Greetings,
I have a samba server deployed. The machine has these:
cpu: amd64 -4 cores
motherboard :GA-990XA-UD3 with 6 sata ports and
StarTech 2 Port SATA 6Gbps PCIe SATA Card
memory: 16GBytes
disks: raid 10 (6 disks)
os: cblfs/linux kernel-3.1.5 pure 64bit and samba-3.6.0
The machine in deployed as a data-storage-device for windows7-based
machines doing 3d graphics rendering using
3d-StudioMax. There are 20 client machines. The rendering machins do
frequent writes and reads to the
samba server which also runs linux software raid (RAID10). The smb.conf
has these:
[global]
workgroup = WORKGROUP
netbios name = WHATEVER
server string = Samba %v on (%L)
encrypt passwords = True
security = user
smb passwd file = /etc/smbpasswd
log file = /var/log/sambaUSER-log.%m
socket options = IPTOS_LOWDELAY TCP_NODELAY
hosts allow = 192.168.0. 127.
host deny = 0.0.0.0/0
interfaces = eth* lo
bind interfaces only = yes
dos charset = cp850
unix charset = ISO-8859-1
load printers = yes
max log size = 50
local master = yes
printing = cups
guest account = guest
os level = 99
dns proxy = no
wins support = yes
#[ipc$]
# hosts allow = 192.168.0.0/24 127.0.0.1
# hosts deny = 0.0.0.0/0
[homes]
comment = Home Directories
valid users = %S
read only = no
browseable = no
[printers]
comment = All Printers
path = /var/spool/samba
browseable = yes
guest ok = yes
printable = yes
browseable = yes
writable = no
create mask = 0700
public = yes
It is a new setup and the server is crashing daily but only on heavy reads
(20 or so clients reading blocks of
~ 20Gbytes each of data). We do not know the cause but a crash typically
ends with a spew like this:-
#############
[<ffffffff81065120>] kthread_worker_fn+0x190/0x150
---
---
RSP <ffff800415b652f8>
-----[ end trace a1e41a9d77b19dc5 ]-------
md0_raid10 used greatest stack depth 2096 bytes left
#############
could this be due to a cache-memory unavailability to write raid data
before it is transferred to network?? And if so how can this be fixed??
suggestions welcomed
sincerely
lux-integ
-----------------------------------------
---------- Forwarded Message ----------
Subject: Fwd: samba on raid10 on blfs setup crashing frequentl;y -help
required
Date: Thursday 08 March 2012, 15:17:23
From: luxInteg <[email protected]>
To: [email protected]
---------- Forwarded Message ----------
Subject: samba on raid10 on blfs setup crashing frequentl;y -help required
Date: Thursday 08 March 2012, 15:14:30
From: luxInteg <[email protected]>
To: [email protected]
Greetings,
I have a samba server deployed. The machine has these:
cpu: amd64 -4 cores
motherboard :GA-990XA-UD3 with 6 sata ports and
StarTech 2 Port SATA 6Gbps PCIe SATA Card
memory: 16GBytes
disks: raid 10 (6 disks)
os: cblfs/linux kernel-3.1.5 pure 64bit and samba-3.6.0
The machine is deployed as a data-storage-device for windows7-based
machines doing 3d graphics rendering using
3d-StudioMax. There are 20 client machines. The rendering machins do
frequent writes and reads to the
samba server which also runs linux software raid (RAID10). The smb.conf
has these:
[global]
workgroup = WORKGROUP
netbios name = WHATEVER
server string = Samba %v on (%L)
encrypt passwords = True
security = user
smb passwd file = /etc/smbpasswd
log file = /var/log/sambaUSER-log.%m
socket options = IPTOS_LOWDELAY TCP_NODELAY
hosts allow = 192.168.0. 127.
host deny = 0.0.0.0/0
interfaces = eth* lo
bind interfaces only = yes
dos charset = cp850
unix charset = ISO-8859-1
load printers = yes
max log size = 50
local master = yes
printing = cups
guest account = guest
os level = 99
dns proxy = no
wins support = yes
#[ipc$]
# hosts allow = 192.168.0.0/24 127.0.0.1
# hosts deny = 0.0.0.0/0
[homes]
comment = Home Directories
valid users = %S
read only = no
browseable = no
[printers]
comment = All Printers
path = /var/spool/samba
browseable = yes
guest ok = yes
printable = yes
browseable = yes
writable = no
create mask = 0700
public = yes
It is a new setup and the server is crashing daily but only on heavy reads
(20 or so clients reading blocks of
~ 20Gbytes each of data). We do not know the cause but a crash typically
ends with a spew like this:-
#############
[<ffffffff81065120>] kthread_worker_fn+0x190/0x150
---
---
RSP <ffff800415b652f8>
-----[ end trace a1e41a9d77b19dc5 ]-------
md0_raid10 used greatest stack depth 2096 bytes left
#############
could this be due to a cache-memory unavailability to write raid data
before it is transferred to network?? And if so how can this be fixed??
suggestions welcomed
sincerely
lux-integ
--
http://linuxfromscratch.org/mailman/listinfo/blfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page
