Package: sg3-utils
Version: 1.21-2.1
I am using Debian etch/amd64 (with backports for postgresql) on a Dell
poweredge 2950.
I have a Perc5 raid card with 6 hard drives :
1 raid10 with 4 hd (/dev/sda) and 1 raid1 with 2 hd (/dev/sdb).
On the bios card, i configure the write policy with "Write Through" (don't use
cache on the raid card).
I start linux and i use MegaCli64 (utils by LSI who make the card) and test the
cache policy :
- get cache policy for virtual drive 0 (/dev/sda) on the first controler:
#./MegaCli64 -LDGetProp -Cache -L0 -a0
Adapter 0-VD 0(target id: 0): Cache Policy:WriteThrough, ReadAdaptive, Direct, No Write Cache if bad BBU
- get cache policy for virtual drive 1 (/dev/sdb) on the first controler:
#./MegaCli64 -LDGetProp -Cache -L1 -a0
Adapter 0-VD 1(target id: 1): Cache Policy:WriteThrough, ReadAdaptive, Direct,
No Write Cache if bad BBU
this results are coherent with my configuration.
Now, i try with sginfo:
#sginfo -c /dev/sda
Caching mode page (0x8)
-----------------------
Initiator Control 0
ABPF 0
CAP 0
DISC 0
SIZE 0
Write Cache Enabled 1
MF 0
Read Cache Disabled 1
Demand Read Retention Priority 0
Demand Write Retention Priority 0
Disable Pre-fetch Transfer Length 0
Minimum Pre-fetch 0
Maximum Pre-fetch 0
Maximum Pre-fetch Ceiling 0
FSW 0
LBCSS 1
DRA 0
NV_DIS 0
Number of Cache Segments 0
Cache Segment size 128
Non-Cache Segment size 0
#sginfo -c /dev/sdb
Caching mode page (0x8)
-----------------------
Initiator Control 0
ABPF 0
CAP 0
DISC 0
SIZE 0
Write Cache Enabled 1
MF 0
Read Cache Disabled 1
Demand Read Retention Priority 0
Demand Write Retention Priority 0
Disable Pre-fetch Transfer Length 0
Minimum Pre-fetch 0
Maximum Pre-fetch 0
Maximum Pre-fetch Ceiling 0
FSW 0
LBCSS 1
DRA 0
NV_DIS 0
Number of Cache Segments 0
Cache Segment size 128
Non-Cache Segment size 0
We show "Write Cache Enabled 1" or we are waiting for "Write Cache Enabled 0"
I try this on 4 Poweredge 2950 with same results.
kernel 2.6.19.7-grsec-drbd, libc6 2.3.6.ds1-13etch2
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
