1 (global) history file

1 (global) history file

[Bengt KLEBERG]

----- Begin Included Message -----
 
|Byron mentioned last week that he uses one history file for all his 
|windows; i tried this last week, and now say i think it is the way to go.

 I tried this in the beginning, but it seemed to jumble up and loose
entries from the history file randomly. It could get into situations
where I'd type a command into one window, and then right after it type
a -p, and it wouldn't find the command. Does anyone else use a single
global history under Ultrix and have it work? (Boyd?)

	- cks


----- End Included Message -----

At the very begining I tried to change to several history files. The
coding was simple but I didn't like the result so I switched back to
one file. I've not had any problems with using one global history file
for all my windows (and on all the machines in the network) at the same
time. Maybe I don't write enough commands. We use sunos 4.1.1

Bengt

Re: 1 (global) history file

[John Mackin]

Paul says Byron says global history is good; and he likes it too.

Chris has trouble with it.

Bengt has no trouble with it.

My story: I've _always_ used one history file per "display" (even pre-X,
i.e. V8, which is why the quotes).  I love it, and hate it when some of
the windows are too remote to write to the history.  _But_ -- ever since
I started getting involved with NFS, and _especially_ since moving to
a Sun platform, I have had a _lot_ of trouble with it.  Clearly, the NFS
writes interfere with each other; I haven't read the source to prove this
but I am sure this is what is happening -- they all think they know where
to append, even though they don't really.  I am wondering if opening
the history O_APPEND would help?  Perhaps someone with NFS source handy
and a tad of time could have a quick read and advise us on this.  Even
though it causes me problems, I won't switch back, because it _mostly_
works and I prefer a mostly-working global history to fully-working
per-window history.

OK,
John.

Re: 1 (global) history file

[John Mackin]

As Byron pointed out to me, opening the history O_APPEND isn't going
to improve matters -- since rc already does it.  Sigh.

I really hope someone besides me can be encouraged to have a dig through
some NFS source for ideas about this.  Any takers?

OK,
John.

Re: 1 (global) history file

[Chris Siebenmann]

|I really hope someone besides me can be encouraged to have a dig through
|some NFS source for ideas about this.  Any takers?

 Okay, it's time for some NFS trivia. NFS has no notion of an O_APPEND
write; in fact, NFS doesn't even have a notion of a 'current file
position', being a stateless protocol (har har). NFS writes specify the
file, position in the file, and data to write. O_APPEND is implemented
by looking up the file's current size, and setting the position to
write to to that.

 This means that if two different machines are writing in O_APPEND mode
to the same file, there is no guarantee that both won't try writing to
the same offset, thinking they're appending cleanly. Worse, if two
machines are writing to the same file normally, they can and will stomp
all over each other's data in a merry dance, as the current position is
maintained entirely by the client.

 NFS is basically a crock. Avoid it if possible. Real disks for real
performance and real Unix filesystem semantics.

	- cks, who knows far too much about the insides of the NFS code
	       for his own good.