makedumpfile 1.5.0 takes much more time to dump
kumagai-atsushi at mxc.nes.nec.co.jp
Mon Nov 5 22:37:16 EST 2012
On Thu, 25 Oct 2012 05:09:44 -0600
Lisa Mitchell <lisa.mitchell at hp.com> wrote:
> Thanks, Atsushi!
> I tried the dump on the 4 TB system with --cyclic-buffer 131072, and the
> dump completed overnight, and I collected a complete vmcore for dump
> level 31. It looks like from the console log the system "cycled" twice
> with this setting, two passes of excluding and copying, before the dump
> was completed. I am in the process of making a more precise timing
> measurement of the dump time today. Looks like each cycle takes about 1
> hour for this system, with the majority of this time spent in "Excluding
> unnecessary pages" phase of each cycle.
Sorry for my lack of description, the excluding phase will be run twice at
a cycle. So in your case, the number of cycle is 1.
> However if I understand what you are doing with the cyclic-buffer
> parameter, it seems we are taking up 128 MB of the crash kernel memory
> space for this buffer, and it may have to scale larger to get decent
> performance on larger memory.
> Is that conclusion correct?
Yes, but to increase cyclic-buffer is just workaround for v1.5.0.
(Additionally, the enhancement of v1.5.1 is just automation of this.)
I think that the dump time should be constant regardless of the buffer size
ideally, because the purpose of cyclic process is to work on constant
memory, so to increase cyclic-buffer is putting the cart before the horse.
> I was only successful with the new makedumpfile with cyclic-buffer set
> to 128 MB when I set crashkernel=384 MB, but ran out of memory trying to
> start dump (Out of memory killer killed makedumpfile) when
> crashkernel=256 MB, on this system.
> Will we be able to dump larger memory systems, up to 12 TB for instance,
> with any kind of reasonable performance, with a crashkernel size limited
> to 384 MB, as I understand all current upstream kernels are now?
I think v1.5.2 can do it because the most of overhead of cyclic process
will be removed with the optimization of the logic of excluding free pages
in v1.5.2. I expect v1.5.2 to work in constant time regardless of the
number of cycle.
> If the ratio of memory size to total bitmap space is assumed linear,
> this would predict a 12 TB system would take about 6 cycles to dump. And
> larger memory will need even more cycles, etc. I can see where
> performance improvements in getting through each cycle will make this
> better, so more cycles will not mean that much increase in dump time
> over the copy time, but I am concerned for crashkernel size being able
> to stay at 384MB, and still be able to accommodate a large enough
> cyclic-buffer size to maintain a reasonable dump time on future large
> memory systems.
> What other things on a large system will effect usable crashkernel size,
> that will make it insufficient to support a 128 MB cyclic-buffer size?
> Or will the cycle performance fixes proposed for the future makedumpfile
> versions improve things enough that performance penalties for having a
> large number of cycles to dump will be small enough not to matter?
I hope so, but some overhead of cyclic process may be unavoidable and
I can't assume how much time will be spent in them now.
So we need to see the measurement of v1.5.2.
More information about the kexec