
--
Troy Piggins | http://piggo.com/~troy/slrn ___ | __ __
* Currently using slrn pre0.9.9-62 |___ | |/ |/ |
Blinky the Shark
> I'll keep an eye on mine also, although I suspect I don't get involved
> in as long a threads as you do, you little power-usenetter you!
I don't contribute to them much any more, but this bug will bite even if
I post once to a mondo thread. :)
Blinky the Shark
New Info: I looked at the post I remember, where slrn didn't insert a
References header, and the post I was responding to didn't have all that
long a References header to add to. That post (that I replied to) only
has this (I've wrapped it here for convenience):
References:
<64fac514-416d-4352...@i29g2000prf.googlegroups.com>
<5442cdfe-fd44-4356...@i12g2000prf.googlegroups.com>
<QqCdnYiKCZFkVtva...@comcast.com>
<4746f342$0$27043$6c5e...@news.westnet.com>
The MID of that post itself (which should've been added in mine) was:
<Xns99F241AC5D2CFop...@127.0.0.1>
I've *certainly* seen (and had slrn correctly process) longer References
headers than what that should've produced.
Troy Piggins
* & my replies are inline below :
> Troy Piggins wrote:
>>> Thanks for the input, Peter.
>>
>>:) We are not the same person!
>>
>>> I'll just look up at my headers on each of my posts and make sure
>>> there's a reference header there. If not, I won't post that article.
>>
>> I'll keep an eye on mine also, although I suspect I don't get
>> involved in as long a threads as you do, you little
>> power-usenetter you!
>
> New Info: I looked at the post I remember, where slrn didn't insert a
> References header, and the post I was responding to didn't have all that
> long a References header to add to. That post (that I replied to) only
> has this (I've wrapped it here for convenience):
>
> References:
> <64fac514-416d-4352...@i29g2000prf.googlegroups.com>
> <5442cdfe-fd44-4356...@i12g2000prf.googlegroups.com>
> <QqCdnYiKCZFkVtva...@comcast.com>
> <4746f342$0$27043$6c5e...@news.westnet.com>
>
> The MID of that post itself (which should've been added in mine) was:
>
> <Xns99F241AC5D2CFop...@127.0.0.1>
>
> I've *certainly* seen (and had slrn correctly process) longer References
> headers than what that should've produced.
Yeah, that M-ID is unique. Never seen that hostname before...
Blinky the Shark
1. MIDs are supposed to be unique.
2. slrn shouldn't be looking up domain names.
3. I've responded to hundreds of posts from that participant without
issues.
Troy Piggins
* & my replies are inline below :
> Troy Piggins wrote:
>>> <4746f342$0$27043$6c5e...@news.westnet.com>
>>>
>>> The MID of that post itself (which should've been added in mine) was:
>>>
>>> <Xns99F241AC5D2CFop...@127.0.0.1>
>>>
>>> I've *certainly* seen (and had slrn correctly process) longer References
>>> headers than what that should've produced.
>>
>> Yeah, that M-ID is unique. Never seen that hostname before...
>
> 1. MIDs are supposed to be unique.
I know, I was being sarcastic but that doesn't translate through
keyboards very well. I was having a dig at the hostname which is
as diametrically opposite to unique as you can get.
> 2. slrn shouldn't be looking up domain names.
>
> 3. I've responded to hundreds of posts from that participant without
> issues.
Agreed and fair enough. Shouldn't be happening regardless.
Blinky the Shark
> * Blinky the Shark is quoted * & my replies are inline below :
>> Troy Piggins wrote: [---=| Quote block shrinked by t-prot: 24 lines
>> snipped |=---]
>>>> <4746f342$0$27043$6c5e...@news.westnet.com>
>>>>
>>>> The MID of that post itself (which should've been added in mine)
>>>> was:
>>>>
>>>> <Xns99F241AC5D2CFop...@127.0.0.1>
>>>>
>>>> I've *certainly* seen (and had slrn correctly process) longer
>>>> References headers than what that should've produced.
>>>
>>> Yeah, that M-ID is unique. Never seen that hostname before...
>>
>> 1. MIDs are supposed to be unique.
>
> I know, I was being sarcastic but that doesn't translate through
> keyboards very well. I was having a dig at the hostname which is as
> diametrically opposite to unique as you can get.
It's worse than the Agent MIDs, IMO.
>> 2. slrn shouldn't be looking up domain names.
>>
>> 3. I've responded to hundreds of posts from that participant without
>> issues.
>
> Agreed and fair enough. Shouldn't be happening regardless.
That group just bristles with Groperhuggers, and they delighted in
pointing out that my post broke a thread (out of my tens of thousands of
posts) and suggesting I use something better like GG. Until I change to
an older (or, less likely, newer) version of slrn, I'm avoiding a
recurrance by switching to Pan for that group. I use Pan also, anyway
and have it set up with several servers, including news.individual.net
through which I do most of my Usenetting, so that's no big loss -- it's
just a bug loss.
Wolfgang Schelongowski
>Troy Piggins wrote:
>
>> I'll keep an eye on mine also, although I suspect I don't get involved
>> in as long a threads as you do, you little power-usenetter you!
>
>I don't contribute to them much any more, but this bug will bite even if
>I post once to a mondo thread. :)
Try to test it. Start with a F'up identical to the one with the
omitted References header, but set Newsgroups to one of the test
groups local to your server. Add
Distribution: local
for good measure. If that article is broken, too, try to narrow the
problem - you know the drill.
--
The first entry of Sin into the mind occurs when, out of cowardice or
conformity or vanity, the Real is replaced by a comforting lie.
-- Integritas, Consonantia, Claritas
Blinky the Shark
> Blinky the Shark <no....@box.invalid> writes:
>
>>Troy Piggins wrote:
>>
>>> I'll keep an eye on mine also, although I suspect I don't get involved
>>> in as long a threads as you do, you little power-usenetter you!
>>
>>I don't contribute to them much any more, but this bug will bite even if
>>I post once to a mondo thread. :)
>
> Try to test it. Start with a F'up identical to the one with the
> omitted References header, but set Newsgroups to one of the test
> groups local to your server. Add
> Distribution: local
> for good measure. If that article is broken, too, try to narrow the
> problem - you know the drill.
Yeah, I thought of that and will when I get around to it. As for now,
I've switched to Pan in that group (since no others would make a big
deal out of the issue). I use Pan every day for a few groups/servers,
even thought I use slrn more; so that's not a big hardship.
Shmuel Metz
at 12:38 AM, Blinky the Shark <no....@box.invalid> said:
>In another group, today I've had a couple reports that one of my posts
>has broken threading.
Am I the only one to see the irony in this?
>but instead of showing an error it's just dropping the
>header silently and therefore, yes, breaking threading?
How's it feel to be wearing that shoe?
--
Shmuel (Seymour J.) Metz, SysProg and JOAT <http://patriot.net/~shmuel>
Unsolicited bulk E-mail subject to legal action. I reserve the
right to publicly post or ridicule any abusive E-mail. Reply to
domain Patriot dot net user shmuel+news to contact me. Do not
reply to spam...@library.lspace.org
Blinky the Shark
> In <slrnfkhh15....@thurston.blinkynet.net>, on 11/25/2007 at
> 12:38 AM, Blinky the Shark <no....@box.invalid> said:
>
>>In another group, today I've had a couple reports that one of my posts
>>has broken threading.
>
> Am I the only one to see the irony in this?
>
>>but instead of showing an error it's just dropping the header silently
>>and therefore, yes, breaking threading?
>
> How's it feel to be wearing that shoe?
I dunno, because I'm not -- for two reasons. 1. It's one post of tens
of thousands, not "just the way the program works". 2. I immediately
adjusted by switching for that group to one of the several other clients
that I have installed and configured and am fluent in; if I get more
datapoints elsewhere we'll have more to work with.
Peter J Ross
<no....@box.invalid> wrote:
> In another group, today I've had a couple reports that one of my posts
> has broken threading. In each of those two posts the References header
> was missing.
>
> WTF?
>
> Has anyone else seen this?
Not with slrn 0.9.9. I've checked some of my recent posts and long
lines in References are present and correct. If there was a problem in
0.9.8.1pl2 and/or pl2.1, nobody drew it to my attention.
> I broke my resolve to only stay with stable
> versions of slrn, a couple months ago, and tried one or two of the
> "custom" slrns floating about. One of the issues discovered there was a
> problem or two with long headers. Might this be related to one of
> these issues -- but instead of showing an error it's just dropping the
> header silently and therefore, yes, breaking threading?
That seems likely.
One solution is to upgrade to the latest slrn sources from the SVN
repository and keep them up to date. I don't really recommend this to
anybody who doesn't enjoy bug-hunting, unless they absolutely can't do
without UTF-8 support.
The alternative is to stick to the stable 0.9.8.1 release until 0.9.9
is ready for release. The wait is unlikely to be a long one.
--
PJR :-)
Blinky the Shark
> In news.software.readers on 25 Nov 2007 00:38:52 GMT, Blinky the Shark
><no....@box.invalid> wrote:
>
>> In another group, today I've had a couple reports that one of my posts
>> has broken threading. In each of those two posts the References header
>> was missing.
>>
>> WTF?
>>
>> Has anyone else seen this?
>
> Not with slrn 0.9.9. I've checked some of my recent posts and long
> lines in References are present and correct. If there was a problem in
> 0.9.8.1pl2 and/or pl2.1, nobody drew it to my attention.
And as I've posted since, the References header would only have
contained, IIRC, five MIDs.
>> I broke my resolve to only stay with stable
>> versions of slrn, a couple months ago, and tried one or two of the
>> "custom" slrns floating about. One of the issues discovered there was a
>> problem or two with long headers. Might this be related to one of
>> these issues -- but instead of showing an error it's just dropping the
>> header silently and therefore, yes, breaking threading?
>
> That seems likely.
>
> One solution is to upgrade to the latest slrn sources from the SVN
> repository and keep them up to date. I don't really recommend this to
> anybody who doesn't enjoy bug-hunting, unless they absolutely can't do
> without UTF-8 support.
>
> The alternative is to stick to the stable 0.9.8.1 release until 0.9.9
> is ready for release. The wait is unlikely to be a long one.
I"ll probably revert. I've obviously already done too much "upgrading".
:)
Peter J Ross
<no....@box.invalid> wrote:
> Peter J Ross wrote:
>> In news.software.readers on 25 Nov 2007 00:38:52 GMT, Blinky the Shark
>><no....@box.invalid> wrote:
>>
>>> In another group, today I've had a couple reports that one of my posts
>>> has broken threading. In each of those two posts the References header
>>> was missing.
>>>
>>> WTF?
>>>
>>> Has anyone else seen this?
>>
>> Not with slrn 0.9.9. I've checked some of my recent posts and long
>> lines in References are present and correct. If there was a problem in
>> 0.9.8.1pl2 and/or pl2.1, nobody drew it to my attention.
>
> And as I've posted since, the References header would only have
> contained, IIRC, five MIDs.
It doesn't matter how long the References header is, as long as it can
be folded. The problem is probably with the length of an individual
Message-ID in the References header.
<...>
>> The alternative is to stick to the stable 0.9.8.1 release until
>> 0.9.9 is ready for release. The wait is unlikely to be a long one.
>
> I"ll probably revert. I've obviously already done too much
> "upgrading". :)
Or not quite enough. :-)
--
PJR :-)
Blinky the Shark
> In news.software.readers on 27 Nov 2007 17:35:12 GMT, Blinky the Shark
><no....@box.invalid> wrote:
>
>> Peter J Ross wrote:
>>> In news.software.readers on 25 Nov 2007 00:38:52 GMT, Blinky the Shark
>>><no....@box.invalid> wrote:
>>>
>>>> In another group, today I've had a couple reports that one of my posts
>>>> has broken threading. In each of those two posts the References header
>>>> was missing.
>>>>
>>>> WTF?
>>>>
>>>> Has anyone else seen this?
>>>
>>> Not with slrn 0.9.9. I've checked some of my recent posts and long
>>> lines in References are present and correct. If there was a problem in
>>> 0.9.8.1pl2 and/or pl2.1, nobody drew it to my attention.
>>
>> And as I've posted since, the References header would only have
>> contained, IIRC, five MIDs.
>
> It doesn't matter how long the References header is, as long as it can
> be folded. The problem is probably with the length of an individual
> Message-ID in the References header.
I posted all of those upthread.
><...>
>
>>> The alternative is to stick to the stable 0.9.8.1 release until
>>> 0.9.9 is ready for release. The wait is unlikely to be a long one.
>>
>> I"ll probably revert. I've obviously already done too much
>> "upgrading". :)
>
> Or not quite enough. :-)
Having stepped one foot out of the boat, I'm kind of averse to following
with the other. Shark jokes aside.
Shmuel Metz
at 10:41 PM, Blinky the Shark <no....@box.invalid> said:
>I dunno, because I'm not -- for two reasons. 1. It's one post of tens of
>thousands, not "just the way the program works".
In my case it was also not "just the way the program works", despite the
claims of some loonies and conspiracy nuts.
Mike Yetto
> In <slrnfkmiu4....@thurston.blinkynet.net>, on 11/26/2007
> at 10:41 PM, Blinky the Shark <no....@box.invalid> said:
>
>>I dunno, because I'm not -- for two reasons. 1. It's one post of tens of
>>thousands, not "just the way the program works".
>
> In my case it was also not "just the way the program works", despite the
> claims of some loonies and conspiracy nuts.
>
IIRC it *was* the way the broken beta worked.
Mike "or was that a different killer bug?" Yetto
--
Pluralitas non est ponenda sine necessitas.
- William of Ockham
Shmuel Metz
at 08:07 AM, Mike Yetto <mye...@nycap.invalid> said:
>IIRC it *was* the way the broken beta worked.
ObClinton That depends on what "it" was. The way the beta worked was that
in a highly unusual situation[1] the References: header field got lost;
the circumstances weren't the same as the ones affecting slrn, but if
anything they were less common. I suspect that a bypass for the bug in
slrn would be harder to find than a bypass for the bug in MR/2 ICE.
[1] Blinky's description of the problem in slrn applies equally well
to MR/2 ICE: It's one post of tens of thousands, not "just the
way the program works".
--
Blinky the Shark
> In <slrn200711280...@nntp.motzarella.org>, on 11/28/2007
> at 08:07 AM, Mike Yetto <mye...@nycap.invalid> said:
>
>>IIRC it *was* the way the broken beta worked.
>
> ObClinton That depends on what "it" was. The way the beta worked was that
> in a highly unusual situation[1] the References: header field got lost;
> the circumstances weren't the same as the ones affecting slrn, but if
> anything they were less common. I suspect that a bypass for the bug in
> slrn would be harder to find than a bypass for the bug in MR/2 ICE.
>
> [1] Blinky's description of the problem in slrn applies equally well
> to MR/2 ICE: It's one post of tens of thousands, not "just the
> way the program works".
I thought yours was breaking stuff routinely.
Shmuel Metz
at 12:32 AM, Blinky the Shark <no....@box.invalid> said:
>I thought yours was breaking stuff routinely.
c 'thought' 'believed'
A minimal check of google or of your archives would have shown that to be
false. The code was broken for a case rarer than the one that you ran
into. You simply took the word of a few raving lunatics instead of looking
at the data.
The good news is that none of them told *you* that the version of slrn you
used did not exist and that you must have removed the References header
fields yourself.
Thomas Wiegner
> The good news is that none of them told *you* that the version of slrn you
> used did not exist and that you must have removed the References header
> fields yourself.
Could you explain, why the version blinky used does not exist?
Cheers,
Thomas
Blinky the Shark
> In <slrnfks253....@thurston.blinkynet.net>, on 11/29/2007
> at 12:32 AM, Blinky the Shark <no....@box.invalid> said:
>
>>I thought yours was breaking stuff routinely.
>
> c 'thought' 'believed'
>
> A minimal check of google or of your archives would have shown that to be
> false. The code was broken for a case rarer than the one that you ran
> into. You simply took the word of a few raving lunatics instead of looking
> at the data.
Okay, so I remembered wrong. Since that's never happened with you, go
ahead and shoot me.
Shmuel Metz
at 01:03 PM, Blinky the Shark <no....@box.invalid> said:
>Okay, so I remembered wrong. Since that's never happened with you, go
>ahead and shoot me.
It's not the fact of your being wrong that I was commenting on, but the
manner.
Shmuel Metz
at 08:29 PM, Thomas Wiegner <wie...@gmx.de> said:
>Could you explain, why the version blinky used does not exist?
For the same reason the the version I used didn't exist;-)
The claim was not mine, only the ridicule of the claim.