🚀 go-pugleaf

the -f flag in freeBSD shows information about swapped out processes, but is
only honored if you're root.  GNU ps (procps, for linux) uses option f to
display the process tree (which processes are whose children).  BSD ps
(!= freeBSD ps) does not accept flag f.  someone probably recompiled and
replaced ps with a regular BSD ps version, which was not modified for freeBSD.

ari

Kim DeVaughn <kimdv@best.com> wrote:
> FYI ...

> Looks like there has been a recent (undocumented) change to the ps(1)
> utility.

> Using the -f flag now gives an "illegal option" error, even though it
> is still listed as a valid option in the man page.

> I've never been quite sure as to what -f is supposed to do ("work harder",
> per the man page), but since it implies more accurate numbers, I've used
> it in my various script wrappers.  Until today.

> I guess the change was made during the last "make world" of a couple days
> ago.

> /kim (who wonders if this is a local hack, since the man page hasn't changed)

> ================================================================
> "Never believe in anything until it has been officially denied."
>                                --Otto von Bismarck

ari <myo@shell6.ba.best.com> wrote:
| the -f flag in freeBSD shows information about swapped out processes, but is
| only honored if you're root.  GNU ps (procps, for linux) uses option f to
| display the process tree (which processes are whose children).  BSD ps
| (!= freeBSD ps) does not accept flag f.  someone probably recompiled and
| replaced ps with a regular BSD ps version which was not modified for freeBSD.
| 
| ari
| 
| Kim DeVaughn <kimdv@best.com> wrote:
| > FYI ...
| 
| > Looks like there has been a recent (undocumented) change to the ps(1)
| > utility.
| 
| > Using the -f flag now gives an "illegal option" error, even though it
| > is still listed as a valid option in the man page.
| 
| > I've never been quite sure as to what -f is supposed to do ("work harder",
| > per the man page), but since it implies more accurate numbers, I've used
| > it in my various script wrappers.  Until today.
| 
| > I guess the change was made during the last "make world" of a couple days
| > ago.
| 
| > /kim (who wonders if this is a local hack, since the man page hasn't changed)

Gee, and just last week we were discussing source-based local site hacks that
prevented programs and procedures from operating on all systems with the "same"
OS and I was brought to task for implying it could happen here (at Best).

:-)

Thad

In article <36520add$0$12779@nntp1.ba.best.com>,
Thad Floryan <thad@thadlabs.com> wrote:
>ari <myo@shell6.ba.best.com> wrote:
>| the -f flag in freeBSD shows information about swapped out processes, but is
>| only honored if you're root.
>Gee, and just last week we were discussing source-based local site
>hacks that prevented programs and procedures from operating on all
>systems with the "same" OS and I was brought to task for implying
>it could happen here (at Best).

It's never worked... you had to be root.  Instead of failing silently,
it probably started to display an error in a recent OS revision.

In article <3651d85a$0$12753@nntp1.ba.best.com>,
	ari <myo@shell6.ba.best.com> wrote:
|
| Kim DeVaughn <kimdv@best.com> wrote:
| >
| > Looks like there has been a recent (undocumented) change to the ps(1)
| > utility.
| >
| > Using the -f flag now gives an "illegal option" error, even though it
| > is still listed as a valid option in the man page.
|
| the -f flag in freeBSD shows information about swapped out processes, but is
| only honored if you're root.  GNU ps (procps, for linux) uses option f to
| display the process tree (which processes are whose children).  BSD ps
| (!= freeBSD ps) does not accept flag f.  someone probably recompiled and
| replaced ps with a regular BSD ps version, which was not modified for freeBSD

Thanks for the info ... I certainly would not have guessed the actual
function from the man page description on a FreeBSD system.  To wit:

 >     -f      Work harder to get detailed process information.

No mention of "root only" options/information anywhere in the page,
either.

Phffft ...

/kim

===============================================================================
"If you can't dazzle 'em with brilliance, baffle 'em with bullshit."  --Unknown

Kim DeVaughn <kimdv@best.com> wrote:
> 
> Thanks for the info ... I certainly would not have guessed the actual
> function from the man page description on a FreeBSD system.  To wit:
> 
>  >     -f      Work harder to get detailed process information.
> 
> No mention of "root only" options/information anywhere in the page,
> either.

Hey, you're more than welcome to submit man page changes to the FreeBSD
community instead of complaining ...

> 
> Phffft ...
> 
> /kim
> 
> ===============================================================================
> "If you can't dazzle 'em with brilliance, baffle 'em with bullshit."  --Unknown


-- dima

In article <36529534$0$12756@nntp1.ba.best.com>,
	Dima Ruban <dima@best.net> wrote:
|
| Kim DeVaughn <kimdv@best.com> wrote:
| >
| > Thanks for the info ... I certainly would not have guessed the actual
| > function from the man page description on a FreeBSD system.  To wit:
| >
| >  >     -f      Work harder to get detailed process information.
| >
| > No mention of "root only" options/information anywhere in the page,
| > either.
|
| Hey, you're more than welcome to submit man page changes to the FreeBSD
| community instead of complaining ...

Point was ... I didn't know that the man page even *needed* changing, since
I actually believed what it said (about -f).  I know ... silly me ...!

Now that "someone" has recently seen fit to change the ps code, and remove
that option, I would *think* they should be the one responsible for changing
the man page documentation to reflect reality.  I know ... silly me ...!

But since you asked ... here ...

diff -uNr /usr.orig/share/man/man1/ps.1 /usr/share/man/man1/ps.1
--- /usr.orig/share/man/man1/ps.1	Fri Nov 13 04:52:34 1998
+++ /usr/share/man/man1/ps.1	Wed Nov 18 01:46:50 1998
@@ -88,8 +88,6 @@
 no effect).
 .It Fl e
 Display the environment as well.
-.It Fl f
-Work harder to get detailed process information.
 .It Fl h
 Repeat the information header as often as necessary to guarantee one
 header per page of information.

Of course I have no idea as to whether any other changes are needed for that
particular utility, and I don't feel an overwhelming urge to walk thru the
code.  Silly me ... :-) ...

/kim

======================================
"I hate prototypes!"   --Chief O'Brien

In article <36529e4e$0$12764@nntp1.ba.best.com>,
Kim DeVaughn <kimdv@best.com> wrote:
>Point was ... I didn't know that the man page even *needed* changing, since
>I actually believed what it said (about -f).  I know ... silly me ...!

And what did you believe that "work harder" meant?  Sounded like
nonsense to me....

>Now that "someone" has recently seen fit to change the ps code,
>and remove that option, I would *think* they should be the one
>responsible for changing the man page documentation to reflect
>reality.

I can't see the source on shellN, and I am not running the most
recent sources here, but I imagine the option is still there.  It
just gives an error instead of failing silently when called as
non-root.

In article <3651c43e$0$12775@nntp1.ba.best.com>, I said:
|
| Looks like there has been a recent (undocumented) change to the ps(1)
| utility.
|
| Using the -f flag now gives an "illegal option" error, even though it
| is still listed as a valid option in the man page.

Hmmm.  Strange things have been happening behind that curtain, Dorothy ...

With lastnight's 1462 (binary) file update, it seems that the old behavior
pf ps(1) is back ... ie, the -f flag is once again valid (or at least it
is not rejected as "illegal").

Whether it actually *does* anything (given what else has been said in this
thread), is anyone's guess.

What is really strange is that:

 1 - the f-flag doesn't show up in the usage blurb, to wit:

     usage: ps [-aChjlmrSTuvwx] [-O|o fmt] [-p pid] [-t tty] [-U user]
               [-M core] [-N system] [-W swap]
            ps [-L]
and

 2 - the filesize *and* timestamp on the current binary are exactly the same
     as they were for the 604 (binary) file update that was done last Friday
     (the 13th), when this whole business seems to have started.

Not that this -f flag stuff for ps is all that important ... but it *does*
make you wonder what *other* gotcha's are lurking out there ...!

As may be ... I don't think we're in Kansas anymore, Toto ...

/kim

===============================================================
"I think I speak for everyone here when I say, 'Huh?'"  --Buffy

In article <72uq49$84$1@machaut.medieval.org>,
	Todd Michel McComb <mccomb@medieval.org> wrote:
|
| In article <36529e4e$0$12764@nntp1.ba.best.com>,
| 	  Kim DeVaughn <kimdv@best.com> wrote:
| >
| >Point was ... I didn't know that the man page even *needed* changing, since
| >I actually believed what it said (about -f).  I know ... silly me ...!
|
| And what did you believe that "work harder" meant?  Sounded like
| nonsense to me....

Oh, I don't know ... just some hand-waving shorthand for something like
"roots around more thoroughly in kmem-space, making fewer assumptions, and/
or making more precise calculations, thereby providing better 'detailed
process information'", or something like that, I suppose.


| I can't see the source on shellN, and I am not running the most
| recent sources here, but I imagine the option is still there.  It
| just gives an error instead of failing silently when called as
| non-root.

It's strangely "legal" again, after lastnight's even bigger update (see my
other post).  As to whether it's *functional* or not ...

/kim

===============================================================================
"Anything free is worth what you pay for it." --Lazarus Long [aka R.A.Heinlein]

In article <3652f5fa$0$12774@nntp1.ba.best.com>,
Kim DeVaughn <kimdv@best.com> wrote:
>As to whether it's *functional* or not ...

Well, in my FreeBSD source (stable of this summer), the first thing
the option checks is for uid=0 and it's ignored otherwise.

In article <72ut6k$ac$1@machaut.medieval.org>,
	Todd Michel McComb <mccomb@medieval.org> wrote:
|
| In article <3652f5fa$0$12774@nntp1.ba.best.com>,
| 	  Kim DeVaughn <kimdv@best.com> wrote:
| >
| >As to whether it's *functional* or not ...
|
| Well, in my FreeBSD source (stable of this summer), the first thing
| the option checks is for uid=0 and it's ignored otherwise.

Just out of curiousity, since you have the sources handy ... does it show
up in the "usage" blurb?

And if, as has been stated, that option controls whether information about
swapped processes is displayed, why should you have to be "root" to see it,
since you can see damn near anything else if you want to ...?

/kim

=============================================================
"There are lies ... damn lies ... and benchmarks."  --Unknown

In article <3652ff74$0$12776@nntp1.ba.best.com>,
Kim DeVaughn <kimdv@best.com> wrote:
>Just out of curiousity, since you have the sources handy ... does it show
>up in the "usage" blurb?

No, it doesn't.

>And if, as has been stated, that option controls whether information about
>swapped processes is displayed, why should you have to be "root" to see it,
>since you can see damn near anything else if you want to ...?

I have no thoughts on that, but I see no evidence that -f ever did
anything for non-root users.

Kim DeVaughn wrote:
> /kim
> 
> ===============================================================
> "I think I speak for everyone here when I say, 'Huh?'"  --Buffy

Okay, I can no longer resist... Kim, do you choose these quotes "by
hand," or does your randomizer have an uncanny knack for choosing apt
quotes?

-mikehan

In article <3653B969.CD861BDF@best.com>, Michael Han <mikehan@best.com> wrote:
|
| Kim DeVaughn wrote:
| >
| > ===============================================================
| > "I think I speak for everyone here when I say, 'Huh?'"  --Buffy
|
| Okay, I can no longer resist... Kim, do you choose these quotes "by
| hand," or does your randomizer have an uncanny knack for choosing apt
| quotes?

Yes.

/kim  (who doesn't really exist, but is just a fairly successful AI construct
       that was loosed on the net some years ago during a little-known, but
       highly successful hacking incident, that involved the NSA)

===============================================================================
"If you can't dazzle 'em with brilliance, baffle 'em with bullshit."  --Unknown

On 20 Nov 1998 21:27:11 GMT, kimdv@best.com wrote:
>
>/kim  (who doesn't really exist, but is just a fairly successful AI construct
>       that was loosed on the net some years ago during a little-known, but
>       highly successful hacking incident, that involved the NSA)

Hey baby can I read your source?


>===============================================================================
>"If you can't dazzle 'em with brilliance, baffle 'em with bullshit."  --Unknown


-- 
 --------------------------------------------------
"initiating.. 'getting the hell out of here' maneouver" - Lennier, babylon5

On 20 Nov 1998 21:48:17 GMT, phil.s2@bolthole.no-bots.com wrote:
>On 20 Nov 1998 21:27:11 GMT, kimdv@best.com wrote:
>>
>>/kim  (who doesn't really exist, but is just a fairly successful AI construct
>>       that was loosed on the net some years ago during a little-known, but
>>       highly successful hacking incident, that involved the NSA)
>
>Hey baby can I read your source?

Okay, "See your source" would have been better...

-- 
[trim the no-bots from my address to reply to me by email!]
 --------------------------------------------------
"initiating.. 'getting the hell out of here' maneouver" - Lennier, babylon5

In article <slrn75bouh.4c1.phil.s2@shell3.ba.best.com>,
Philip Brown <phil.s2@bolthole.no-bots.com> wrote:
>On 20 Nov 1998 21:48:17 GMT, phil.s2@bolthole.no-bots.com wrote:
>>On 20 Nov 1998 21:27:11 GMT, kimdv@best.com wrote:
>>>
>>>/kim  (who doesn't really exist, but is just a fairly successful AI construct
>>>       that was loosed on the net some years ago during a little-known, but
>>>       highly successful hacking incident, that involved the NSA)
>>
>>Hey baby can I read your source?
>
>Okay, "See your source" would have been better...
>

I think "show us your source!" says it even better.


	Bob O`Bob

-- 
+ email replies without this line will be discarded le4m5tvt (expires 28dec98) +

"On the internet, *everyone* will have their own fifteen minutes of blame."