report for merged logs

[Francis J. Lacoste]

On lun, 2003-01-06 at 15:21, Edward Quick wrote:
> [...]
> I have applied the patch, but now I'm getting a new error:
> 
> $smpd9 lr_run lr_xml2report -U email -S postfix -o html  \
> /home/ftp/src/rsl59/xml/report-20021230.235500.xml \
> /home/ftp/src/rsl59/xml/report-20021231.235500.xml \
> /home/ftp/src/rsl59/xml/report-20030101.235500.xml \
> /home/ftp/src/rsl59/xml/report-20030102.235500.xml \
> >/home/ftp/src/rsl59/reports/overall-report.tar
> 
> bogus message: Segmentation Fault - core dumped
> email postfix lr_tag-20030106201244-11016 lr_xml2report err lr_xml_merge 
> failed
> email postfix lr_tag-20030106201244-11016 lr_xml2report notice keeping 
> /home/ftp/src/rsl59/tmp/lr_xml2report.email.lr_tag-20030106201244-11016.xml 
> for debugging
> smpd9$
> 
> Reading symbolic information for perl
> core file header read successfully
> Reading symbolic information for rtld /usr/lib/ld.so.1
> dbx: program is not active
> Reading symbolic information for libsocket.so.1
> Reading symbolic information for libnsl.so.1
> Reading symbolic information for libdl.so.1
> Reading symbolic information for libc.so.1
> Reading symbolic information for libmp.so.2
> Reading symbolic information for libc_psr.so.1
> Reading symbolic information for en_UK.so.1
> Reading symbolic information for Fcntl.so
> Reading symbolic information for IO.so
> Reading symbolic information for Expat.so
> Reading symbolic information for libexpat.so.0
> Reading symbolic information for DBI.so
> Reading symbolic information for POSIX.so
> program terminated by signal SEGV (no mapping at the fault address)
> (dbx)
> 
> (dbx) where
> =>[1] Perl_hv_fetch_ent(0xdd4080, 0x3c0250, 0x0, 0x41ad82, 0xdd3498, 
> 0x3dd2a4), at 0x631d0
>   [2] Perl_pp_helem(0x3dc7a0, 0xdd6a0, 0x0, 0x0, 0xdd400, 0x40c920), at 
> 0x6b9d0
>   [3] Perl_runops_standard(0xdd6a0, 0x5d9cb8, 0x1, 0x0, 0x1996e0, 
> 0x7000000), at 0xb1148
>   [4] perl_run(0xdd000, 0xdcc00, 0xdd800, 0xdd000, 0xdd800, 0x800000), at 
> 0x26e14
>   [5] main(0x0, 0xeffff684, 0xdc8e8, 0xd1400, 0x0, 0x0), at 0x24550
> (dbx)
> 
> 
> I really can't see what's causing this problem, though I suspect it may have 
> something to do with the size of the logs I'm using:
> 
> smpd9$ ls -l xml
> total 5708
> -rw-r--r--   1 u752359  staff     508146 Jan  1 11:38 
> report-20021230.235500.xml
> -rw-r--r--   1 u752359  staff     469383 Jan  1 18:15 
> report-20021231.235500.xml
> -rw-r--r--   1 u752359  staff     500364 Jan  6 11:41 
> report-20030101.235500.xml
> -rw-r--r--   1 u752359  staff    1442646 Jan  5 17:58 
> report-20030102.235500.xml
> smpd9$
> 

It is possible that the segfault is related to the 1.4MB XML report
that you are trying to merge. One way to confirm this would be to
only the three 500k reports. 

Anyway, this looks like a perl/XS compilation issue. On what platform
are you running Lire? Solaris is my guess. In that case, which Solaris
version and which perl version are you using ? 

Another users had some problem with the Solaris' perl (he also got
segfaults when processing *some* XML reports). He compiled
perl 5.6.1 and used the system malloc instead of the perl builtin malloc
and after that he didn't any other problems. You might want to try that.

You can check which malloc version is used by running 
perl -V:usemymalloc

Don't hesitate to post a follow-ups here on that issue.

Kind regards

Francis J. Lacoste

-- 
Francis J. Lacoste              . .           http://www.logreport.org
/^LogReport$/               . .               flacoste at logreport.org
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
Url : http://lists.logreport.org/pipermail/questions/attachments/20030110/db9c53ea/attachment.bin