[gnus.gnus-bug] Wrong article count

[gnus.gnus-bug] Wrong article count

[Ted Zlatanov]

[-- Attachment #1: Type: text/plain, Size: 131 bytes --]

Has anyone looked at this bug report?  I don't know the NNTP protocol
well, but I can examine the report if no one else can.

Ted


[-- Attachment #2: Type: message/rfc822, Size: 135337 bytes --]

[-- Attachment #2.1.1: Type: text/plain, Size: 1926 bytes --]

A newsgroup I subscribe to, comp.compilers, has the wrong article
count. I never see any new News unless I enter the group and ask for
all articles.

If I ask for the latest 200 articles, i.e. by C-u 1 0 0 RET on the
group in the Group Buffer, I get nothing.

Gnus seems to think there are 10651 articles in the group, but if I
telnet to the server it says:

  group comp.compilers
  211 411 8841 9514 comp.compilers

I tried to kill the group and subscribe again, but that didn't help. 

I tried to find all mentions of the group in .newsrc.eld and other
files and remove them. Then I restarted Emacs and Gnus, but that
didn't help. 

Gnus still thinks there are 10651 articles in the group and gives me
nothing when I ask for, say, the last 200.

Any ideas?

To generate this report, I restarted Emacs, then started Gnus with
nntp+lysator:comp.compilers killed, changed nntp-record-commands to t
and set gnus-verbose to 10. Then I subscribed to group and did C-u 1 0
0 RET to get the last 100 articles. Nothing seemed to happen.

This is some of the  *Messages* buffer:

Denied server
Opening nntp server on axis...failed
Opening nntp server on axis...
Denied server
Opening nntp server on axis...failed
Opening nntp server on lysator...done
Checking new news...done
Loading gnus-topic...done
t
10
Woof! Woof!
Retrieving newsgroup: nntp+lysator:comp.compilers...
Loading gnus-ml...done
Loading parse-time...done
Fetching headers for nntp+lysator:comp.compilers...
NNTP: Receiving headers...done
Fetching headers for nntp+lysator:comp.compilers...done
No articles in the group
Loading gnus-async...done
No unread news
Auto-saving...
Mark set

Attached is the NNTP log and the .newsrc.eld.

Here's what M-x gnus-bug RET generated:

No Gnus v0.6
GNU Emacs 21.3.1 (i386--freebsd, X toolkit, Xaw3d scroll bars)
 of 2005-04-04 on gohan30.freebsd.org
200 news.lysator.liu.se InterNetNews NNRP server INN 2.4.3 ready (posting ok).


[-- Attachment #2.1.2: User settings --]
[-- Type: application/emacs-lisp, Size: 3510 bytes --]

[-- Attachment #2.1.3: NNTP log --]
[-- Type: text/plain, Size: 4302 bytes --]

20070209T143508.315 localhost GROUP comp.compilers
20070209T143515.641 localhost GROUP comp.compilers
20070209T143515.694 localhost XOVER 19392-19491
20070209T143515.713 localhost XOVERVIEW 19392-19491
20070209T143515.730 localhost HEAD 19392
20070209T143515.730 localhost HEAD 19393
20070209T143515.730 localhost HEAD 19394
20070209T143515.730 localhost HEAD 19395
20070209T143515.731 localhost HEAD 19396
20070209T143515.731 localhost HEAD 19397
20070209T143515.731 localhost HEAD 19398
20070209T143515.731 localhost HEAD 19399
20070209T143515.731 localhost HEAD 19400
20070209T143515.731 localhost HEAD 19401
20070209T143515.731 localhost HEAD 19402
20070209T143515.731 localhost HEAD 19403
20070209T143515.731 localhost HEAD 19404
20070209T143515.732 localhost HEAD 19405
20070209T143515.732 localhost HEAD 19406
20070209T143515.732 localhost HEAD 19407
20070209T143515.732 localhost HEAD 19408
20070209T143515.732 localhost HEAD 19409
20070209T143515.732 localhost HEAD 19410
20070209T143515.732 localhost HEAD 19411
20070209T143515.732 localhost HEAD 19412
20070209T143515.732 localhost HEAD 19413
20070209T143515.733 localhost HEAD 19414
20070209T143515.733 localhost HEAD 19415
20070209T143515.733 localhost HEAD 19416
20070209T143515.733 localhost HEAD 19417
20070209T143515.733 localhost HEAD 19418
20070209T143515.733 localhost HEAD 19419
20070209T143515.733 localhost HEAD 19420
20070209T143515.733 localhost HEAD 19421
20070209T143515.733 localhost HEAD 19422
20070209T143515.734 localhost HEAD 19423
20070209T143515.734 localhost HEAD 19424
20070209T143515.734 localhost HEAD 19425
20070209T143515.734 localhost HEAD 19426
20070209T143515.734 localhost HEAD 19427
20070209T143515.734 localhost HEAD 19428
20070209T143515.734 localhost HEAD 19429
20070209T143515.734 localhost HEAD 19430
20070209T143515.734 localhost HEAD 19431
20070209T143515.735 localhost HEAD 19432
20070209T143515.735 localhost HEAD 19433
20070209T143515.735 localhost HEAD 19434
20070209T143515.735 localhost HEAD 19435
20070209T143515.736 localhost HEAD 19436
20070209T143515.736 localhost HEAD 19437
20070209T143515.736 localhost HEAD 19438
20070209T143515.736 localhost HEAD 19439
20070209T143515.736 localhost HEAD 19440
20070209T143515.736 localhost HEAD 19441
20070209T143515.736 localhost HEAD 19442
20070209T143515.736 localhost HEAD 19443
20070209T143515.736 localhost HEAD 19444
20070209T143515.737 localhost HEAD 19445
20070209T143515.737 localhost HEAD 19446
20070209T143515.737 localhost HEAD 19447
20070209T143515.737 localhost HEAD 19448
20070209T143515.737 localhost HEAD 19449
20070209T143515.737 localhost HEAD 19450
20070209T143515.737 localhost HEAD 19451
20070209T143515.737 localhost HEAD 19452
20070209T143515.737 localhost HEAD 19453
20070209T143515.738 localhost HEAD 19454
20070209T143515.738 localhost HEAD 19455
20070209T143515.738 localhost HEAD 19456
20070209T143515.738 localhost HEAD 19457
20070209T143515.738 localhost HEAD 19458
20070209T143515.738 localhost HEAD 19459
20070209T143515.738 localhost HEAD 19460
20070209T143515.738 localhost HEAD 19461
20070209T143515.738 localhost HEAD 19462
20070209T143515.739 localhost HEAD 19463
20070209T143515.739 localhost HEAD 19464
20070209T143515.739 localhost HEAD 19465
20070209T143515.739 localhost HEAD 19466
20070209T143515.739 localhost HEAD 19467
20070209T143515.739 localhost HEAD 19468
20070209T143515.739 localhost HEAD 19469
20070209T143515.739 localhost HEAD 19470
20070209T143515.739 localhost HEAD 19471
20070209T143515.739 localhost HEAD 19472
20070209T143515.740 localhost HEAD 19473
20070209T143515.740 localhost HEAD 19474
20070209T143515.740 localhost HEAD 19475
20070209T143515.740 localhost HEAD 19476
20070209T143515.740 localhost HEAD 19477
20070209T143515.740 localhost HEAD 19478
20070209T143515.740 localhost HEAD 19479
20070209T143515.740 localhost HEAD 19480
20070209T143515.740 localhost HEAD 19481
20070209T143515.741 localhost HEAD 19482
20070209T143515.741 localhost HEAD 19483
20070209T143515.741 localhost HEAD 19484
20070209T143515.741 localhost HEAD 19485
20070209T143515.741 localhost HEAD 19486
20070209T143515.741 localhost HEAD 19487
20070209T143515.741 localhost HEAD 19488
20070209T143515.741 localhost HEAD 19489
20070209T143515.742 localhost HEAD 19490
20070209T143515.742 localhost HEAD 19491

[-- Attachment #2.1.4: .newsrc.eld --]
[-- Type: application/octet-stream, Size: 92472 bytes --]

Re: [gnus.gnus-bug] Wrong article count

[Andreas Seltenreich]

Ted Zlatanov writes:

> Has anyone looked at this bug report?  I don't know the NNTP protocol
> well, but I can examine the report if no one else can.

I wonder if the symptoms are caused by invalid numbers in the agent
cache.  gnus-agent-possibly-alter-active seems to prefer data from the
agent, and AFAICT bogus data would survive the killing of the group or
gnus-group-clear-data.

Michael: If you are using the agent's caching feature with this method
(it is enabled by default), do the symptoms still occur when you
disable gnus-agent-cache or remove the method from the agent in the
server buffer (J r)?

regards,
andreas

Re: [gnus.gnus-bug] Wrong article count

[Michael Widerkrantz]

Andreas Seltenreich <andreas+ding@gate450.dyndns.org> writes:

> Michael: If you are using the agent's caching feature with this
> method (it is enabled by default), do the symptoms still occur when
> you disable gnus-agent-cache or remove the method from the agent in
> the server buffer (J r)?

For some reason that I can't account for, both the newsgroups with
this problem have the right article count now. I can't explain what I
did. Sorry. If it ever shows up again, I'll keep this in mind.