Re: ZFS kmem_map too small.

Pawel Jakub Dawidek wrote:
> On Tue, Nov 06, 2007 at 10:04:43AM +0100, Borja Marcos wrote:
>> On Nov 6, 2007, at 12:31 AM, Niki Denev wrote:
>> the kernel memory limits to 1.5 GB but it still crashes with the  
>> kmem_too_small panic. I also
>> tried the patch supplied by Pawel.
>>
>> The machine has 6 hard disks configured as a raidz2 pool.
>>
>> Please let me know of any test anyone might need.
> 
> If you use vm_kern.c.2.patch, can you show loader.conf and exact command
> that can provke the panic?
> 


I'd like to mention that I have the same problem - running FreeBSD 
ganymede.dailycal.org 7.0-BETA1 FreeBSD 7.0-BETA1 #0: Tue Oct 30 
12:43:09 PDT 2007 
root_at_ganymede.dailycal.org:/usr/obj/usr/src/sys/GENERIC  i386

on a CPU: Intel(R) Xeon(TM) CPU 2.80GHz (2793.20-MHz 686-class CPU)
   Origin = "GenuineIntel"  Id = 0xf41  Stepping = 1
with 1GB ram.

I also have the vm patch applied.  It helped, but it still panics.

I run a huge rsync from a remote machine to a zpool set up as follows:
[sschlansker_at_ganymede /usr/obj/usr/src/sys/GENERIC]$ sudo zpool status
   pool: dailycal
  state: ONLINE
  scrub: none requested
config:

         NAME        STATE     READ WRITE CKSUM
         dailycal    ONLINE       0     0     0
           raidz2    ONLINE       0     0     0
             da1s1   ONLINE       0     0     0
             da1s2   ONLINE       0     0     0
             ad4     ONLINE       0     0     0
             ad6     ONLINE       0     0     0
             ad8     ONLINE       0     0     0
             ad10    ONLINE       0     0     0

/boot/loader.conf:
vfs.zfs.arc_max="104857600"
vm.kmem_size_max="838860800"
vm.kmem_size="402653184"



The panic is as follows:

Unread portion of the kernel message buffer:
panic: kmem_malloc(131072): kmem_map too small: 358408192 total allocated
cpuid = 0
Uptime: 20h0m56s
Physical memory: 1011 MB
Dumping 430 MB: 415 399 383 367 351 335 319 303 287 271 255 239 223 207 
191 175 159 143 127 111 95 79 63 47 31 15

#0  doadump () at pcpu.h:195
195             __asm __volatile("movl %%fs:0,%0" : "=r" (td));
(kgdb) bt
#0  doadump () at pcpu.h:195
#1  0xc07506a7 in boot (howto=260) at /usr/src/sys/kern/kern_shutdown.c:409
#2  0xc0750969 in panic (fmt=Variable "fmt" is not available.
) at /usr/src/sys/kern/kern_shutdown.c:563
#3  0xc096b9fa in kmem_malloc (map=0xc147108c, size=131072, flags=2) at 
/usr/src/sys/vm/vm_kern.c:312
#4  0xc0962217 in page_alloc (zone=0x0, bytes=131072, pflag=0xe892da33 
"\002", wait=2) at /usr/src/sys/vm/uma_core.c:955
#5  0xc0964ce0 in uma_large_malloc (size=131072, wait=2) at 
/usr/src/sys/vm/uma_core.c:2709
#6  0xc0741d98 in malloc (size=131072, mtp=0xc4405400, flags=2) at 
/usr/src/sys/kern/kern_malloc.c:364
#7  0xc4384a60 in ?? ()
#8  0x00020000 in ?? ()
#9  0xc4405400 in ?? ()
...


It survives a scrub, but the rsync kills it every time.

I'm also willing to help test out patches and whatnot - hope something 
here helps figure this problem out.

Steven