Bug#134742: Spying at http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=134742
Peter T. Breuer:
> oboe:/usr/oboe/ptb% elm
> I can't understand keyword "local-fast-lookup" in line 18 in ".elm/elmrc"
> file
> Fix .elm/elmrc or let elm rebuild elmrc with option '-w'
>
> followed by bailout.
>
> So the workaround is untenable on a system with NFS shared home
> directories, which people may be accessing from different computers,
> with different versions of elm.
Well, Is different version of Elm different _global_ configuration
directory ("library directory")? There is also elm.rc file.
elm -
gives location.
/ Kari Hurtta
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#134742: Spying at http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=134742
Peter T. Breuer: > > That option have following description: > > > > # If set indicates that directory listing not needed for lookup > > ?? Double dutch! Lookup of what? What directory listing? > > > # Need to be unset, if same character on local-fs-charset is several > > # repesentations (or encodings) > > Again, this is not english. What? "is several representations"? Do they > mean "HAS several representations"? > > > # > > # See also: local-fs-charset, imap-fast-lookup > > ### local-fast-lookup = OFF > > The comments are incomprehensible, I am afraid, which is a bug in > itself. Sorry, my english is not very good. / Kari Hurtta -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#134742: Spying at http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=134742
"A month of sundays ago Kari Hurtta wrote:"
> Peter T. Breuer:
> > oboe:/usr/oboe/ptb% elm
> > I can't understand keyword "local-fast-lookup" in line 18 in ".elm/elmrc"
> > file
> > Fix .elm/elmrc or let elm rebuild elmrc with option '-w'
> >
> > followed by bailout.
> >
> > So the workaround is untenable on a system with NFS shared home
> > directories, which people may be accessing from different computers,
> > with different versions of elm.
>
> Well, Is different version of Elm different _global_ configuration
> directory ("library directory")? There is also elm.rc file.
Oh, I see, you suggest making the elm configuration machine specific by
using a machine specific defaults file! That's not a bad idea, but
whether it's a good or a bad idea, it's in any case a workaround for a
problem that is in itself a result of a workaround for a different
problem - namely that the default behaviour has changed: one now
needs a special new option in order to get the old (and appropriate)
behaviour on any system with an automounter. If not, then new
elm will cause unrelated cds and floppies to try and mount themselves,
causing elm itself to block while each tiems out.
IMO it's simply a programming error, and the simplest thing is to
correct it. Are you saying that this behaviour is _intentional_? What is
its objective?
Peter
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#134742: Spying at http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=134742
Peter T. Breuer: > IMO it's simply a programming error, and the simplest thing is to > correct it. Are you saying that this behaviour is _intentional_? What is > its objective? Doing directory lookup is _intentional_. Statting perhaps can be delayed until someone asks flags for entry from Elm's internal directory browser. Default value of local-fast-lookup can perhaps made function of local-fast-lookup, but that is little more complicated. (ie. US-ASCII, and quite likely all ISO-8859-x are OK for local-fast-lookup=YES. However default value is SYSTEM. On other with UTF-8 definately local-fast-lookup=NO is needed if we not know that unicode strings are canonified everywhere. ) Kari Hurtta -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#134742: Spying at http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=134742
"A month of sundays ago Kari Hurtta wrote:" > Peter T. Breuer: > > IMO it's simply a programming error, and the simplest thing is to > > correct it. Are you saying that this behaviour is _intentional_? What is > > its objective? > > Doing directory lookup is _intentional_. Then what is its intention, if I may ask? The english text about it is indecipherable to my eyes, and I speak english well, and am technically competent. > Statting perhaps can be delayed until someone asks flags > for entry from Elm's internal directory browser. Flags for entry? Asks FOR flags? I'm afraid I don't follow .. what is this about flags? And what is an internal directory browser .. elm doesn't browse, does it? It shows you all mailboxes if you type an asterisk, but I've never seen anything that could remotely be called a browser! Yes, statting is precisely the problem. Or more accurately, statting all the entries BENEATH all the directories that form initial segments of the path to the system mailbox is the problem. That will eventually stat things like /mnt and /cdrom, for being offshoots of the / initial segment of /var/spool/mail. And here it also somehow gets to stat the links in /usr/foo that are links into the NFS automounter's area. > Default value of local-fast-lookup can perhaps made function of > local-fast-lookup, but that is little more complicated. Eh? X can perhaps be made a function of X? You must have meant something different. > (ie. US-ASCII, and quite likely all ISO-8859-x are OK for > local-fast-lookup=YES. However default value is SYSTEM. > On other with UTF-8 definately local-fast-lookup=NO is needed > if we not know that unicode strings are canonified everywhere. ) Can someone give me a clue as to WHY elm is now doing stats on everything below every initial segment of the path to the system mailbox? Or whatever it is doing? It seems from your remarks to have something to do with unicode characters, but I fail to see any logical connection whatsoever! Peter -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#134742: Spying at http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=134742
Peter T. Breuer:
> Yes, statting is precisely the problem. Or more accurately, statting
> all the entries BENEATH all the directories that form initial segments
> of the path to the system mailbox is the problem.
You claim that.
elm -f
does
new_browser(..) [see src/init.c]
but without -f option it does
enter_new_folder()
In other words I claim that it is not "the path to the system mailbox".
_Leaving_ of folder does[src/leavembox.c]
struct folder_browser * recv_browser = new_browser(selection_folder);
struct string * recv_name= new_string(system_charset);
add_ascii_to_string(recv_name,s2us(">"));/* Received folder */
can_store = select_dir_item(recv_browser,&recv_name);
/ Kari Hurtta
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#134742: Spying at http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=134742
Peter T. Breuer: > > Statting perhaps can be delayed until someone asks flags > > for entry from Elm's internal directory browser. > > Flags for entry? Asks FOR flags? I'm afraid I don't follow .. what is > this about flags? And what is an internal directory browser .. elm > doesn't browse, does it? It shows you all mailboxes if you type an > asterisk, but I've never seen anything that could remotely be called a > browser! That it is. You can browse files and imap mailboxes -- that includes imap mailbox hierarcy and local directory structure. You can also use TABulator character -- ie write directory name or [EMAIL PROTECTED] and press TABulator. That is on chapter "Folder and file browser" on file README.ME+ (That file is on top level directory on Elm ME+ distribution. Also http://www.ozone.fmi.fi/KEH/elm-2.4ME+.README ) I do not quote chapter, you then probably just complain my english .-) / Kari Hurtta -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#134742: Spying at http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=134742
"A month of sundays ago Kari E. Hurtta wrote:"
> Peter T. Breuer:
> > Yes, statting is precisely the problem. Or more accurately, statting
> > all the entries BENEATH all the directories that form initial segments
> > of the path to the system mailbox is the problem.
>
> You claim that.
>
> elm -f
I have never used the -f option. All my trials have been with plain
"elm" as the commandline call. I don't know why you claim what you claim
I claim above! ;)
> does
> new_browser(..) [see src/init.c]
I have no idea. I think that you will have to _describe_ what it does
in terms more universally communicative than function names. I don't
see why the name "new_browser" should evoke in my mind the concept of
statting every directory BENEATH every directory in the path to
somewhere!
Perhaps I am not getting through to you on this .. I don't care that it
stats things. What I care about is that it stats _irrelevant_ things,
or perhaps stats them at irrelevant moments, because that triggers
irrelevant system events, such as attempted mounts of cdroms and
floppys, blocking both elm and various other aspects of the system (a
mount attempt will take and hold the unique /etc/mtab~ system lockfile
for the duration of the attempt, for example, which is 30s).
Anyway, have we confirmed this? Certainly my strace showed that
originally, and your suggested config file option has cured the
apparently similar behaviour in the newest elm. Does that support the
hypothesis? Remember that my original tests were carried out with an
elm about number -81, if I recall correctly. the most recent one
appears to be -95, and it quite possibly has an entirely different set
of behaviours, as far as I know. I would be grateful if you confined
your analysis to the version that the report was about, as we will make
progress faster that way.
> but without -f option it does
>
> enter_new_folder()
>
> In other words I claim that it is not "the path to the system mailbox".
I have no idea what it is that it is taking the initial segments of the
path of. Does it matter? It seemed from the original straces made to be
the path to $MAIL. Isn't that the "system mailbox"? If I am using the
incorrect term for it, I beg your pardon, but it's not very important
to me what it happens to be termed.
> _Leaving_ of folder does[src/leavembox.c]
>
> struct folder_browser * recv_browser = new_browser(selection_folder);
> struct string * recv_name= new_string(system_charset);
>
> add_ascii_to_string(recv_name,s2us(">"));/* Received folder */
> can_store = select_dir_item(recv_browser,&recv_name);
Is this supposed to mean something? I can assure you that as an
explanation, it fails completely even to communicate itself! Are you
saying that there is in fact a search for some file containing charset
info, and that THAT is what is causing the problem?
If you are saying that, I suggest you "say it". It will make
communication a lot easier. Both ways.
Peter
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#134742: Spying at http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=134742
"A month of sundays ago Kari E. Hurtta wrote:" > Peter T. Breuer: > > Flags for entry? Asks FOR flags? I'm afraid I don't follow .. what is > > this about flags? And what is an internal directory browser .. elm > > doesn't browse, does it? It shows you all mailboxes if you type an > > asterisk, but I've never seen anything that could remotely be called a > > browser! > > That it is. It is a browser? > You can browse files and imap mailboxes -- that includes imap mailbox hierarcy > and local directory structure. Is that the "asterisk" that i referred to above, or something else? > You can also use TABulator character -- ie write directory name > or [EMAIL PROTECTED] and press TABulator. Interesting. I didn't know it now had imap support. Pop too? In any case, I am not interested in new features, useful though they may be to other people! I simply want it to continue working as well as it has always worked ... > That is on chapter "Folder and file browser" on file README.ME+ Never read it, I'm afraid. I don't use any language features. I strip everything to straight ascii, and mail in 8bit. As a matter of fact, I would like everything to be treated as 8bit and that's that, but sometimes people send me things with weird charsets in, and then I have to add another charset to the list of us-ascii equivs in the config file! Otherwise I can't edit a reply, because elm gets rid of the contents, claiming it can't understand them. > (That file is on top level directory on Elm ME+ distribution. Quite possibly! > Also http://www.ozone.fmi.fi/KEH/elm-2.4ME+.README > ) > > I do not quote chapter, you then probably just complain my english .-) Well, if it's in the documentation, I should probably enter a bug report :-). I'm quite willing to translate, provided I can find out what it was intended to mean in the first place. I however have been unsuccessful so far in getting elm to compile with only dotlocking, it appears. At any rate, this version (95) is currently hanging and counting up to 7 trying to lock my mail in an NFS directory mount: Timed out on locking mailbox. Leaving program. You have new mail in /usr/spool/mail/ptb sh-2.03$ ls -ld /var/spool/mail drwxrwxrwt6 mail mail 164864 Jun 5 16:19 /var/spool/mail sh-2.03$ ls -ld /usr/bin/elm -rwxr-sr-x1 root mail 396032 Jun 4 17:04 /usr/bin/elm and I don't see any reason why. This looks like it's not doing dotlocking. I've mounted the directory nolock, but no change: arpa:/var/mail on /var/spool/mail type nfs (rw,soft,intr,rsize=4096,wsize=4096,nolock,addr=163.117.139.31) What I did was go into the Configure file and add has_flock= has_fcntl= which as far as I can see, should do the trick. What's wrong with using an ordinary autoconf? How am I supposed to set particular options? Peter -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#134742: Spying at http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=134742
"A month of sundays ago Kari E. Hurtta wrote:"
> That is on chapter "Folder and file browser" on file README.ME+
Ecch .. I've just looked at that file. If it has chapters, that's
news. It looks like a mess and nothing more.
OK! I've now compiled elm with JUST DOTLOCKING, yay, and it no longer
hangs on an nfs mounted spool. There probably was a system-way out
of that, but I was too impatient to spend time thinking of it (what
could it be that nolock doesn't cure in the mount optioons?)
The trick appeared to be to edit the config.sh file by hand, and
setting to ?ndef' the flock option.
Then make wouln't let me cun again without a Config -S. The message
was incomprehensible, I may add, and the blurb for the meaning
of the ards to Config is also incomprehensible, grumble, grumble,...
Now, to show you that it is the $MAIL variable of which it is statting the
entries BELOW the initial segments of the path, I have set
MAIL=/usr/spool/mail/ptb
and straced elm. The result includes ...
access("/usr/", R_OK) = 0
open("/usr/", O_RDONLY|O_NONBLOCK|0x1) = 9
fstat(9, {st_mode=S_IFDIR|0755, st_size=1024, ...}) = 0
fcntl(9, F_SETFD, FD_CLOEXEC) = 0
getdents(9, /* 31 entries */, 3933) = 544
stat("/usr/lost+found", {st_mode=S_IFDIR|0755, st_size=12288, ...}) = 0
stat("/usr/share", {st_mode=S_IFDIR|0755, st_size=3072, ...}) = 0
stat("/usr/local", {st_mode=S_IFDIR|0755, st_size=1024, ...}) = 0
stat("/usr/bin", {st_mode=S_IFDIR|0755, st_size=40960, ...}) = 0
stat("/usr/doc", {st_mode=S_IFDIR|0755, st_size=21504, ...}) = 0
stat("/usr/lm", 0xb2dc) = -1 ENOENT (No such file or
directory)
stat("/usr/info", {st_mode=S_IFDIR|0755, st_size=10240, ...}) = 0
stat("/usr/man", {st_mode=S_IFDIR|0755, st_size=1024, ...}) = 0
stat("/usr/sbin", {st_mode=S_IFDIR|0755, st_size=7168, ...}) = 0
stat("/usr/X11R6", {st_mode=S_IFDIR|0755, st_size=1024, ...}) = 0
stat("/usr/batuta", {st_mode=S_IFDIR|0755, st_size=1024, ...}) = 0
stat("/usr/spool", {st_mode=S_IFDIR|0755, st_size=1024, ...}) = 0
stat("/usr/games", {st_mode=S_IFDIR|0755, st_size=1024, ...}) = 0
stat("/usr/lab", {st_mode=S_IFDIR|0755, st_size=2048, ...}) = 0
stat("/usr/oboe", {st_mode=S_IFDIR|0755, st_size=1024, ...}) = 0
stat("/usr/include", {st_mode=S_IFDIR|0755, st_size=9216, ...}) = 0
stat("/usr/lib", {st_mode=S_IFDIR|0755, st_size=38912, ...}) = 0
stat("/usr/openwin", {st_mode=S_IFDIR|0755, st_size=1024, ...}) = 0
stat("/usr/src", {st_mode=S_IFDIR|S_ISGID|0755, st_size=1024, ...}) = 0
stat("/usr/ssl", {st_mode=S_IFDIR|0755, st_size=1024, ...}) = 0
stat("/usr/arpa", {st_mode=S_IFDIR|0755, st_size=1024, ...}) = 0
stat("/usr/bajo", {st_mode=S_IFDIR|0755, st_size=1024, ...}) = 0
stat("/usr/karajan", 0xb2dc)= -1 ENOENT (No such file or
directory)
stat("/usr/cuerno", {st_mode=S_IFDIR|0755, st_size=4096, ...}) = 0
stat("/usr/dist", {st_mode=S_IFDIR|0755, st_size=1024, ...}) = 0
stat("/usr/guitarra",
which is running through a set of links to the autofs NFS mounts. This
happens on quit from elm, but I think it used to do it on entry as well
in earlier versions.
As it happens, MAIL is normally set to /var/spool/mail here, which avoids
the issue for me, but on some systems here there are both autofs mounts
below /var, and/or the system mailboxes are in /usr/spool (via links),
so the problem got triggered there spectacularly.
Peter
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#149131: clisp accepts, as accessor to object, name that conflicts with build-in function
Package: clisp Version: 1:2.27-0.5 Severity: wishlist I wrote in clisp program that defines class with accessor to one of its slots called type. I tried to run this program under Harlequin Lisp and Alegro. Both reported an error, Harlequin said that type conflicts with function visible from COMMON-LISP. This seems to me that clisp dont check accessor names with build-in functions. -- System Information Debian Release: testing/unstable Architecture: i386 Kernel: Linux pisidlo 2.4.18 #1 Tue Jun 4 12:08:32 CEST 2002 i686 Locale: LANG=cs_CZ, LC_CTYPE=cs_CZ Versions of packages clisp depends on: ii libc62.2.5-4 GNU C Library: Shared libraries an ii libncurses5 5.2.20020112a-7 Shared libraries for terminal hand -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
