Hello,
No idea about this issue from anybody ?
We know that a workaround to this issue is to create a symlink to the
executable file called a.out
ln -s /laki2/singapore/dbs/dbs_3/dbs_xrdb/mx a.out
before running the pstack command on the core file.
Thanks,
William.
On 01/19/07 William Roche wrote:
>
> Hello,
>
> I'd like to know if this problem rings a bell to anyone:
>
> A customer has a program that generates a core, and on the machine where
> the core is generated, they run a "pstack core" command.
>
> On S10 FCS (03/05), the stack has most of the symbols resolved --> OK
>
>
> Now when they run the same program on S10 Update 1 (01/06) or S10 Update
> 3, it generates a core and the "pstack core" output on this machine
> misses the symbols from the executable itself.
>
> Running the pstack command with the LIBPROC_DEBUG flag gave us some
> interesting elements, but I'd like to understand how those differences
> could be triggered ?
>
> I've attached an HTML table with some differences I noticed, and the
> full outputs for the failing case (01/06) and the succeeding case (03/05).
>
> If you have any hint on how to reproduce a test case that would fail the
> same way (we can't use the customer's application and can't get their
> core file). Or even better, how to fix it ;)
>
>
> Thanks in advance for any hint you could give me.
>
>
> Cheers,
> William.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://mail.opensolaris.org/pipermail/tools-linking/attachments/20070124/3d0444d9/attachment.html>
