love the wrap, so it looks like 8.1 will be my next testbed. hummmm....... "wondering aloud" I have a number of people that look after different 'bits' of a number of webservers for non-profit organisations, and they do it for free. :-) I wonder if the next version could allow me as overall administrator to give them limited access to their bits but not all of it? It would certainly cut my workload down considerably (to just co-ordinating and checking the change instead of doing). scsijon At 06:42 AM 9/14/02, Curtis Rey wrote:

Woohoo, looks like they got the message. Very promising. I think this is a wonderful addition to the distro and if it holds out to be what I think it is it will just put SuSE way ahead of the rest - as usual.

Thanks for the FYI. Curtis :)

On Friday 13 September 2002 15:20, Michael Hasenstein wrote:

...

See http://www.suse.de/~sh/YaST2-Package-Manager/

-- SuSE 8.0 Pro Gigabyte GA-7VRXP AMD XP 1800+ Registered Linux User # 285235 Registered Linux Machine # 167063

-- Check the headers for your unsubscription address For additional commands send e-mail to suse-linux-e-help@suse.com Also check the archives at http://lists.suse.com Please read the FAQs: suse-linux-e-faq@suse.com

* Jay Vollmer (jvollmer@visi.com) [020913 17:43]: :: ::So, when is SL8.1 due to be released? :: Sept 25th in Europe and if they hold to pattern then about 10-15 days later in the states. -- Ben Rosenberg ---===---===---===--- mailto:ben@whack.org Tell me what you believe.. I tell you what you should see.

* Salman Khilji [Sep 14. 2002 00:39]:

Can 8.0 users upgrade to this?

Of course. -- Mads Martin Jørgensen, http://mmj.dk "Why make things difficult, when it is possible to make them cryptic and totally illogic, with just a little bit more effort?" -- A. P. J.

* Erik Jakobsen [Sep 14. 2002 11:06]:

...

...

Can 8.0 users upgrade to this?

Of course.

Where is it to be found ?.

Ah, I though if 8.0 users could upgrade to 8.1 using this, and they can. But no, there is none for 8.0. It's an 8.1 thing. -- Mads Martin Jørgensen, http://mmj.dk "Why make things difficult, when it is possible to make them cryptic and totally illogic, with just a little bit more effort?" -- A. P. J.

Erik Jakobsen wrote:

...

...

...

...

Can 8.0 users upgrade to this?

Of course.

Where is it to be found ?.

Ah, I though if 8.0 users could upgrade to 8.1 using this, and they can. But no, there is none for 8.0. It's an 8.1 thing.

I see. Thanks for the information.

The upgrade for SuSE Linux N is always SuSE Linux N+1. What else? The upgrade for the Honda Accord N is also the Honda Accord N+1, and no one asks them why they don't put that new engine and head airbag into 2000 models... ;-) Michael

The upgrade for SuSE Linux N is always SuSE Linux N+1. What else? The upgrade for the Honda Accord N is also the Honda Accord N+1, and no one asks them why they don't put that new engine and head airbag into 2000 models... ;-)

????? We are talking about an add on package, not the core kernel. Why should we not expect that an updated version of this package should not be available for older versions? That seems to be part of the problem with all of this. We should NOT have to update everthing everytime there is a minor improvement to something that simply does not work properly in previous releases. I am still waiting for a version of YOU that will actually allow me to make these updates because the current release DOES NOT WORK WITH A PROXY. I use fou4s and it does the job so why can't the official version do it. I will probably update to 8.1, but I should not have to do so to get simple bug fixes! By the way does this new Package Manager actually work with proxies? I could not see any references to that. -- Lester Caine ----------------------------- L.S.Caine Electronic Services

* Keith Winston (kwinston@twmi.rr.com) [020914 09:11]: ::On Sat, 2002-09-14 at 11:39, lester@lsces.co.uk wrote: ::> We are talking about an add on package, not the core kernel. ::> Why should we not expect that an updated version of this ::> package should not be available for older versions? That ::> seems to be part of the problem with all of this. We should ::> NOT have to update everthing everytime there is a minor ::> improvement to something that simply does not work properly ::> in previous releases. I am still waiting for a version of ::> YOU that will actually allow me to make these updates ::> because the current release DOES NOT WORK WITH A PROXY. I ::> use fou4s and it does the job so why can't the official ::> version do it. ::> ::> I will probably update to 8.1, but I should not have to do ::> so to get simple bug fixes! :: ::I have to disagree with you about the new package manager. This is a ::newly developed program, not a bug fix. Exactly. YaST2's whole packaging subsystem has been rewritten from scratch as I've been told. No they didn't change from RPM..I mean the way that YaST2 handles RPM's. So of course there may be major system changes so that YaST2's new found ability works right. The user will most likely not have to worry about such things but maybe..just maybe it's to much of a change to have for 8.0 :) Beside if you have 8.0 then just buy that upgrade which is 1/2 the price of the Pro package and by the time it hits the states it will have been 10 months since the previous release. That's not so bad..is it..is it really? :) Cheers! -- Ben Rosenberg ---===---===---===--- mailto:ben@whack.org Tell me what you believe.. I tell you what you should see.

...

I am still waiting for a version of ::> YOU that will actually allow me to make these updates ::> because the current release DOES NOT WORK WITH A PROXY.

it works here behind a proxy, so i don't know what your problem is ..

any error message etc ?

Just read all the messages on fou4s and apt - they all agree that you will not update via a proxy. -- Lester Caine ----------------------------- L.S.Caine Electronic Services

1) That still doesn't explain why i don't have a problem with a proxy server in Yast, and a lot of other users here in Belgium who have a cable internet connections via a proxy don't have a problem also. 2) I'm not subscribed to this mailinglist to go to another forum, trying to find the problem of a user there, and discussed it here. So where did you define the proxy, works ftp and wget etc.

?? This list has all the messages - Early June Quote from Richard Bos 19/6/2002

Hi Lester,

this is a known problem with YOU and should have been fixed with SuSE-8.0, however as you noticed it is still not easy.

He goes on to suggest apt, but I went with fou4s and have not bothered to switch back, but it is my understanding that if your 'dial out' connection is not on the machine that is using YOU then even in 8.0, the proxy problem had not been fixed last time a wasted time trying to set it up. -- Lester Caine ----------------------------- L.S.Caine Electronic Services

On Sat, 14 Sep 2002 20:42:17 -0700 Salman Khilji wrote:

Having to upgrade to 8.1 just for this is unacceptable. I paid $80 for 8.0 Pro and I am NOT paying $80 again within 6 months! This would mean LInux is more expensive than Windows XP!!

I think this should be made available for 8.0 users.

true, but isn't better, hey there are cheaper good options ... slackware

Plus I don't want to upgrade becuase 8.1 is not going to have gcc 2.95.3. I don't want to downgrade to gcc 3.x since the stuff that I need is broken on all versions of 3.x that I have worked with. (Last time I tested was 3.1...haven't tried with 3.2). Salman On Saturday 14 September 2002 17:52, Lao Toma wrote:

AFAIK, you don't pay FULL price for an upgrade, or am I wrong?

After I compile my program I need access to the debug symbols (addresses and tyeps). I use a modified version of objdump (from binutils) to get to the debug symbols. My application monitors the value of static variables without stopping the target application unlike a source line debugger which has to stop the application to look at the values of variables. Even the unmodified latest version of objdump could not read the debug info generated by gcc 3.1. Haven't tried 3.2. It would say something like "bad mangled name" when you tried to get objdump to read debug symbols generated by gcc. Someone told me that this was a problme in the past and would be in the future because binutils and gcc was maintained by different groups. I tried all versions till 3.1 when I gave up on it and decided to stick with 2.95.3. That was one of the reasons I dumped Mandrake 8.0 and switched to SuSE 8.0. I won't switch for a long time to any newer version till they get evertying straigthen out with gcc. Salman On Sunday 15 September 2002 11:43, Mads Martin Joergensen wrote:

* Salman Khilji [Sep 15. 2002 03:22]:

...

Plus I don't want to upgrade becuase 8.1 is not going to have gcc 2.95.3. I don't want to downgrade to gcc 3.x since the stuff that I need is broken on all versions of 3.x that I have worked with. (Last time I tested was 3.1...haven't tried with 3.2).

What are these things? We've ported a lot of applications, so it could be the apps. you're mentioning is among them, and the fix just needs to be sent upstream.

* Graham Murray [Sep 16. 2002 07:59]:

...

Even the unmodified latest version of objdump could not read the debug info generated by gcc 3.1. Haven't tried 3.2.

It would say something like "bad mangled name" when you tried to get objdump to read debug symbols generated by gcc. Someone told me that this was a problme in the past and would be in the future because binutils and gcc was maintained by different groups.

Have you tried applying your objdump modifications to the latest binutils. I am using 2.13.90.0.4 (from ftp.kernel.org) and "objdump -t 'progname'" where progname was built with gcc3.2, lists the symbols perfectly.

One solution to the problem of binutils and gcc being out of step is to build them from the same source tree. I have never done this myself, but believe that it is frequently used when building a cross-compile toolchain.

When you upgrade the compiler, you need newest binutils as well. They changed debugging format to DWARF2. I don't think gcc is to blame here. -- Mads Martin Jørgensen, http://mmj.dk "Why make things difficult, when it is possible to make them cryptic and totally illogic, with just a little bit more effort?" -- A. P. J.

Salman Khilji writes:

Could you please do an objdump -g <prog> and let me know what you get. graham@home:~> objdump -g /usr/local/bin/emacs

/usr/local/bin/emacs: file format elf32-i386 /usr/src/packages/BUILD/glibc-2.2.5/csu/init.c: typedef int32 int; typedef int8 char; typedef int32 long int; typedef uint32 unsigned int; typedef uint32 long unsigned int; typedef int64 long long int; typedef uint64 long long unsigned int; typedef int16 short int; typedef uint16 short unsigned int; typedef int8 signed char; typedef uint8 unsigned char; typedef float float; typedef double double; typedef float96 long double; typedef struct %anon1 { /* size 8 */ int real; /* bitsize 32, bitpos 0 */ int imag; /* bitsize 32, bitpos 32 */ } complex int; typedef complex float complex float; typedef complex double complex double; typedef complex float96 complex long double; typedef void void; int _IO_stdin_used /* 0x81c0da4 */; struct __gconv_loaded_object { /* id 0 */ }; struct _pthread_descr_struct { /* id 0 */ }; ../include/libc-symbols.h: /usr/src/packages/BUILD/glibc-2.2.5/cc/config.h: ../sysdeps/gnu/_G_config.h: typedef struct %anon2 { /* size 12 */ long int __pos; /* bitsize 32, bitpos 0 */ struct %anon3 { /* size 8 */ int __count; /* bitsize 32, bitpos 0 */ union %anon4 { /* size 4 */ unsigned int __wch; /* bitsize 32, bitpos 0 */ char __wchb[4]; /* bitsize 32, bitpos 0 */ } __value; /* bitsize 32, bitpos 32 */ } __state; /* bitsize 64, bitpos 32 */ } _G_fpos_t; typedef struct %anon5 { /* size 16 */ long long int __pos; /* bitsize 64, bitpos 0 */ struct %anon3 __state; /* bitsize 64, bitpos 64 */ } _G_fpos64_t; typedef union %anon6 { /* size 44 */ struct __gconv_info /* id 7 */ __cd; /* bitsize 64, bitpos 0 */ struct %anon8 { /* size 44 */ struct __gconv_info /* id 7 */ __cd; /* bitsize 64, bitpos 0 */ struct __gconv_step_data /* id 9 */ __data; /* bitsize 288, bitpos 64 */ } __combined; /* bitsize 352, bitpos 0 */ } _G_iconv_t; typedef short int _G_int16_t; typedef int _G_int32_t; typedef short unsigned int _G_uint16_t; typedef unsigned int _G_uint32_t; ../sysdeps/unix/sysv/linux/bits/types.h: typedef unsigned char __u_char; typedef short unsigned int __u_short; typedef unsigned int __u_int; typedef long unsigned int __u_long; typedef long long unsigned int __u_quad_t; typedef long long int __quad_t; typedef signed char __int8_t; typedef unsigned char __uint8_t; typedef short int __int16_t; typedef short unsigned int __uint16_t; typedef int __int32_t; typedef unsigned int __uint32_t; typedef long long int __int64_t; typedef long long unsigned int __uint64_t; typedef __quad_t *__qaddr_t; typedef __u_quad_t __dev_t; typedef __u_int __uid_t; typedef __u_int __gid_t; typedef __u_long __ino_t; typedef __u_int __mode_t; typedef __u_int __nlink_t; typedef long int __off_t; typedef __quad_t __loff_t; typedef int __pid_t; typedef int __ssize_t; typedef __u_long __rlim_t; typedef __u_quad_t __rlim64_t; typedef __u_int __id_t; typedef struct %anon10 { /* size 8 */ int __val[2]; /* bitsize 64, bitpos 0 */ } __fsid_t; typedef int __daddr_t; typedef char *__caddr_t; typedef long int __time_t; typedef unsigned int __useconds_t; typedef long int __suseconds_t; typedef long int __swblk_t; typedef long int __clock_t; typedef int __clockid_t; typedef int __timer_t; typedef int __key_t; typedef short unsigned int __ipc_pid_t; typedef long int __blksize_t; typedef long int __blkcnt_t; typedef __quad_t __blkcnt64_t; typedef __u_long __fsblkcnt_t; typedef __u_quad_t __fsblkcnt64_t; typedef __u_long __fsfilcnt_t; typedef __u_quad_t __fsfilcnt64_t; typedef __u_quad_t __ino64_t; typedef __loff_t __off64_t; typedef long int __t_scalar_t; typedef long unsigned int __t_uscalar_t; typedef int __intptr_t; typedef unsigned int __socklen_t; ../include/features.h: ../include/sys/cdefs.h: ../misc/sys/cdefs.h: /usr/lib/gcc-lib/i486-suse-linux/2.95.3/include/stddef.h: typedef unsigned int size_t; typedef long int wchar_t; typedef unsigned int wint_t; ../linuxthreads/sysdeps/pthread/bits/pthreadtypes.h: struct _pthread_fastlock { /* size 8 id 11 */ long int __status; /* bitsize 32, bitpos 0 */ int __spinlock; /* bitsize 32, bitpos 32 */ }; typedef struct _pthread_descr_struct /* id 0 */ *_pthread_descr; struct __pthread_attr_s { /* size 36 id 12 */ int __detachstate; /* bitsize 32, bitpos 0 */ int __schedpolicy; /* bitsize 32, bitpos 32 */ struct __sched_param /* id 13 */ __schedparam; /* bitsize 32, bitpos 64 */ int __inheritsched; /* bitsize 32, bitpos 96 */ int __scope; /* bitsize 32, bitpos 128 */ size_t __guardsize; /* bitsize 32, bitpos 160 */ int __stackaddr_set; /* bitsize 32, bitpos 192 */ void *__stackaddr; /* bitsize 32, bitpos 224 */ size_t __stacksize; /* bitsize 32, bitpos 256 */ }; typedef struct __pthread_attr_s /* id 12 */ pthread_attr_t; typedef struct %anon14 { /* size 12 */ struct _pthread_fastlock /* id 11 */ __c_lock; /* bitsize 64, bitpos 0 */ _pthread_descr __c_waiting; /* bitsize 32, bitpos 64 */ } pthread_cond_t; typedef struct %anon15 { /* size 4 */ int __dummy; /* bitsize 32, bitpos 0 */ } pthread_condattr_t; typedef unsigned int pthread_key_t; typedef struct %anon16 { /* size 24 */ int __m_reserved; /* bitsize 32, bitpos 0 */ int __m_count; /* bitsize 32, bitpos 32 */ _pthread_descr __m_owner; /* bitsize 32, bitpos 64 */ int __m_kind; /* bitsize 32, bitpos 96 */ struct _pthread_fastlock /* id 11 */ __m_lock; /* bitsize 64, bitpos 128 */ } pthread_mutex_t; typedef struct %anon17 { /* size 4 */ int __mutexkind; /* bitsize 32, bitpos 0 */ } pthread_mutexattr_t; typedef int pthread_once_t; struct _pthread_rwlock_t { /* size 32 id 18 */ struct _pthread_fastlock /* id 11 */ __rw_lock; /* bitsize 64, bitpos 0 */ int __rw_readers; /* bitsize 32, bitpos 64 */ _pthread_descr __rw_writer; /* bitsize 32, bitpos 96 */ _pthread_descr __rw_read_waiting; /* bitsize 32, bitpos 128 */ _pthread_descr __rw_write_waiting; /* bitsize 32, bitpos 160 */ int __rw_kind; /* bitsize 32, bitpos 192 */ int __rw_pshared; /* bitsize 32, bitpos 224 */ }; typedef struct _pthread_rwlock_t /* id 18 */ pthread_rwlock_t; typedef struct %anon19 { /* size 8 */ int __lockkind; /* bitsize 32, bitpos 0 */ int __pshared; /* bitsize 32, bitpos 32 */ } pthread_rwlockattr_t; typedef int pthread_spinlock_t; typedef struct %anon20 { /* size 20 */ struct _pthread_fastlock /* id 11 */ __ba_lock; /* bitsize 64, bitpos 0 */ int __ba_required; /* bitsize 32, bitpos 64 */ int __ba_present; /* bitsize 32, bitpos 96 */ _pthread_descr __ba_waiting; /* bitsize 32, bitpos 128 */ } pthread_barrier_t; typedef struct %anon21 { /* size 4 */ int __pshared; /* bitsize 32, bitpos 0 */ } pthread_barrierattr_t; typedef long unsigned int pthread_t; ../sysdeps/unix/sysv/linux/bits/sched.h: struct __sched_param { /* size 4 id 13 */ int __sched_priority; /* bitsize 32, bitpos 0 */ }; ../include/wchar.h: ../wcsmbs/wchar.h: typedef struct %anon3 __mbstate_t; ../sysdeps/unix/sysv/linux/i386/bits/wchar.h: ../include/gconv.h: ../iconv/gconv.h: typedef int (*__gconv_fct) (/* unknown */); typedef int (*__gconv_init_fct) (/* unknown */); typedef void (*__gconv_end_fct) (/* unknown */); typedef int (*__gconv_trans_fct) (/* unknown */); typedef int (*__gconv_trans_context_fct) (/* unknown */); typedef int (*__gconv_trans_query_fct) (/* unknown */); typedef int (*__gconv_trans_init_fct) (/* unknown */); typedef void (*__gconv_trans_end_fct) (/* unknown */); struct __gconv_trans_data { /* size 20 id 22 */ __gconv_trans_fct __trans_fct; /* bitsize 32, bitpos 0 */ __gconv_trans_context_fct __trans_context_fct; /* bitsize 32, bitpos 32 */ __gconv_trans_end_fct __trans_end_fct; /* bitsize 32, bitpos 64 */ void *__data; /* bitsize 32, bitpos 96 */ struct __gconv_trans_data /* id 22 */ *__next; /* bitsize 32, bitpos 128 */ }; struct __gconv_step { /* size 56 id 23 */ struct __gconv_loaded_object /* id 0 */ *__shlib_handle; /* bitsize 32, bitpos 0 */ char *__modname; /* bitsize 32, bitpos 32 */ int __counter; /* bitsize 32, bitpos 64 */ char *__from_name; /* bitsize 32, bitpos 96 */ char *__to_name; /* bitsize 32, bitpos 128 */ __gconv_fct __fct; /* bitsize 32, bitpos 160 */ __gconv_init_fct __init_fct; /* bitsize 32, bitpos 192 */ __gconv_end_fct __end_fct; /* bitsize 32, bitpos 224 */ int __min_needed_from; /* bitsize 32, bitpos 256 */ int __max_needed_from; /* bitsize 32, bitpos 288 */ int __min_needed_to; /* bitsize 32, bitpos 320 */ int __max_needed_to; /* bitsize 32, bitpos 352 */ int __stateful; /* bitsize 32, bitpos 384 */ void *__data; /* bitsize 32, bitpos 416 */ }; struct __gconv_step_data { /* size 36 id 9 */ unsigned char *__outbuf; /* bitsize 32, bitpos 0 */ unsigned char *__outbufend; /* bitsize 32, bitpos 32 */ int __flags; /* bitsize 32, bitpos 64 */ int __invocation_counter; /* bitsize 32, bitpos 96 */ int __internal_use; /* bitsize 32, bitpos 128 */ __mbstate_t *__statep; /* bitsize 32, bitpos 160 */ __mbstate_t __state; /* bitsize 64, bitpos 192 */ struct __gconv_trans_data /* id 22 */ *__trans; /* bitsize 32, bitpos 256 */ }; struct __gconv_info { /* size 8 id 7 */ size_t __nsteps; /* bitsize 32, bitpos 0 */ struct __gconv_step /* id 23 */ *__steps; /* bitsize 32, bitpos 32 */ struct __gconv_step_data /* id 9 */ __data[]; /* bitpos 64 */ }; typedef struct __gconv_info /* id 7 */ *__gconv_t; graham@home:~>

On Saturday 14 September 2002 14.48, Vitaly Shishakov wrote:

On Saturday 14 September 2002 00:20, Michael Hasenstein wrote:

...

See http://www.suse.de/~sh/YaST2-Package-Manager/

Will it have a feature to browse directories with packages like package manager in YaST1 ?

Vitaly.

That *is* an important feature. I hope they will have it.

On Sunday 15 September 2002 08.59, Graham Murray wrote:

Michael Hasenstein writes:

...

See http://www.suse.de/~sh/YaST2-Package-Manager/

Will this solve the "anomaly" whereby current package managers (eg kpackage) think that release candidates are later versions than the actual release? eg they think that "koffice-1.2rc1-20.rpm" is newer than "koffice-1.2-3.i386.rpm".

I think the problem lies with the people who create the rpms. rpm doesn't have advanced AI. In the above case, "2rc1" comes after "2" in any sorting algorithm I can think of off the top of my head. How about naming 1.2rc1 1.1.99 or something?! //Anders