Hi.

Latelly I'm using samba to mount two fileshares from a w2k pro box. Made a
small script to make the smbmounts. All works fine, when the script is run
by root, manually. The mount or df commands report the w2k shares
correctly mounted.

The problem I'm going to describe occured with the original version of samba
that came along with RH4. So, the linux box is a RH 4, linux version 2.0.27.
Int this box, I upgraded samba to the latest Samba version 1.9.16p9, but the
problem still persists. The w2k has the SP2 installed.

The two boxes are 24h online, 7 days a week. There are no shutdowns or
restarts.

The problem:
Everyday, by the morning, the two w2k fileshares are unmounted in samba.
There are no problems reported in /var/log/messages, /var/log/samba,
/var/log/samba.nmb and /var/log/samba.smb that could be related to this
problem. In this state, when I issue a df command, the linux reports two
"smb_statfs: dskatr error = 5" lines.

In /var/log/messages there is everyday a registered problem, related to the
linux box scsi bus, when the tape reaches the end of a dayly copy. The scsi
bus gets a reset, but I don't see how that could be related to the samba
mounts. The filesystem where the shares are mounted is not a scsi
filesystem, it's a IDE one.

As I didn't found out why the samba mounts get unmounted, I tried to make an
automatic mount of the w2k shares, by using cron, inserting a call to my
script, made by root, in /etc/crontab, at 21:59, everyday. The script
actually is run, but this is a source of problems: every login process, when
finish, becomes a zombie, sync hangs, a init 0 does nothing. The linux
machine has to be taken down by cutting it's power. It's a mess. In
/var/log/messages there is then plenty of error info:

when my script was run:
Sep 9 21:59:02 linux kernel: smb_dont_catch_keepalive: server->data_ready
== NULL

when a job tried do write on the w2k samba share:
Sep 9 22:12:14 linux kernel: smb_dont_catch_keepalive: server->data_ready
== NULL
Sep 9 22:12:14 linux kernel: smb_dont_catch_keepalive: server->data_ready
== NULL

when another job tried to write on the other w2k samba share:
Sep 9 22:25:27 linux kernel: Unable to handle kernel paging request at
virtual address c3ab15ac
Sep 9 22:25:27 linux kernel: current->tss.cr3 = 01f6b000, 8r3 = 01f6b000
Sep 9 22:25:27 linux kernel: *pde = 00001067
Sep 9 22:25:27 linux kernel: *pte = 00000000
Sep 9 22:25:27 linux kernel: Oops: 0000
Sep 9 22:25:27 linux kernel: CPU: 0
Sep 9 22:25:27 linux kernel: EIP: 0010:[<03ab15ac>]
Sep 9 22:25:27 linux kernel: EFLAGS: 00010293
Sep 9 22:25:27 linux kernel: eax: ffffff92 ebx: 0240b018 ecx: 0193ebcc
edx: 0000006e
Sep 9 22:25:27 linux kernel: esi: 00000fe6 edi: 0240b018 ebp: 001ef9c8
esp: 0193ebf4
Sep 9 22:25:27 linux kernel: ds: 0018 es: 0018 fs: 002b gs: 002b
ss: 0018
Sep 9 22:25:27 linux kernel: Process smbmount (pid: 26740, process nr: 51,
stackpage=0193e000)
Sep 9 22:25:27 linux kernel: Stack: 0119f088 0240b144 00000010 00000000
0193ecec 03ab028f 0240b018 0240b018
Sep 9 22:25:27 linux kernel: 0193eee8 0240b2c0 001ef9c8 00002000
01354000 01d9a000 001ee208 00118d73
Sep 9 22:25:27 linux kernel: 00118df7 0203d12c 0244d067 00118c70
0804b174 022c0b98 0223e414 022c0b98
Sep 9 22:25:27 linux kernel: Call Trace: [<03ab028f>] [do_no_page+367/776]
[do_no_page+499/776] [do_no_page+108/776] [lcall7+0/84]
[do_page_fault+8/712] [mem_init+400/556]
Sep 9 22:25:27 linux kernel: [lcall7+0/84] [dcache_add+307/404]
[<03ab3677>] [<03ab368e>] [<03ab36a6>] [<03ab36bd>] [<03ab36c7>]
[<03ab36d1>]
Sep 9 22:25:27 linux kernel: [<03ab36dc>] [<03ab0d8e>] [<03ab1e5d>]
[<03ab3e3c>] [read_super+205/208] [do_mount+255/304] [<03ab3ba5>]
[do_mount_root+58/444]
Sep 9 22:25:27 linux kernel: [<03ab3ba5>] [<03ab3ba5>]
[shm_swap_in+586/692] [do_signal+273/628]
Sep 9 22:25:27 linux kernel: Code: <1>Unable to handle kernel paging
request at virtual address c3ab15ac
Sep 9 22:25:27 linux kernel: current->tss.cr3 = 01f6b000, 8r3 = 01f6b000
Sep 9 22:25:27 linux kernel: *pde = 00001067
Sep 9 22:25:27 linux kernel: *pte = 00000000
Sep 9 22:25:27 linux kernel: Oops: 0000
Sep 9 22:25:27 linux kernel: CPU: 0
Sep 9 22:25:27 linux kernel: EIP: 0010:[die_if_kernel+80/684]
Sep 9 22:25:27 linux kernel: EFLAGS: 00010216
Sep 9 22:25:27 linux kernel: eax: 00000010 ebx: 0000002b ecx: 03ab15ac
edx: 025ebc0c
Sep 9 22:25:27 linux kernel: esi: 00000000 edi: 0193f000 ebp: 0193ebb8
esp: 0193eb5c
Sep 9 22:25:27 linux kernel: ds: 0018 es: 0018 fs: 0010 gs: 002b
ss: 0018
Sep 9 22:25:27 linux kernel: Process smbmount (pid: 26740, process nr: 51,
stackpage=0193e000)
Sep 9 22:25:27 linux kernel: Stack: 0000002b 00000000 002b1000 0193ebb8
0223e414 03800000 04000000 03800000
Sep 9 22:25:27 linux kernel: 02230018 0010ff5a 001b295b 0193ebb8
00000000 0010fca0 00000fe6 0240b018
Sep 9 22:25:27 linux kernel: 001ef9c8 0293d414 022c0b98 0000012c
0010a738 0193ebb8 00000000 0240b018
Sep 9 22:25:27 linux kernel: Call Trace: [ne:ne_probe+-2686980/5676]
[<04000000>] [ne:ne_probe+-2686980/5676] [do_page_fault+422/712]
[mem_init+400/556] [lcall7+0/84] [<03ab15ac>]
Sep 9 22:25:27 linux kernel: [<03ab028f>] [do_no_page+367/776]
[do_no_page+499/776] [do_no_page+108/776] [lcall7+0/84]
[do_page_fault+8/712] [mem_init+400/556] [lcall7+0/84]
Sep 9 22:25:27 linux kernel: [dcache_add+307/404] [<03ab3677>]
[<03ab368e>] [<03ab36a6>] [<03ab36bd>] [<03ab36c7>] [<03ab36d1>]
[<03ab36dc>]
Sep 9 22:25:27 linux kernel: [<03ab0d8e>] [<03ab1e5d>] [<03ab3e3c>]
[read_super+205/208] [do_mount+255/304] [<03ab3ba5>] [do_mount_root+58/444]
[<03ab3ba5>]
Sep 9 22:25:27 linux kernel: [<03ab3ba5>] [shm_swap_in+586/692]
[do_signal+273/628]
Sep 9 22:25:27 linux kernel: Code: 64 8a 04 0e 0f a1 88 c2 81 e2 ff 00 00
00 89 54 24 10 52 68


Well, of course I want to keep using samba. By now, I have to run my script,
manually, everyday, if I want to have the shares online. But that is
annoying.

Any thougts or clues would be much appreciated.

Cheers, L.A.