On May 17, 2010, at 8:59 PM, Leslie P. Polzer wrote:

> On May 17, 4:47 pm, Jan Rychter <[email protected]> wrote:
> [...]
>>  12: SB-KERNEL::COLLECT-GARBAGE
>>  13: (COMMON-LISP::FLET SB-THREAD::WITH-MUTEX-THUNK)
>>  14: (COMMON-LISP::FLET WITHOUT-INTERRUPTS-BODY-[CALL-WITH-MUTEX]293)
>>  15: (COMMON-LISP::FLET WITHOUT-INTERRUPTS-BODY-[SUB-GC]19)
>>  16: (SB-C::TL-XEP SB-KERNEL::SUB-GC)
>>  17: Foreign fp = 0xb6241f88, ra = 0x8066e4b
>>  18: Foreign fp = 0xb6241fa8, ra = 0x8052a0d
>>  19: Foreign fp = 0xb6241fc8, ra = 0x805365e
>>  20: Foreign fp = 0xb6241ff8, ra = 0x80586bb
>>  21: Foreign fp = 0xb6242028, ra = 0x805755b
>>  22: Foreign fp = 0xb62423e8, ra = 0xb7fcde70
>>  23: SB-IMPL::EXTRACT-NAME-TYPE-AND-VERSION
>>  24: SB-IMPL::PARSE-UNIX-NAMESTRING
>>  25: SB-IMPL::%PARSE-NAMESTRING
>>  26: (SB-C::VARARGS-ENTRY COMMON-LISP::PARSE-NAMESTRING)
>>  27: (SB-C::TL-XEP COMMON-LISP::MAKE-PATHNAME)
>>  28: WEBLOCKS::PUBLIC-FILE-RELATIVE-PATH
>>  29: (SB-C::TL-XEP WEBLOCKS::MAKE-LOCAL-DEPENDENCY)
>>  30: WEBLOCKS::DEPENDENCIES-BY-SYMBOL
>>  31: (COMMON-LISP::LAMBDA (SB-PCL::.PV. SB-PCL::.NEXT-METHOD-CALL. 
>> SB-PCL::.ARG0.))
>>  32: (SB-PCL::FAST-METHOD WEBLOCKS::DEPENDENCIES KEYWORD::AROUND 
>> (COMMON-LISP::T))
> [...]
>> When this happens, SBCL refuses to continue and drops into LDB. I had to 
>> disable LDB so that the application can be automatically restarted.
>> 
>> Reproduced on two different Linux systems. On several SBCL versions.
>> 
>> Any hints appreciated,
> 
> What I would do:
> 
> * try to get more information about the wacky data that is causing
> this.

This could be difficult. It happens unpredictably, it could be months, days, or 
hours before the bug is triggered. The backtrace isn't always as clear as 
above, in fact sometimes I only see a "heap exhausted" message.

It is of course possible that I'm exhausting the heap in some runaway code — 
but if so, how do I track it down?

> * send this to the SBCL list (or file a bug in their Launchpad
> tracker)

I did send it to the SBCL list, but there were no replies.

This is bad.

--J.

-- 
You received this message because you are subscribed to the Google Groups 
"weblocks" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/weblocks?hl=en.

Reply via email to