samba - Aug 2007 - setresuid failed with EAGAIN, samba crashed

Hello,

I was upgrade our Linux Samba server (Fedora Core 4, kernel 2.6.17,
samba 3.0.23d) to Fedora 7, kernel 2.6.22.2, samba 3.0.25b
(SELinux in permissive mode), i386 Pentium D 3GHz CPU, 2GB RAM,
LAN with approx. <10 users.
But now users smb daemon sporadically (but frequently, 1-5 times/hour)
crashes. Machine log (Level 2) contain records like this:

[2007/08/21 08:53:18, 0] lib/util_sec.c:set_effective_uid(205)
   setresuid failed with EAGAIN. uid(502) might be over its NPROC limit
[2007/08/21 08:53:18, 0] lib/util_sec.c:assert_uid(101)
   Failed to set uid privileges to (-1,502) now set to (0,0)
[2007/08/21 08:53:18, 0] lib/util.c:smb_panic(1654)
   PANIC (pid 13035): failed to set uid

[2007/08/21 08:53:18, 0] lib/util.c:log_stack_trace(1758)
   BACKTRACE: 22 stack frames:
    #0 smbd(log_stack_trace+0x2d) [0x802439fd]
    #1 smbd(smb_panic+0x5d) [0x80243b2d]
    #2 smbd [0x80249f4e]
    #3 smbd [0x800b595c]
    #4 smbd(pop_sec_ctx+0xa2) [0x800b5af2]
    #5 smbd(unbecome_root+0x17) [0x800a9267]
    #6 smbd(uid_to_sid+0x15c) [0x801fbc4c]
    #7 smbd [0x800bc51d]
    #8 smbd(get_nt_acl+0x4ac) [0x800c268c]
    #9 smbd [0x800d887b]
    #10 smbd(is_visible_file+0x2a0) [0x80060ee0]
    #11 smbd [0x8006160d]
    #12 smbd(dptr_ReadDirName+0x51) [0x80061671]
    #13 smbd [0x8009a521]
    #14 smbd [0x8009e935]
    #15 smbd(handle_trans2+0x237) [0x800a2737]
    #16 smbd(reply_trans2+0x6bb) [0x800a8eeb]
    #17 smbd [0x800c54d0]
    #18 smbd(smbd_process+0x836) [0x800c6566]
    #19 smbd(main+0xbdd) [0x8032079d]
    #20 /lib/libc.so.6(__libc_start_main+0xe0) [0x37ff70]
    #21 smbd [0x8004a181]
[2007/08/21 08:53:18, 0] lib/fault.c:dump_core(181)
   dumping core in /var/log/samba/cores/smbd

Share definition in smb.conf:
[eko_rw]
         path = /home/DATA/eko_rw
         read list = @eko_ro
         write list = @eko
         force group = eko
         default case = lower
         short preserve case = no
         create mask = 0660
         force create mode = 0660
         directory mask = 0770
         force directory mode = 0770
         hide unreadable = Yes
         valid users = @eko, @eko_ro

"ulimit -u" say "max user processes" = 32752, this look (to
me)
be enough. I have no idea about solving this error, and all my
documentation and internet searches were unsuccessful.

Thanks for your replies and any help in solving this issue,
Yours
Franta Hanzlik