do_IRQ stack overflow on 32-bit CentOS 4
Just hit a:
do_IRQ: stack overflow: 452
on 32-bit CentOS 4 machine with xfs-kmod-0.4-2.el4. Is there anything that can be done to minimize this risk? Running xfs over md raid5 array, exporting via nfs. More kernel dump messages are posted at:
http://bugs.centos.org/view.php?id=3136
Thanks!
_______________________________________________
xfs mailing list
[email protected]
http://oss.sgi.com/mailman/listinfo/xfs
Quoted 9 lines · Orion Poplawski
Just hit a:
do_IRQ: stack overflow: 452on 32-bit CentOS 4 machine with xfs-kmod-0.4-2.el4. Is there anything that can be done to minimize this risk? Running xfs over md raid5 array, exporting via nfs. More kernel dump messages are posted at:
get an x86_64 machine ;)
The stack overflow message is misleading; it means you have 452 -left-
unfortunately on some kernels, simply printing that message actually consumes what's left, and more. See http://lkml.org/lkml/2007/8/29/245
This was fixed in 2.6.18-108.el5 and beyond.
Upstream has some xfs stack size fixes; this may eventually make it to a centos near you.
-Eric
_______________________________________________
xfs mailing list
[email protected]
http://oss.sgi.com/mailman/listinfo/xfs
Quoted 11 lines · Eric Sandeen
Quoted 9 lines · Orion Poplawski
Just hit a:
do_IRQ: stack overflow: 452on 32-bit CentOS 4 machine with xfs-kmod-0.4-2.el4. Is there anything that can be done to minimize this risk? Running xfs over md raid5 array, exporting via nfs. More kernel dump messages are posted at:
Quoted 8 lines
get an
x86_64machine ;)The stack overflow message is misleading; it means you have 452 -left-
unfortunately on some kernels, simply printing that message actually consumes what's left, and more. See http://lkml.org/lkml/2007/8/29/245
This was fixed in 2.6.18-108.el5 and beyond.
Oh, for RHEL^wCENTOS4 it was fixed in 2.6.9-78.8.EL and beyond.
-Eric
_______________________________________________
xfs mailing list
[email protected]
http://oss.sgi.com/mailman/listinfo/xfs