* hang on entry to group
@ 1999-10-02 19:17 Laura Conrad
1999-10-02 19:22 ` Laura Conrad
1999-10-02 22:05 ` Shenghuo ZHU
0 siblings, 2 replies; 6+ messages in thread
From: Laura Conrad @ 1999-10-02 19:17 UTC (permalink / raw)
pgnus 0.97 is hanging on entry to comp.os.linux.announce at the
moment. The backtrace is:
Signaling: (quit)
accept-process-output(#<network connection "nntpd" ("nntp" . "localhost") state:run> 1)
nntp-accept-process-output(#<network connection "nntpd" ("nntp" . "localhost") state:run>)
nntp-send-command-nodelete("\r?\n\\.\r?\n" "XOVER" "117-117")
nntp-send-xover-command(117 117)
nntp-retrieve-headers-with-xover((117 122 356 357 358 359 360 361 362) nil)
nntp-retrieve-headers((117 122 356 357 358 359 360 361 362) "comp.os.linux.announce" "localhost" nil)
gnus-retrieve-headers((117 122 356 357 358 359 360 361 362) "comp.os.linux.announce" nil)
gnus-cache-retrieve-headers((117 122 356 357 358 359 360 361 362) "comp.os.linux.announce" nil)
gnus-retrieve-headers((117 122 356 357 358 359 360 361 362) "comp.os.linux.announce" nil)
gnus-select-newsgroup("comp.os.linux.announce" nil nil)
gnus-summary-read-group-1("comp.os.linux.announce" nil nil nil nil nil)
gnus-summary-read-group("comp.os.linux.announce" nil nil nil nil nil nil)
gnus-group-read-group(nil)
call-interactively(gnus-group-read-group)
Telnetting to the server, I get:
[lconrad@serpent tmp]$ telnet localhost 119
Trying 127.0.0.1...
Connected to localhost.
Escape character is '^]'.
200 Leafnode NNTP Daemon, version 1.9.4 running at serpent.local
group comp.os.linux.announce
211 278 85 362 comp.os.linux.announce group selected
xover 117-117
420 No articles in specified range.
quit
205 Always happy to serve!
Connection closed by foreign host.
[lconrad@serpent tmp]$
I reported a bug like this before, and what happened was that the
group I was having trouble on continued to have trouble until I
restarted xemacs. (That is, restarting gnus did not help, but
restarting xemacs did.)
--
Laura (mailto:lconrad@world.std.com , http://www.world.std.com/~lconrad/ )
(617) 661-8097 fax: (801) 365-6574
233 Broadway, Cambridge, MA 02139
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: hang on entry to group
1999-10-02 19:17 hang on entry to group Laura Conrad
@ 1999-10-02 19:22 ` Laura Conrad
1999-10-02 22:05 ` Shenghuo ZHU
1 sibling, 0 replies; 6+ messages in thread
From: Laura Conrad @ 1999-10-02 19:22 UTC (permalink / raw)
>>>>> "Laura" == Laura Conrad <lconrad@world.std.com> writes:
Laura> I reported a bug like this before, and what happened was that the
Laura> group I was having trouble on continued to have trouble until I
Laura> restarted xemacs. (That is, restarting gnus did not help, but
Laura> restarting xemacs did.)
However, in this case, restarting gnus did clear the problem, and I
can now get into comp.os.linux.announce fine.
--
Laura (mailto:lconrad@world.std.com , http://www.world.std.com/~lconrad/ )
(617) 661-8097 fax: (801) 365-6574
233 Broadway, Cambridge, MA 02139
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: hang on entry to group
1999-10-02 19:17 hang on entry to group Laura Conrad
1999-10-02 19:22 ` Laura Conrad
@ 1999-10-02 22:05 ` Shenghuo ZHU
1999-10-03 4:15 ` Shenghuo ZHU
1 sibling, 1 reply; 6+ messages in thread
From: Shenghuo ZHU @ 1999-10-02 22:05 UTC (permalink / raw)
[-- Attachment #1: Type: text/plain, Size: 1385 bytes --]
>>>>> "LC" == Laura Conrad <lconrad@world.std.com> writes:
LC> pgnus 0.97 is hanging on entry to comp.os.linux.announce at the
LC> moment. The backtrace is:
LC> Signaling: (quit)
LC> accept-process-output(#<network connection "nntpd" ("nntp" . "localhost") state:run> 1)
LC> nntp-accept-process-output(#<network connection "nntpd" ("nntp" . "localhost") state:run>)
LC> nntp-send-command-nodelete("\r?\n\\.\r?\n" "XOVER" "117-117")
LC> nntp-send-xover-command(117 117)
LC> nntp-retrieve-headers-with-xover((117 122 356 357 358 359 360 361 362) nil)
[...]
I guess I've located the bug. The ChangeLog and patch follow and have
been committed to CVS.
LC> [lconrad@serpent tmp]$ telnet localhost 119
LC> Trying 127.0.0.1...
LC> Connected to localhost.
LC> Escape character is '^]'.
LC> 200 Leafnode NNTP Daemon, version 1.9.4 running at serpent.local
LC> group comp.os.linux.announce
LC> 211 278 85 362 comp.os.linux.announce group selected
LC> xover 117-117
LC> 420 No articles in specified range.
The response code 420 is a bit weird. In nntp-send-xover-command, Gnus
try to find out the XOVER command when nntp-server-xover is try. The
response code 420 cause Gnus disable XOVER, while here XOVER is right
command.
[...]
--
Shenghuo ZHU
1999-10-02 Shenghuo ZHU <zsh@cs.rochester.edu>
* nntp.el (nntp-send-xover-command): Wait for nothing if not
wait-for-reply.
[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: 97-zsh3.diff --]
[-- Type: text/x-patch, Size: 692 bytes --]
Index: nntp.el
===================================================================
RCS file: /usr/local/cvsroot/gnus/lisp/nntp.el,v
retrieving revision 5.20
diff -u -r5.20 nntp.el
--- nntp.el 1999/09/27 15:19:05 5.20
+++ nntp.el 1999/10/02 21:55:07
@@ -1189,7 +1189,7 @@
(nntp-send-command-nodelete
"\r?\n\\.\r?\n" nntp-server-xover range)
;; We do not wait for the reply.
- (nntp-send-command-nodelete "\r?\n\\.\r?\n" nntp-server-xover range))
+ (nntp-send-command-nodelete nil nntp-server-xover range))
(let ((commands nntp-xover-commands))
;; `nntp-xover-commands' is a list of possible XOVER commands.
;; We try them all until we get at positive response.
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: hang on entry to group
1999-10-02 22:05 ` Shenghuo ZHU
@ 1999-10-03 4:15 ` Shenghuo ZHU
1999-10-03 6:27 ` Shenghuo ZHU
0 siblings, 1 reply; 6+ messages in thread
From: Shenghuo ZHU @ 1999-10-03 4:15 UTC (permalink / raw)
>>>>> "ZSH" == Shenghuo ZHU <zsh@cs.rochester.edu> writes:
>>>>> "LC" == Laura Conrad <lconrad@world.std.com> writes:
LC> pgnus 0.97 is hanging on entry to comp.os.linux.announce at the
LC> moment. The backtrace is:
LC> Signaling: (quit)
LC> accept-process-output(#<network connection "nntpd" ("nntp" . "localhost") state:run> 1)
LC> nntp-accept-process-output(#<network connection "nntpd" ("nntp" . "localhost") state:run>)
LC> nntp-send-command-nodelete("?\n\\.?\n" "XOVER" "117-117")
LC> nntp-send-xover-command(117 117)
LC> nntp-retrieve-headers-with-xover((117 122 356 357 358 359 360 361 362) nil)
ZSH> I guess I've located the bug. The ChangeLog and patch follow and
ZSH> have been committed to CVS.
There are still other bugs which cause Gnus hanging.
--
Shenghuo
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: hang on entry to group
1999-10-03 4:15 ` Shenghuo ZHU
@ 1999-10-03 6:27 ` Shenghuo ZHU
1999-10-03 7:07 ` Shenghuo ZHU
0 siblings, 1 reply; 6+ messages in thread
From: Shenghuo ZHU @ 1999-10-03 6:27 UTC (permalink / raw)
[-- Attachment #1: Type: text/plain, Size: 364 bytes --]
>>>>> "ZSH" == Shenghuo ZHU <zsh@cs.rochester.edu> writes:
ZSH> There are still other bugs which cause Gnus hanging.
Fixed.
The ChangeLog and the patch follow and have been committed to CVS.
Apply the previous patch also.
--
Shenghuo ZHU
1999-10-03 Shenghuo ZHU <zsh@cs.rochester.edu>
* nntp.el (nntp-retrieve-headers-with-xover): Fix hanging problem.
[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: 97-zsh4.diff --]
[-- Type: text/x-patch, Size: 3025 bytes --]
Index: nntp.el
===================================================================
RCS file: /usr/local/cvsroot/gnus/lisp/nntp.el,v
retrieving revision 5.21
diff -u -r5.21 nntp.el
--- nntp.el 1999/10/02 21:57:40 5.21
+++ nntp.el 1999/10/03 05:59:13
@@ -1118,9 +1118,10 @@
((numberp nntp-nov-gap)
(let ((count 0)
(received 0)
- (last-point (point-min))
+ last-point
+ in-process-buffer-p
(buf nntp-server-buffer)
- ;;(process-buffer (nntp-find-connection (current-buffer))))
+ (process-buffer (nntp-find-connection-buffer nntp-server-buffer))
first)
;; We have to check `nntp-server-xover'. If it gets set to nil,
;; that means that the server does not understand XOVER, but we
@@ -1133,40 +1134,50 @@
(< (- (nth 1 articles) (car articles)) nntp-nov-gap))
(setq articles (cdr articles)))
- (when (nntp-send-xover-command first (car articles))
- (setq articles (cdr articles)
- count (1+ count))
-
+ (setq in-process-buffer-p (stringp nntp-server-xover))
+ (nntp-send-xover-command first (car articles))
+ (setq articles (cdr articles))
+
+ (when (and nntp-server-xover in-process-buffer-p)
+ ;; Don't count tried request.
+ (setq count (1+ count))
+
;; Every 400 requests we have to read the stream in
;; order to avoid deadlocks.
(when (or (null articles) ;All requests have been sent.
(zerop (% count nntp-maximum-request)))
- (accept-process-output)
+
+ (nntp-accept-response)
;; On some Emacs versions the preceding function has
;; a tendency to change the buffer. Perhaps. It's
;; quite difficult to reproduce, because it only
;; seems to happen once in a blue moon.
- (set-buffer buf)
+ (set-buffer process-buffer)
(while (progn
- (goto-char last-point)
+ (goto-char (or last-point (point-min)))
;; Count replies.
(while (re-search-forward "^[0-9][0-9][0-9] " nil t)
(setq received (1+ received)))
(setq last-point (point))
(< received count))
- (accept-process-output)
- (set-buffer buf)))))
+ (nntp-accept-response)
+ (set-buffer process-buffer))
+ (set-buffer buf))))
(when nntp-server-xover
- ;; Wait for the reply from the final command.
- (goto-char (point-max))
- (re-search-backward "^[0-9][0-9][0-9] " nil t)
- (when (looking-at "^[23]")
- (while (progn
- (goto-char (point-max))
- (forward-line -1)
- (not (looking-at "^\\.\r?\n")))
- (nntp-accept-response)))
+ (when in-process-buffer-p
+ (set-buffer process-buffer)
+ ;; Wait for the reply from the final command.
+ (goto-char (point-max))
+ (re-search-backward "^[0-9][0-9][0-9] " nil t)
+ (when (looking-at "^[23]")
+ (while (progn
+ (goto-char (point-max))
+ (forward-line -1)
+ (not (looking-at "^\\.\r?\n")))
+ (nntp-accept-response)
+ (set-buffer process-buffer)))
+ (set-buffer buf))
;; We remove any "." lines and status lines.
(goto-char (point-min))
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: hang on entry to group
1999-10-03 6:27 ` Shenghuo ZHU
@ 1999-10-03 7:07 ` Shenghuo ZHU
0 siblings, 0 replies; 6+ messages in thread
From: Shenghuo ZHU @ 1999-10-03 7:07 UTC (permalink / raw)
[-- Attachment #1: Type: text/plain, Size: 342 bytes --]
>>>>> "ZSH" == Shenghuo ZHU <zsh@cs.rochester.edu> writes:
ZSH> Fixed.
ZSH> The ChangeLog and the patch follow and have been committed to CVS.
ZSH> Apply the previous patch also.
I am sorry that I made a mistake on the last patch. Now I put all
changes I made on nntp.el since 0.97 in one patch. ChangeLog is
reusable.
--
Shenghuo ZHU
[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: 97-zsh4+.diff --]
[-- Type: text/x-patch, Size: 2355 bytes --]
Index: nntp.el
===================================================================
RCS file: /usr/local/cvsroot/gnus/lisp/nntp.el,v
retrieving revision 5.20
retrieving revision 5.23
diff -r5.20 -r5.23
1121c1121,1122
< (last-point (point-min))
---
> last-point
> in-process-buffer-p
1123c1124
< ;;(process-buffer (nntp-find-connection (current-buffer))))
---
> (process-buffer (nntp-find-connection-buffer nntp-server-buffer))
1136,1139c1137,1144
< (when (nntp-send-xover-command first (car articles))
< (setq articles (cdr articles)
< count (1+ count))
<
---
> (setq in-process-buffer-p (stringp nntp-server-xover))
> (nntp-send-xover-command first (car articles))
> (setq articles (cdr articles))
>
> (when (and nntp-server-xover in-process-buffer-p)
> ;; Don't count tried request.
> (setq count (1+ count))
>
1144c1149,1150
< (accept-process-output)
---
>
> (nntp-accept-response)
1149c1155
< (set-buffer buf)
---
> (set-buffer process-buffer)
1151c1157
< (goto-char last-point)
---
> (goto-char (or last-point (point-min)))
1157,1158c1163,1165
< (accept-process-output)
< (set-buffer buf)))))
---
> (nntp-accept-response)
> (set-buffer process-buffer))
> (set-buffer buf))))
1161,1169c1168,1185
< ;; Wait for the reply from the final command.
< (goto-char (point-max))
< (re-search-backward "^[0-9][0-9][0-9] " nil t)
< (when (looking-at "^[23]")
< (while (progn
< (goto-char (point-max))
< (forward-line -1)
< (not (looking-at "^\\.\r?\n")))
< (nntp-accept-response)))
---
> (when in-process-buffer-p
> (set-buffer process-buffer)
> ;; Wait for the reply from the final command.
> (goto-char (point-max))
> (re-search-backward "^[0-9][0-9][0-9] " nil t)
> (when (looking-at "^[23]")
> (while (progn
> (goto-char (point-max))
> (forward-line -1)
> (not (looking-at "^\\.\r?\n")))
> (nntp-accept-response)
> (set-buffer process-buffer)))
> (set-buffer buf)
> (goto-char (point-max))
> (insert-buffer-substring process-buffer)
> (set-buffer process-buffer)
> (erase-buffer)
> (set-buffer buf))
1192c1208
< (nntp-send-command-nodelete "\r?\n\\.\r?\n" nntp-server-xover range))
---
> (nntp-send-command-nodelete nil nntp-server-xover range))
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~1999-10-03 7:07 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
1999-10-02 19:17 hang on entry to group Laura Conrad
1999-10-02 19:22 ` Laura Conrad
1999-10-02 22:05 ` Shenghuo ZHU
1999-10-03 4:15 ` Shenghuo ZHU
1999-10-03 6:27 ` Shenghuo ZHU
1999-10-03 7:07 ` Shenghuo ZHU
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).