[Dxspider-support] DX-Spider problem Caught signal BUS at /spider/perl/cluster.pl line 482.

[Dirk Koopman]

You may find this useful:

http://en.wikipedia.org/wiki/Bus_error

If the problem persists I would still be looking at a hardware mediated 
corruption issue (e.g. a corrupt swap file). It could even be a failing 
memory stick or motherboard. I am running the same version in production 
on a number of busy nodes and have *never* seen a BUS error. In fact, I 
am pretty certain that I have never seen a BUS error in any program I 
have written over the last 30 odd years for i86 hardware. Used to see 
them occasionally on Sparc/Motorola and other word oriented hardware 
before that though and it was always a data misalignment bug.

Dirk

On 21/11/14 14:24, Jens Schoon wrote:
> Hello Dirk and hello list-users,
> thanks for the fast replay.
>
> Perl-Version is:
> This is perl 5, version 18, subversion 2 (v5.18.2) built for
> x86_64-linux-gnu-thread-multi
> (with 42 registered patches, see perl -V for more detail)
>
> The record for VE1SYN seems to be OK:
> VE1SYN  bless( {call => 'VE1SYN',lastoper => '1398732343',qth => 'Sydney
> NS',sort => 'U',lat => '46.15',name => 'Greg Macinnis',long =>
> '-60.2',qra => 'FN96VD'}, 'DXUser' )
>
> A "perl user_asc" (after remove users.v3) shows:
> There are 66184 user records and 0 errors
>
> Lets see, if it helps.
>
> 73 de DH6BB
> Jens
>
>
>
> Am 21.11.2014 um 14:46 schrieb Dirk Koopman:
>> Please send emails of this type to the support list so that everyone
>> gets a chance to see what's going on (and please join if you have not
>> yet done so).
>>
>> A bus error is a perl bug (actually one that one rarely sees at all on
>> Intel/AMD hardware). Programs written in perl should not be able to
>> generate these.
>>
>> However, I do use the perl package "Storable" which seems to be very
>> intolerant of disc errors, perhaps caused by uncontrolled
>> interruptions of the program (probably power related).
>>
>> I would be inclined to rebuild the user database and see whether that
>> fixes it. Problems of this sort usually seem to be cured by this.
>>
>> So:
>>
>> stop the node.
>>
>> cd /spider
>> rm users.v3
>> perl user_asc
>>
>> # time passes
>>
>> restart the node.
>>
>> Now it may be that there is some problem with the actual record for
>> VE1SYN, so it might be worth just opening the user_asc file with your
>> favourite editor and searching for VE1SYN. Then, have a look at the
>> record and compare it with similar ones. If there long strings of 0000
>> (or other characters) in fields then just delete the record before
>> doing the 'perl user_asc'. If it isn't one of your regular users it
>> won't make any difference to the running of your node.
>>
>> Dirk G1TLH
>>
>> On 21/11/14 13:14, Jens Schoon wrote:
>>> Hello Dirk,
>>> i have a problem with my DX-Spider at DB0FHO-4. I get the following
>>> message multiple time. Sometimes after a day or two, sometimes after a
>>> week. Any ideas?
>>>
>>> <- I DB0SUE-7
>>> PC92^EA4RH-5^61826^C^5EA4RH-5:5455^5LA3WAA-6^5EA7URG-5^1SV1GSV^5PY1NB-4^5LA7Q-6^1EA4RH-2^5EA6VQ-1^1EA1DDU^H97^
>>>
>>> -> D DB0ERF-5
>>> PC92^EA4RH-5^61826^C^5EA4RH-5:5455^5LA3WAA-6^5EA7URG-5^1SV1GSV^5PY1NB-4^5LA7Q-6^1EA4RH-2^5EA6VQ-1^1EA1DDU^H96^
>>>
>>> <- I DB0SUE-7 PC92^VE1CFB^61848^A^^1YO2MFZ^H93^
>>> -> D DB0ERF-5 PC92^VE1CFB^61848^A^^1YO2MFZ^H92^
>>> <- I DB0SUE-7 PC92^9A0DXC^61831^D^^1SP8FHM^H94^
>>> -> D DB0ERF-5 PC92^9A0DXC^61831^D^^1SP8FHM^H93^
>>> <- I DB0ERF-5 PC92^VE1CFB^61848^A^^1YO2MFZ^H92^
>>> <- I DB0SUE-7 PC92^9A0DXC^61831.01^A^^19A2SW^1SP9LAS^H96^
>>> -> D DB0ERF-5 PC92^9A0DXC^61831.01^A^^19A2SW^1SP9LAS^H95^
>>> <- I DB0ERF-5 PC92^9A0DXC^61831^D^^1SP8FHM^H94^
>>> <- I DB0ERF-5 PC92^9A0DXC^61831.01^A^^19A2SW^1SP9LAS^H95^
>>> <- I DB0SUE-7 PC41^VE1SYN^1^Greg^H24^~
>>> -> D DB0ERF-5 PC41^VE1SYN^1^Greg^H23^~
>>> <- I DB0ERF-5 PC41^VE1SYN^1^Greg^H25^~
>>> <- I DB0SUE-7 PC41^VE1SYN^2^Sydney NS^H28^~
>>> Caught signal BUS at /spider/perl/cluster.pl line 482.
>>> DXChannel DB0ERF-5 destroyed (4)
>>> DXChannel DB0DY destroyed (3)
>>> DXChannel DB0SUE-7 destroyed (2)
>>>           (in cleanup) Can't call method "year" on an undefined value at
>>> /spider/perl/DXLog.pm line 151 during global destruction.
>>>           (in cleanup) Can't call method "year" on an undefined value at
>>> /spider/perl/DXLog.pm line 151 during global destruction.
>>> DXChannel DB0FHO-4 destroyed (2)
>>>           (in cleanup) Can't call method "year" on an undefined value at
>>> /spider/perl/DXLog.pm line 151 during global destruction.
>>>           (in cleanup) Can't call method "year" on an undefined value at
>>> /spider/perl/DXLog.pm line 151 during global destruction.
>>>
>>>
>>> DB0ERF-5 and DB0SUE-7 are my link-partners. DB0DY is a user.
>>>
>>> System is Debian Linux:
>>> Linux version 3.14-2-amd64 (debian-kernel at lists.debian.org) (gcc version
>>> 4.8.3 (Debian 4.8.3-7) ) #1 SMP Debian 3.14.15-2 (2014-08-09)
>>>
>>> Spider-Version is:
>>> DXSpider V1.55, build 0.163 (git: 365d2aa[r])
>>>
>>> Many thanks!
>>> 73 de DH6BB
>>> Jens
>>>
>>
>>
>