RE: [PATCH] Be verbose during shl_load on HP-UX

(I hit the wrong reply button, and my response just went to Bill - resending again) -Original Message- From: William A. Rowe, Jr. [mailto:[EMAIL PROTECTED] At 11:21 AM 6/5/2003, MATHIHALLI,MADHUSUDAN (HP-Cupertino,ex1) wrote: Well.. in that case, I'd suppose it's better to dup stderr to

RE: [PATCH] Be verbose during shl_load on HP-UX

Madhu, Can we please end this thread? I think we had this discussion a year ago on the exact same topic. The outcome of these discussions have consistently been: outputting error messages to the user is the responsibility of the application using the library, not of the library itself. As to

Re: [PATCH] Be verbose during shl_load on HP-UX

Sander Striker wrote: Madhu, Can we please end this thread? I think we had this discussion a year ago on the exact same topic. The outcome of these discussions have consistently been: outputting error messages to the user is the responsibility of the application using the library, not of the

Re: [PATCH] Be verbose during shl_load on HP-UX

On Fri, May 30, 2003 at 01:54:21PM -0400, MATHIHALLI,MADHUSUDAN (HP-Cupertino,ex1) wrote: ... Now, what should be the behaviour under the following scenarios : - everything is fine : program runs fine. No problems. - Unresolved symbols and stderr is open : program will report the

RE: [PATCH] Be verbose during shl_load on HP-UX

-Original Message- From: Joe Orton [mailto:[EMAIL PROTECTED] If that is the case, can we can put a specific case for HPUX, for 1. when BIND_VERBOSE is enabled, and 2. stderr is closed, and 3. you get this message, we can print a particular error message (because we know whatz

RE: [PATCH] Be verbose during shl_load on HP-UX

From: MATHIHALLI,MADHUSUDAN (HP-Cupertino,ex1) [mailto:[EMAIL PROTECTED] Sent: Thursday, June 05, 2003 2:07 AM Anyways, the disconnect is because you're trying to solve the case when stderr is closed/disabled - and I'm not. I don't think we're converging on any solution here. I'd appreciate

RE: [PATCH] Be verbose during shl_load on HP-UX

-Original Message- From: Sander Striker [mailto:[EMAIL PROTECTED] [SNIP] Anyways, the disconnect is because you're trying to solve the case when stderr is closed/disabled - and I'm not. I don't think we're converging on any solution here. I'd appreciate if we could get more inputs

RE: [PATCH] Be verbose during shl_load on HP-UX

-Original Message- From: Joe Orton [mailto:[EMAIL PROTECTED] [SNIP] 2. how do you tell reliably whether stderr is closed? Actually, I was just wondering : why would somebody want to close the stderr ? I'd think the best they'd do is to dup stderr to /dev/null or something. -Madhu

RE: [PATCH] Be verbose during shl_load on HP-UX

--On Wednesday, June 4, 2003 9:10 PM -0400 MATHIHALLI,MADHUSUDAN (HP-Cupertino,ex1) [EMAIL PROTECTED] wrote: Actually, I was just wondering : why would somebody want to close the stderr ? I'd think the best they'd do is to dup stderr to /dev/null or something. IIRC, httpd will specifically close

RE: [PATCH] Be verbose during shl_load on HP-UX

--On Wednesday, June 4, 2003 9:01 PM -0400 MATHIHALLI,MADHUSUDAN (HP-Cupertino,ex1) [EMAIL PROTECTED] wrote: If I understand you correctly, libraries should return the errno (because errors may mean strings also), and not try writing anything to stderr. The application should do the job of

RE: [PATCH] Be verbose during shl_load on HP-UX

On Wed, 4 Jun 2003, MATHIHALLI,MADHUSUDAN (HP-Cupertino,ex1) wrote: If I understand you correctly, libraries should return the errno (because errors may mean strings also), and not try writing anything to stderr. The application should do the job of writing to stderr. Is that just your

RE: [PATCH] Be verbose during shl_load on HP-UX

At 08:34 PM 6/4/2003, Justin Erenkrantz wrote: --On Wednesday, June 4, 2003 9:10 PM -0400 MATHIHALLI,MADHUSUDAN (HP-Cupertino,ex1) [EMAIL PROTECTED] wrote: Actually, I was just wondering : why would somebody want to close the stderr ? I'd think the best they'd do is to dup stderr to /dev/null or

RE: [PATCH] Be verbose during shl_load on HP-UX

-Original Message- From: Justin Erenkrantz [mailto:[EMAIL PROTECTED] --On Wednesday, June 4, 2003 9:10 PM -0400 MATHIHALLI,MADHUSUDAN (HP-Cupertino,ex1) [EMAIL PROTECTED] wrote: Actually, I was just wondering : why would somebody want to close the stderr ? I'd think the best they'd

RE: [PATCH] Be verbose during shl_load on HP-UX

At 11:21 AM 6/5/2003, MATHIHALLI,MADHUSUDAN (HP-Cupertino,ex1) wrote: Well.. in that case, I'd suppose it's better to dup stderr to /dev/null (even better - the apache error_log) rather than closing stderr. If the intention is to avoid children from writing to stderr, we ought to dup it so that

RE: [PATCH] Be verbose during shl_load on HP-UX

Hi, Here's what I did : I wrote a small program (attached below). In Step 1, I close stdout, and try running the program.. Since the program doesn't have all the ap_* symbols, it complains, and the output I got was : $ ./a /usr/lib/dld.sl: Unresolved symbol: ap_chroot (data) from

RE: [PATCH] Be verbose during shl_load on HP-UX

-Original Message- From: Garrett Rooney [mailto:[EMAIL PROTECTED] Be that as it may, it's not APR's place to start spewing random stuff (however interesting to the programmer) to stderr. To begin with, let us not forget the fact that all applications have to be developed by a

Re: [PATCH] Be verbose during shl_load on HP-UX

On Wednesday, May 28, 2003, at 07:17 PM, MATHIHALLI,MADHUSUDAN (HP-Cupertino,ex1) wrote: APR is a general purpose library, and you have no idea what kind of application it is being used in. Perhaps it's not appropriate for there to be any output at all to the terminal? Perhaps there is no

Re: [PATCH] Be verbose during shl_load on HP-UX

--On Wednesday, May 28, 2003 7:32 PM -0400 Garrett Rooney [EMAIL PROTECTED] wrote: Well, if you feel so strongly about having the option (and it should be an option, turned off by default, IMO), then I suggest revising the patch such that it makes apr_dso_load take a flags parameter (changing

RE: [PATCH] Be verbose during shl_load on HP-UX

-Original Message- From: Justin Erenkrantz [mailto:[EMAIL PROTECTED] [SNIP] Is it possible for HP-UX's DSO's error to be captured in a string rather than forcibly sent to stderr? On the top of my head, I don't think it's possible. The man page for shl_laod says Print verbose

Re: [PATCH] Be verbose during shl_load on HP-UX

On Wed, May 28, 2003 at 04:44:53PM -0700, MATHIHALLI,MADHUSUDAN (HP-Cupertino,ex1) wrote: -Original Message- From: Justin Erenkrantz [mailto:[EMAIL PROTECTED] [SNIP] Is it possible for HP-UX's DSO's error to be captured in a string rather than forcibly sent to stderr? On

[PATCH] Be verbose during shl_load on HP-UX

Hi, 'wondering if anybody has objections for the below patch. By having the BIND_VERBOSE option, we can know what really failed during the shl_load. -Madhu Index: dso.c === RCS file: /home/cvs/apr/dso/unix/dso.c,v retrieving

Re: [PATCH] Be verbose during shl_load on HP-UX

On Wed, May 28, 2003 at 02:46:28PM -0700, MATHIHALLI,MADHUSUDAN (HP-Cupertino,ex1) wrote: 'wondering if anybody has objections for the below patch. By having the BIND_VERBOSE option, we can know what really failed during the shl_load. Hi Madhu, BIND_VERBOSE was removed on purpose - the

RE: [PATCH] Be verbose during shl_load on HP-UX

, May 28, 2003 3:10 PM To: MATHIHALLI,MADHUSUDAN (HP-Cupertino,ex1) Cc: 'dev@apr.apache.org' Subject: Re: [PATCH] Be verbose during shl_load on HP-UX On Wed, May 28, 2003 at 02:46:28PM -0700, MATHIHALLI,MADHUSUDAN (HP-Cupertino,ex1) wrote: 'wondering if anybody has objections for the below