Re: recent amanda on Debian?

[Jean-Francois Malouin]

* Jean-Francois Malouin  [20170210 
11:22]:
> * Jean-Louis Martineau  [20170210 11:10]:
> > valgrind is a useful tool to find memory corruption.
> > 
> > Try the attached patch.
> 
> Thanks Jean-Louis.
> 
> I'll try as soon as the server finishes a flush run. 
> Should take a few more hours to complete.

This as seemingly solved the glitch with amadmin:

~# ~amanda/sbin/amadmin top tape
The next Amanda should go onto 3 new tapes.
The next 3 new tapes already labelled are: NEO-T48-B00028, NEO-T48-B00029, 
NEO-T48-B00030

Thanks Jean-Louis,

jf
> 
> Thanks,
> jf
> 
> > 
> > Jean-Louis
> > 
> > On 10/02/17 09:35 AM, Jean-Francois Malouin wrote:
> > > * Jean-Louis Martineau  [20170210 07:54]:
> > >> Jean-Francois,
> > >>
> > >> I was not able to find the problem with the stacktrace.
> > >>
> > >> Do you have valgrind installed on your system?, if yes, can you run the
> > >> following command:
> > >> G_DEBUG=gc-friendly G_SLICE=always-malloc valgrind --trace-children=yes
> > >> --log-file=valgrind.amtape.output --leak-check=full
> > >> --leak-resolution=high --track-origins=yes amadmin top tape
> > >>
> > >> And send me the resulting valgrind.amtape.output file.
> > > The output from valgrind file is attached.
> > >
> > > cheers,
> > > jf
> > >
> > >> Jean-Louis
> > >>
> > >> On 09/02/17 02:19 PM, Jean-Francois Malouin wrote:
> > >>> * Jean-Louis Martineau  [20170209 14:09]:
> >  Jean-Francois,
> > 
> >  Can you run amadmin in gdb and get a stacktrace?
> > 
> >  $ gdb ~amanda/sbin/amadmin
> >  (gdb) run top tape
> >  (gdb) bt
> >  (gdb) quit
> > 
> >  Jean-Louis
> > >>> Here you go:
> > >>>
> > >>> amanda@dow:~$ gdb ~amanda/sbin/amadmin
> > >>> GNU gdb (Debian 7.7.1+dfsg-5) 7.7.1
> > >>> Copyright (C) 2014 Free Software Foundation, Inc.
> > >>> License GPLv3+: GNU GPL version 3 or later
> > >>> 
> > >>> This is free software: you are free to change and redistribute it.
> > >>> There is NO WARRANTY, to the extent permitted by law.  Type "show 
> > >>> copying"
> > >>> and "show warranty" for details.
> > >>> This GDB was configured as "x86_64-linux-gnu".
> > >>> Type "show configuration" for configuration details.
> > >>> For bug reporting instructions, please see:
> > >>> .
> > >>> Find the GDB manual and other documentation resources online at:
> > >>> .
> > >>> For help, type "help".
> > >>> Type "apropos word" to search for commands related to "word"...
> > >>> Reading symbols from /opt/amanda/sbin/amadmin...(no debugging symbols
> > >>> found)...done.
> > >>> (gdb) run top tape
> > >>> Starting program: /opt/amanda-3.4.2/sbin/amadmin top tape
> > >>> [Thread debugging using libthread_db enabled]
> > >>> Using host libthread_db library 
> > >>> "/lib/x86_64-linux-gnu/libthread_db.so.1".
> > >>> The next Amanda should go onto 3 new tapes.
> > >>> *** Error in `/opt/amanda-3.4.2/sbin/amadmin': corrupted double-linked 
> > >>> list: 0x006f4710 ***
> > >>> The next 3 new tapes already labelled are: NEO-T48-B00022,
> > >>> NEO-T48-B00023, NEO-T48-B00024
> > >>> Program received signal SIGABRT, Aborted.
> > >>> 0x750ee067 in __GI_raise (sig=sig@entry=6) at 
> > >>> ../nptl/sysdeps/unix/sysv/linux/raise.c:56
> > >>> 56  ../nptl/sysdeps/unix/sysv/linux/raise.c: No such file or directory.
> > >>> (gdb) bt
> > >>> #0  0x750ee067 in __GI_raise (sig=sig@entry=6) at 
> > >>> ../nptl/sysdeps/unix/sysv/linux/raise.c:56
> > >>> #1  0x750ef448 in __GI_abort () at abort.c:89
> > >>> #2  0x7512c1b4 in __libc_message (do_abort=do_abort@entry=1, 
> > >>> fmt=fmt@entry=0x75221210 "*** Error in `%s': %s: 0x%s ***\n") at 
> > >>> ../sysdeps/posix/libc_fatal.c:175
> > >>> #3  0x7513198e in malloc_printerr (action=1, str=0x7521d2ec 
> > >>> "corrupted double-linked list", ptr=) at malloc.c:4996
> > >>> #4  0x75131c4c in malloc_consolidate 
> > >>> (av=av@entry=0x7545e620 ) at malloc.c:4165
> > >>> #5  0x75132e98 in _int_malloc (av=0x7545e620 , 
> > >>> bytes=32816) at malloc.c:3423
> > >>> #6  0x75135020 in __GI___libc_malloc (bytes=32816) at 
> > >>> malloc.c:2891
> > >>> #7  0x7516ee31 in __alloc_dir (fd=3, close_fd=, 
> > >>> flags=, statp=) at 
> > >>> ../sysdeps/posix/opendir.c:207
> > >>> #8  0x774a5d19 in debug_unlink_old () from 
> > >>> /opt/amanda/lib/amanda/libamanda-3.4.2.so
> > >>> #9  0x774a6cf2 in debug_close () from 
> > >>> /opt/amanda/lib/amanda/libamanda-3.4.2.so
> > >>> #10 0x004038b7 in main ()
> > >>> (gdb) quit
> > >>> A debugging session is active.
> > >>>
> > >>>   Inferior 1 [process 4173] will be killed.
> > >>>
> > >>> Quit anyway? (y or n) y
> > >>>
> > >>> cheers,
> > >>> jf
> > >>>
> >  On 09/02/17 

Re: recent amanda on Debian?

[Jean-Francois Malouin]

* Jean-Louis Martineau  [20170210 11:10]:
> valgrind is a useful tool to find memory corruption.
> 
> Try the attached patch.

Thanks Jean-Louis.

I'll try as soon as the server finishes a flush run. 
Should take a few more hours to complete.

Thanks,
jf

> 
> Jean-Louis
> 
> On 10/02/17 09:35 AM, Jean-Francois Malouin wrote:
> > * Jean-Louis Martineau  [20170210 07:54]:
> >> Jean-Francois,
> >>
> >> I was not able to find the problem with the stacktrace.
> >>
> >> Do you have valgrind installed on your system?, if yes, can you run the
> >> following command:
> >> G_DEBUG=gc-friendly G_SLICE=always-malloc valgrind --trace-children=yes
> >> --log-file=valgrind.amtape.output --leak-check=full
> >> --leak-resolution=high --track-origins=yes amadmin top tape
> >>
> >> And send me the resulting valgrind.amtape.output file.
> > The output from valgrind file is attached.
> >
> > cheers,
> > jf
> >
> >> Jean-Louis
> >>
> >> On 09/02/17 02:19 PM, Jean-Francois Malouin wrote:
> >>> * Jean-Louis Martineau  [20170209 14:09]:
>  Jean-Francois,
> 
>  Can you run amadmin in gdb and get a stacktrace?
> 
>  $ gdb ~amanda/sbin/amadmin
>  (gdb) run top tape
>  (gdb) bt
>  (gdb) quit
> 
>  Jean-Louis
> >>> Here you go:
> >>>
> >>> amanda@dow:~$ gdb ~amanda/sbin/amadmin
> >>> GNU gdb (Debian 7.7.1+dfsg-5) 7.7.1
> >>> Copyright (C) 2014 Free Software Foundation, Inc.
> >>> License GPLv3+: GNU GPL version 3 or later
> >>> 
> >>> This is free software: you are free to change and redistribute it.
> >>> There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
> >>> and "show warranty" for details.
> >>> This GDB was configured as "x86_64-linux-gnu".
> >>> Type "show configuration" for configuration details.
> >>> For bug reporting instructions, please see:
> >>> .
> >>> Find the GDB manual and other documentation resources online at:
> >>> .
> >>> For help, type "help".
> >>> Type "apropos word" to search for commands related to "word"...
> >>> Reading symbols from /opt/amanda/sbin/amadmin...(no debugging symbols
> >>> found)...done.
> >>> (gdb) run top tape
> >>> Starting program: /opt/amanda-3.4.2/sbin/amadmin top tape
> >>> [Thread debugging using libthread_db enabled]
> >>> Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
> >>> The next Amanda should go onto 3 new tapes.
> >>> *** Error in `/opt/amanda-3.4.2/sbin/amadmin': corrupted double-linked 
> >>> list: 0x006f4710 ***
> >>> The next 3 new tapes already labelled are: NEO-T48-B00022,
> >>> NEO-T48-B00023, NEO-T48-B00024
> >>> Program received signal SIGABRT, Aborted.
> >>> 0x750ee067 in __GI_raise (sig=sig@entry=6) at 
> >>> ../nptl/sysdeps/unix/sysv/linux/raise.c:56
> >>> 56  ../nptl/sysdeps/unix/sysv/linux/raise.c: No such file or directory.
> >>> (gdb) bt
> >>> #0  0x750ee067 in __GI_raise (sig=sig@entry=6) at 
> >>> ../nptl/sysdeps/unix/sysv/linux/raise.c:56
> >>> #1  0x750ef448 in __GI_abort () at abort.c:89
> >>> #2  0x7512c1b4 in __libc_message (do_abort=do_abort@entry=1, 
> >>> fmt=fmt@entry=0x75221210 "*** Error in `%s': %s: 0x%s ***\n") at 
> >>> ../sysdeps/posix/libc_fatal.c:175
> >>> #3  0x7513198e in malloc_printerr (action=1, str=0x7521d2ec 
> >>> "corrupted double-linked list", ptr=) at malloc.c:4996
> >>> #4  0x75131c4c in malloc_consolidate (av=av@entry=0x7545e620 
> >>> ) at malloc.c:4165
> >>> #5  0x75132e98 in _int_malloc (av=0x7545e620 , 
> >>> bytes=32816) at malloc.c:3423
> >>> #6  0x75135020 in __GI___libc_malloc (bytes=32816) at 
> >>> malloc.c:2891
> >>> #7  0x7516ee31 in __alloc_dir (fd=3, close_fd=, 
> >>> flags=, statp=) at 
> >>> ../sysdeps/posix/opendir.c:207
> >>> #8  0x774a5d19 in debug_unlink_old () from 
> >>> /opt/amanda/lib/amanda/libamanda-3.4.2.so
> >>> #9  0x774a6cf2 in debug_close () from 
> >>> /opt/amanda/lib/amanda/libamanda-3.4.2.so
> >>> #10 0x004038b7 in main ()
> >>> (gdb) quit
> >>> A debugging session is active.
> >>>
> >>>   Inferior 1 [process 4173] will be killed.
> >>>
> >>> Quit anyway? (y or n) y
> >>>
> >>> cheers,
> >>> jf
> >>>
>  On 09/02/17 01:56 PM, Jean-Francois Malouin wrote:
> > * Stefan G. Weichinger  [20170209 13:36]:
> >> Does anyone use amanda 3.4.2 packages on debian jessie already?
> >> Which one, do they work?
> > I only see amanda 3.3.6 on Debian/Jessie stable repos at the moment:
> >
> > Inst amanda-common (1:3.3.6-4 Debian:8.7/stable [amd64])
> > Inst amanda-server (1:3.3.6-4 Debian:8.7/stable [amd64])
> > Inst amanda-client (1:3.3.6-4 Debian:8.7/stable [amd64])
> >
> > I've been running a self-compiled amanda-3.4.1 on a Jessie host for a
> 

Re: recent amanda on Debian?

[Jean-Louis Martineau]

valgrind is a useful tool to find memory corruption.

Try the attached patch.

Jean-Louis

On 10/02/17 09:35 AM, Jean-Francois Malouin wrote:
> * Jean-Louis Martineau  [20170210 07:54]:
>> Jean-Francois,
>>
>> I was not able to find the problem with the stacktrace.
>>
>> Do you have valgrind installed on your system?, if yes, can you run the
>> following command:
>> G_DEBUG=gc-friendly G_SLICE=always-malloc valgrind --trace-children=yes
>> --log-file=valgrind.amtape.output --leak-check=full
>> --leak-resolution=high --track-origins=yes amadmin top tape
>>
>> And send me the resulting valgrind.amtape.output file.
> The output from valgrind file is attached.
>
> cheers,
> jf
>
>> Jean-Louis
>>
>> On 09/02/17 02:19 PM, Jean-Francois Malouin wrote:
>>> * Jean-Louis Martineau  [20170209 14:09]:
 Jean-Francois,

 Can you run amadmin in gdb and get a stacktrace?

 $ gdb ~amanda/sbin/amadmin
 (gdb) run top tape
 (gdb) bt
 (gdb) quit

 Jean-Louis
>>> Here you go:
>>>
>>> amanda@dow:~$ gdb ~amanda/sbin/amadmin
>>> GNU gdb (Debian 7.7.1+dfsg-5) 7.7.1
>>> Copyright (C) 2014 Free Software Foundation, Inc.
>>> License GPLv3+: GNU GPL version 3 or later
>>> 
>>> This is free software: you are free to change and redistribute it.
>>> There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
>>> and "show warranty" for details.
>>> This GDB was configured as "x86_64-linux-gnu".
>>> Type "show configuration" for configuration details.
>>> For bug reporting instructions, please see:
>>> .
>>> Find the GDB manual and other documentation resources online at:
>>> .
>>> For help, type "help".
>>> Type "apropos word" to search for commands related to "word"...
>>> Reading symbols from /opt/amanda/sbin/amadmin...(no debugging symbols
>>> found)...done.
>>> (gdb) run top tape
>>> Starting program: /opt/amanda-3.4.2/sbin/amadmin top tape
>>> [Thread debugging using libthread_db enabled]
>>> Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
>>> The next Amanda should go onto 3 new tapes.
>>> *** Error in `/opt/amanda-3.4.2/sbin/amadmin': corrupted double-linked 
>>> list: 0x006f4710 ***
>>> The next 3 new tapes already labelled are: NEO-T48-B00022,
>>> NEO-T48-B00023, NEO-T48-B00024
>>> Program received signal SIGABRT, Aborted.
>>> 0x750ee067 in __GI_raise (sig=sig@entry=6) at 
>>> ../nptl/sysdeps/unix/sysv/linux/raise.c:56
>>> 56  ../nptl/sysdeps/unix/sysv/linux/raise.c: No such file or directory.
>>> (gdb) bt
>>> #0  0x750ee067 in __GI_raise (sig=sig@entry=6) at 
>>> ../nptl/sysdeps/unix/sysv/linux/raise.c:56
>>> #1  0x750ef448 in __GI_abort () at abort.c:89
>>> #2  0x7512c1b4 in __libc_message (do_abort=do_abort@entry=1, 
>>> fmt=fmt@entry=0x75221210 "*** Error in `%s': %s: 0x%s ***\n") at 
>>> ../sysdeps/posix/libc_fatal.c:175
>>> #3  0x7513198e in malloc_printerr (action=1, str=0x7521d2ec 
>>> "corrupted double-linked list", ptr=) at malloc.c:4996
>>> #4  0x75131c4c in malloc_consolidate (av=av@entry=0x7545e620 
>>> ) at malloc.c:4165
>>> #5  0x75132e98 in _int_malloc (av=0x7545e620 , 
>>> bytes=32816) at malloc.c:3423
>>> #6  0x75135020 in __GI___libc_malloc (bytes=32816) at malloc.c:2891
>>> #7  0x7516ee31 in __alloc_dir (fd=3, close_fd=, 
>>> flags=, statp=) at 
>>> ../sysdeps/posix/opendir.c:207
>>> #8  0x774a5d19 in debug_unlink_old () from 
>>> /opt/amanda/lib/amanda/libamanda-3.4.2.so
>>> #9  0x774a6cf2 in debug_close () from 
>>> /opt/amanda/lib/amanda/libamanda-3.4.2.so
>>> #10 0x004038b7 in main ()
>>> (gdb) quit
>>> A debugging session is active.
>>>
>>>   Inferior 1 [process 4173] will be killed.
>>>
>>> Quit anyway? (y or n) y
>>>
>>> cheers,
>>> jf
>>>
 On 09/02/17 01:56 PM, Jean-Francois Malouin wrote:
> * Stefan G. Weichinger  [20170209 13:36]:
>> Does anyone use amanda 3.4.2 packages on debian jessie already?
>> Which one, do they work?
> I only see amanda 3.3.6 on Debian/Jessie stable repos at the moment:
>
> Inst amanda-common (1:3.3.6-4 Debian:8.7/stable [amd64])
> Inst amanda-server (1:3.3.6-4 Debian:8.7/stable [amd64])
> Inst amanda-client (1:3.3.6-4 Debian:8.7/stable [amd64])
>
> I've been running a self-compiled amanda-3.4.1 on a Jessie host for a
> month now with little problems so far, except this
>
> :~# su amanda -c "~amanda/sbin/amadmin top tape"
> The next Amanda should go onto 3 new tapes.
> *** Error in `/opt/amanda/sbin/amadmin': corrupted double-linked list: 
> 0x01153710 ***
> The next 3 new tapes already labelled are: NEO-T48-B00022, 
> NEO-T48-B00023, NEO-T48-B00024
>
> I have seen this on 3.4.1 and 3.4.2 (I just 

Re: recent amanda on Debian?

[Stefan G. Weichinger]

Am 2017-02-10 um 16:43 schrieb Jose M Calhariz:
> On Fri, Feb 10, 2017 at 04:23:29PM +0100, Stefan G. Weichinger
> wrote:
>> btw: I wrote to you on january 31 personally asking for
>> amanda-packages in debian.
>> 
>> my email might have been filtered to spam or so ...
>> 
> 
> I have replied to you, so maybe was my reply that was tagged as
> spam.

seems like. But this mail came in via both ml and directly.
And your spam score is low ;-) good!

my postfix-logs don't show any rejected mails from you.

the amanda.org mailserver is a bit strange (sorry, Todd ;-)), so pls
use li...@xunil.at for direct mail. my fault to write to you via
s...@amanda.org (I am subscribed to the ml with that address, though)

thanks

Re: recent amanda on Debian?

[Jose M Calhariz]

On Fri, Feb 10, 2017 at 04:23:29PM +0100, Stefan G. Weichinger wrote:
> Am 2017-02-10 um 16:13 schrieb Jose M Calhariz:
> > I have started on creating the official Debian packages of amanda
> > 3.4.2, and I would like to have beta testers, in case you are
> > interested.
> 
> sure, yes
> 
> btw: I wrote to you on january 31 personally asking for amanda-packages
> in debian.
> 
> my email might have been filtered to spam or so ...
> 

I have replied to you, so maybe was my reply that was tagged as spam.


-- 
--
Sucrilhos Multi I/O. Com saída Cereal.


signature.asc
Description: Digital signature

Re: recent amanda on Debian?

[Jean-Francois Malouin]

* Jose M Calhariz  [20170210 10:14]:
> I have started on creating the official Debian packages of amanda
> 3.4.2, and I would like to have beta testers, in case you are
> interested.

no problem, count me in.

jf

> 
> On Thu, Feb 09, 2017 at 07:35:58PM +0100, Stefan G. Weichinger wrote:
> > 
> > Does anyone use amanda 3.4.2 packages on debian jessie already?
> > Which one, do they work?
> > 
> > I have to switch OS on a backup server and want to stay with
> > amanda-3.4.x because I need some of the latest features ...
> > 
> > self-compile, you know ... I can do but I'd like to avoid.
> > lower maintenance.
> > 
> 
> Kind regards
> Jose M Calhariz
> 
> 
> -- 
> --
> Sucrilhos Multi I/O. Com saída Cereal.

Re: recent amanda on Debian?

[Stefan G. Weichinger]

Am 2017-02-10 um 16:13 schrieb Jose M Calhariz:
> I have started on creating the official Debian packages of amanda
> 3.4.2, and I would like to have beta testers, in case you are
> interested.

sure, yes

btw: I wrote to you on january 31 personally asking for amanda-packages
in debian.

my email might have been filtered to spam or so ...

Re: recent amanda on Debian?

[Jose M Calhariz]

I have started on creating the official Debian packages of amanda
3.4.2, and I would like to have beta testers, in case you are
interested.

On Thu, Feb 09, 2017 at 07:35:58PM +0100, Stefan G. Weichinger wrote:
> 
> Does anyone use amanda 3.4.2 packages on debian jessie already?
> Which one, do they work?
> 
> I have to switch OS on a backup server and want to stay with
> amanda-3.4.x because I need some of the latest features ...
> 
> self-compile, you know ... I can do but I'd like to avoid.
> lower maintenance.
> 

Kind regards
Jose M Calhariz


-- 
--
Sucrilhos Multi I/O. Com saída Cereal.


signature.asc
Description: Digital signature

Re: recent amanda on Debian?

[Jean-Louis Martineau]

Jean-Francois,

I was not able to find the problem with the stacktrace.

Do you have valgrind installed on your system?, if yes, can you run the 
following command:
G_DEBUG=gc-friendly G_SLICE=always-malloc valgrind --trace-children=yes 
--log-file=valgrind.amtape.output --leak-check=full 
--leak-resolution=high --track-origins=yes amadmin top tape

And send me the resulting valgrind.amtape.output file.

Jean-Louis

On 09/02/17 02:19 PM, Jean-Francois Malouin wrote:
> * Jean-Louis Martineau  [20170209 14:09]:
>> Jean-Francois,
>>
>> Can you run amadmin in gdb and get a stacktrace?
>>
>> $ gdb ~amanda/sbin/amadmin
>> (gdb) run top tape
>> (gdb) bt
>> (gdb) quit
>>
>> Jean-Louis
> Here you go:
>
> amanda@dow:~$ gdb ~amanda/sbin/amadmin
> GNU gdb (Debian 7.7.1+dfsg-5) 7.7.1
> Copyright (C) 2014 Free Software Foundation, Inc.
> License GPLv3+: GNU GPL version 3 or later
> 
> This is free software: you are free to change and redistribute it.
> There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
> and "show warranty" for details.
> This GDB was configured as "x86_64-linux-gnu".
> Type "show configuration" for configuration details.
> For bug reporting instructions, please see:
> .
> Find the GDB manual and other documentation resources online at:
> .
> For help, type "help".
> Type "apropos word" to search for commands related to "word"...
> Reading symbols from /opt/amanda/sbin/amadmin...(no debugging symbols
> found)...done.
> (gdb) run top tape
> Starting program: /opt/amanda-3.4.2/sbin/amadmin top tape
> [Thread debugging using libthread_db enabled]
> Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
> The next Amanda should go onto 3 new tapes.
> *** Error in `/opt/amanda-3.4.2/sbin/amadmin': corrupted double-linked list: 
> 0x006f4710 ***
> The next 3 new tapes already labelled are: NEO-T48-B00022,
> NEO-T48-B00023, NEO-T48-B00024
> Program received signal SIGABRT, Aborted.
> 0x750ee067 in __GI_raise (sig=sig@entry=6) at 
> ../nptl/sysdeps/unix/sysv/linux/raise.c:56
> 56  ../nptl/sysdeps/unix/sysv/linux/raise.c: No such file or directory.
> (gdb) bt
> #0  0x750ee067 in __GI_raise (sig=sig@entry=6) at 
> ../nptl/sysdeps/unix/sysv/linux/raise.c:56
> #1  0x750ef448 in __GI_abort () at abort.c:89
> #2  0x7512c1b4 in __libc_message (do_abort=do_abort@entry=1, 
> fmt=fmt@entry=0x75221210 "*** Error in `%s': %s: 0x%s ***\n") at 
> ../sysdeps/posix/libc_fatal.c:175
> #3  0x7513198e in malloc_printerr (action=1, str=0x7521d2ec 
> "corrupted double-linked list", ptr=) at malloc.c:4996
> #4  0x75131c4c in malloc_consolidate (av=av@entry=0x7545e620 
> ) at malloc.c:4165
> #5  0x75132e98 in _int_malloc (av=0x7545e620 , 
> bytes=32816) at malloc.c:3423
> #6  0x75135020 in __GI___libc_malloc (bytes=32816) at malloc.c:2891
> #7  0x7516ee31 in __alloc_dir (fd=3, close_fd=, 
> flags=, statp=) at 
> ../sysdeps/posix/opendir.c:207
> #8  0x774a5d19 in debug_unlink_old () from 
> /opt/amanda/lib/amanda/libamanda-3.4.2.so
> #9  0x774a6cf2 in debug_close () from 
> /opt/amanda/lib/amanda/libamanda-3.4.2.so
> #10 0x004038b7 in main ()
> (gdb) quit
> A debugging session is active.
>
>  Inferior 1 [process 4173] will be killed.
>
> Quit anyway? (y or n) y
>
> cheers,
> jf
>
>> On 09/02/17 01:56 PM, Jean-Francois Malouin wrote:
>>> * Stefan G. Weichinger  [20170209 13:36]:
 Does anyone use amanda 3.4.2 packages on debian jessie already?
 Which one, do they work?
>>> I only see amanda 3.3.6 on Debian/Jessie stable repos at the moment:
>>>
>>> Inst amanda-common (1:3.3.6-4 Debian:8.7/stable [amd64])
>>> Inst amanda-server (1:3.3.6-4 Debian:8.7/stable [amd64])
>>> Inst amanda-client (1:3.3.6-4 Debian:8.7/stable [amd64])
>>>
>>> I've been running a self-compiled amanda-3.4.1 on a Jessie host for a
>>> month now with little problems so far, except this
>>>
>>> :~# su amanda -c "~amanda/sbin/amadmin top tape"
>>> The next Amanda should go onto 3 new tapes.
>>> *** Error in `/opt/amanda/sbin/amadmin': corrupted double-linked list: 
>>> 0x01153710 ***
>>> The next 3 new tapes already labelled are: NEO-T48-B00022, NEO-T48-B00023, 
>>> NEO-T48-B00024
>>>
>>> I have seen this on 3.4.1 and 3.4.2 (I just upgraded).
>>>
>>> cheers,
>>> jf
>>>
>>>
 I have to switch OS on a backup server and want to stay with
 amanda-3.4.x because I need some of the latest features ...

 self-compile, you know ... I can do but I'd like to avoid.
 lower maintenance.

Re: recent amanda on Debian?

[Jean-Francois Malouin]

* Jean-Louis Martineau  [20170209 14:09]:
> Jean-Francois,
> 
> Can you run amadmin in gdb and get a stacktrace?
> 
> $ gdb ~amanda/sbin/amadmin
> (gdb) run top tape
> (gdb) bt
> (gdb) quit
> 
> Jean-Louis

Here you go:

amanda@dow:~$ gdb ~amanda/sbin/amadmin
GNU gdb (Debian 7.7.1+dfsg-5) 7.7.1
Copyright (C) 2014 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later

This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
and "show warranty" for details.
This GDB was configured as "x86_64-linux-gnu".
Type "show configuration" for configuration details.
For bug reporting instructions, please see:
.
Find the GDB manual and other documentation resources online at:
.
For help, type "help".
Type "apropos word" to search for commands related to "word"...
Reading symbols from /opt/amanda/sbin/amadmin...(no debugging symbols
found)...done.
(gdb) run top tape
Starting program: /opt/amanda-3.4.2/sbin/amadmin top tape
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
The next Amanda should go onto 3 new tapes.
*** Error in `/opt/amanda-3.4.2/sbin/amadmin': corrupted double-linked list: 
0x006f4710 ***
The next 3 new tapes already labelled are: NEO-T48-B00022,
NEO-T48-B00023, NEO-T48-B00024
Program received signal SIGABRT, Aborted.
0x750ee067 in __GI_raise (sig=sig@entry=6) at 
../nptl/sysdeps/unix/sysv/linux/raise.c:56
56  ../nptl/sysdeps/unix/sysv/linux/raise.c: No such file or directory.
(gdb) bt
#0  0x750ee067 in __GI_raise (sig=sig@entry=6) at 
../nptl/sysdeps/unix/sysv/linux/raise.c:56
#1  0x750ef448 in __GI_abort () at abort.c:89
#2  0x7512c1b4 in __libc_message (do_abort=do_abort@entry=1, 
fmt=fmt@entry=0x75221210 "*** Error in `%s': %s: 0x%s ***\n") at 
../sysdeps/posix/libc_fatal.c:175
#3  0x7513198e in malloc_printerr (action=1, str=0x7521d2ec 
"corrupted double-linked list", ptr=) at malloc.c:4996
#4  0x75131c4c in malloc_consolidate (av=av@entry=0x7545e620 
) at malloc.c:4165
#5  0x75132e98 in _int_malloc (av=0x7545e620 , 
bytes=32816) at malloc.c:3423
#6  0x75135020 in __GI___libc_malloc (bytes=32816) at malloc.c:2891
#7  0x7516ee31 in __alloc_dir (fd=3, close_fd=, 
flags=, statp=) at ../sysdeps/posix/opendir.c:207
#8  0x774a5d19 in debug_unlink_old () from 
/opt/amanda/lib/amanda/libamanda-3.4.2.so
#9  0x774a6cf2 in debug_close () from 
/opt/amanda/lib/amanda/libamanda-3.4.2.so
#10 0x004038b7 in main ()
(gdb) quit
A debugging session is active.

Inferior 1 [process 4173] will be killed.

Quit anyway? (y or n) y

cheers,
jf

> 
> On 09/02/17 01:56 PM, Jean-Francois Malouin wrote:
> > * Stefan G. Weichinger  [20170209 13:36]:
> >> Does anyone use amanda 3.4.2 packages on debian jessie already?
> >> Which one, do they work?
> > I only see amanda 3.3.6 on Debian/Jessie stable repos at the moment:
> >
> > Inst amanda-common (1:3.3.6-4 Debian:8.7/stable [amd64])
> > Inst amanda-server (1:3.3.6-4 Debian:8.7/stable [amd64])
> > Inst amanda-client (1:3.3.6-4 Debian:8.7/stable [amd64])
> >
> > I've been running a self-compiled amanda-3.4.1 on a Jessie host for a
> > month now with little problems so far, except this
> >
> > :~# su amanda -c "~amanda/sbin/amadmin top tape"
> > The next Amanda should go onto 3 new tapes.
> > *** Error in `/opt/amanda/sbin/amadmin': corrupted double-linked list: 
> > 0x01153710 ***
> > The next 3 new tapes already labelled are: NEO-T48-B00022, NEO-T48-B00023, 
> > NEO-T48-B00024
> >
> > I have seen this on 3.4.1 and 3.4.2 (I just upgraded).
> >
> > cheers,
> > jf
> >
> >
> >> I have to switch OS on a backup server and want to stay with
> >> amanda-3.4.x because I need some of the latest features ...
> >>
> >> self-compile, you know ... I can do but I'd like to avoid.
> >> lower maintenance.

Re: recent amanda on Debian?

[Jean-Louis Martineau]

Jean-Francois,

Can you run amadmin in gdb and get a stacktrace?

$ gdb ~amanda/sbin/amadmin
(gdb) run top tape
(gdb) bt
(gdb) quit

Jean-Louis

On 09/02/17 01:56 PM, Jean-Francois Malouin wrote:
> * Stefan G. Weichinger  [20170209 13:36]:
>> Does anyone use amanda 3.4.2 packages on debian jessie already?
>> Which one, do they work?
> I only see amanda 3.3.6 on Debian/Jessie stable repos at the moment:
>
> Inst amanda-common (1:3.3.6-4 Debian:8.7/stable [amd64])
> Inst amanda-server (1:3.3.6-4 Debian:8.7/stable [amd64])
> Inst amanda-client (1:3.3.6-4 Debian:8.7/stable [amd64])
>
> I've been running a self-compiled amanda-3.4.1 on a Jessie host for a
> month now with little problems so far, except this
>
> :~# su amanda -c "~amanda/sbin/amadmin top tape"
> The next Amanda should go onto 3 new tapes.
> *** Error in `/opt/amanda/sbin/amadmin': corrupted double-linked list: 
> 0x01153710 ***
> The next 3 new tapes already labelled are: NEO-T48-B00022, NEO-T48-B00023, 
> NEO-T48-B00024
>
> I have seen this on 3.4.1 and 3.4.2 (I just upgraded).
>
> cheers,
> jf
>
>
>> I have to switch OS on a backup server and want to stay with
>> amanda-3.4.x because I need some of the latest features ...
>>
>> self-compile, you know ... I can do but I'd like to avoid.
>> lower maintenance.

Re: recent amanda on Debian?

[Jean-Francois Malouin]

* Stefan G. Weichinger  [20170209 13:36]:
> 
> Does anyone use amanda 3.4.2 packages on debian jessie already?
> Which one, do they work?

I only see amanda 3.3.6 on Debian/Jessie stable repos at the moment:

Inst amanda-common (1:3.3.6-4 Debian:8.7/stable [amd64])
Inst amanda-server (1:3.3.6-4 Debian:8.7/stable [amd64])
Inst amanda-client (1:3.3.6-4 Debian:8.7/stable [amd64])

I've been running a self-compiled amanda-3.4.1 on a Jessie host for a
month now with little problems so far, except this 

:~# su amanda -c "~amanda/sbin/amadmin top tape"
The next Amanda should go onto 3 new tapes.
*** Error in `/opt/amanda/sbin/amadmin': corrupted double-linked list: 
0x01153710 ***
The next 3 new tapes already labelled are: NEO-T48-B00022, NEO-T48-B00023, 
NEO-T48-B00024

I have seen this on 3.4.1 and 3.4.2 (I just upgraded).

cheers,
jf


> I have to switch OS on a backup server and want to stay with
> amanda-3.4.x because I need some of the latest features ...
> 
> self-compile, you know ... I can do but I'd like to avoid.
> lower maintenance.