Re: [Condor-users] 7.0.3 Debian 4 dynamic clipped?

[Richard Palmer <richard.d.palmer@xxxxxxxxx>]

Hi Daniel,

> > setarch doesn't seem to exist on Debian, what is it doing ?.
> 
> It basically calls "personality()" with flags based on the command
> line arguments and then "execs" the program.
> 
> $ gcc -Wall -o setarch setarch.c
> 
> $ ./setarch ./simple.std -_condor_D_ALL
> 
> $ ./setarch ./simple.std -_condor_D_ALL -_condor_restart simple.std.ckpt

and it now works!:

condor-vm-3:/home/condor$ ./setarch ./simple.std -_condor_D_ALL
-_condor_restart simple.std.ckpt 
User Job - $CondorPlatform: X86_64-LINUX_DEBIAN40 $
User Job - $CondorVersion: 7.1.1 Jul 23 2008 $
Condor: Notice: Will restart from simple.std.ckpt
Read headers OK
Read SegMap[0](DATA) OK
Read SegMap[1](STACK) OK
Read all SegMaps OK
Restoring a DATA segment
Found a DATA block, increasing heap from 0x6b2000 to 0x6b3000
About to overwrite 696320 bytes starting at 0x609000(DATA)
About to execute on TmpStk
About to execute on tmpstack.
Beginning Execution on TmpStack.
RestoreStack() Entrance!
Restoring a STACK segment
About to overwrite 40959 bytes starting at 0x7fffffff5000(STACK)
RestoreStack() Exit!
About to restore file state
CondorFileTable::resume
working dir = /home/condor

OPEN FILE TABLE:
working dir = /home/condor

OPEN FILE TABLE:
fd 0
	logical name: default stdin
	offset:       0
	dups:         1
	open flags:   0x0
	not currently bound to a url.
fd 1
	logical name: default stdout
	offset:       67
	dups:         1
	open flags:   0x1
	not currently bound to a url.
fd 2
	logical name: default stderr
	offset:       0
	dups:         1
	open flags:   0x1
	not currently bound to a url.
Done restoring file state
About to restore signal state
About to return to user code
bout to return to user code
We calculated: 8

> A similar test:
> 
> $ diff <(./setarch cat /proc/self/maps) <(./setarch cat /proc/self/maps)
> 
> Should show no differences (other than a sometimes different stack
> size, but always with the same top stack address).

I'll have to read up on memory management...

So the simple sleep() programme seems to work, do you know of any other
programmes I could run to test how stable the checkpointing is ? (or is
this the time to get writing my own code?!).

thanks for all your help,

Richard.

-- 
Richard Palmer
Systems Administration Officer / Centre for E-Research
King's College London          / Centre for Computing in the Humanities
Tel: 0207 848 1973