scsi read error 문의 입니다.

[김봉현_1]

Apr 30 02:43:41 NULL vmunix: SCSI: Read error -- dev: b 31 0x041000, errno: 126, resid: 2048,

Apr 30 02:43:41 NULL vmunix: blkno: 8, sectno: 16, offset: 8192, bcount: 2048.

Apr 30 02:43:41 NULL vmunix: LVM: vg: pvnum=0 (dev_t=0x1f061000) is POWERFAILED

Apr 30 02:43:46 NULL vmunix: LVM: Recovered Path (device 0x1f041000) to PV 0 in VG 7.

Apr 30 02:43:46 NULL vmunix: LVM: Performed a switch for Lun ID = 0 (pv = 0x000000004c77e040),

from raw device 0x1f061000 (with priority: 1, and current flags: 0xc0) to raw device 0x1f041000

(with priority: 0, and current flags: 0x80).

Apr 30 02:43:46 NULL vmunix: LVM: Restored PV 0 to VG 7.

Apr 30 02:43:41 NULL vmunix: LVM: vg: pvnum=0 (dev_t=0x1f061000) is POWERFAILED

Apr 30 02:43:48 NULL vmunix: LVM: Recovered Path (device 0x1f061000) to PV 0 in VG 7.

Apr 30 02:44:52 NULL vmunix: LVM: Recovered Path (device 0x1f041000) to PV 0 in VG 7.

Apr 30 02:44:53 NULL vmunix: LVM: Restored PV 0 to VG 7.

Apr 30 02:45:48 NULL vmunix: LVM: Recovered Path (device 0x1f041000) to PV 0 in VG 7.

Apr 30 02:45:50 NULL vmunix: LVM: Restored PV 0 to VG 7.

Apr 30 02:47:00 NULL vmunix: LVM: Performed a switch for Lun ID = 0 (pv = 0x000000004c77e040),

from raw device 0x1f041000 (with priority: 0, and current flags: 0x40) to raw device 0x1f061000

(with priority: 1, and current flags: 0x0).

Apr 30 02:47:14 NULL vmunix: LVM: Performed a switch for Lun ID = 0 (pv = 0x000000004c77e040),

from raw device 0x1f061000 (with priority: 1, and current flags: 0x0) to raw device 0x1f041000

(with priority: 0, and current flags: 0x0).

Apr 30 02:47:00 NULL vmunix: LVM: Restored PV 0 to VG 7.





NULL:>ioscan -fnC disk

Class I H/W Path Driver S/W State H/W Type Description

========================================================================

disk 3 0/3/0/0.8.0.110.0.0.0 sdisk CLAIMED DEVICE HP A6189B

/dev/dsk/c4t0d0 /dev/rdsk/c4t0d0

disk 5 0/3/0/0.8.0.110.0.0.1 sdisk CLAIMED DEVICE HP A6189B

/dev/dsk/c4t0d1 /dev/rdsk/c4t0d1

disk 7 0/3/0/0.8.0.110.0.0.2 sdisk CLAIMED DEVICE HP A6189B

/dev/dsk/c4t0d2 /dev/rdsk/c4t0d2

disk 9 0/3/0/0.8.0.110.0.0.3 sdisk CLAIMED DEVICE HP A6189B

/dev/dsk/c4t0d3 /dev/rdsk/c4t0d3

disk 11 0/3/0/0.8.0.110.0.0.4 sdisk CLAIMED DEVICE HP A6189B

/dev/dsk/c4t0d4 /dev/rdsk/c4t0d4

disk 13 0/3/0/0.8.0.110.0.0.5 sdisk CLAIMED DEVICE HP A6189B

/dev/dsk/c4t0d5 /dev/rdsk/c4t0d5

disk 15 0/3/0/0.8.0.110.0.0.6 sdisk CLAIMED DEVICE HP A6189B

/dev/dsk/c4t0d6 /dev/rdsk/c4t0d6

disk 17 0/3/0/0.8.0.110.0.1.0 sdisk CLAIMED DEVICE HP A6189B

/dev/dsk/c4t1d0 /dev/rdsk/c4t1d0

disk 4 0/9/0/0.8.0.108.0.0.0 sdisk CLAIMED DEVICE HP A6189B

/dev/dsk/c6t0d0 /dev/rdsk/c6t0d0

disk 6 0/9/0/0.8.0.108.0.0.1 sdisk CLAIMED DEVICE HP A6189B

/dev/dsk/c6t0d1 /dev/rdsk/c6t0d1

disk 8 0/9/0/0.8.0.108.0.0.2 sdisk CLAIMED DEVICE HP A6189B

/dev/dsk/c6t0d2 /dev/rdsk/c6t0d2

disk 10 0/9/0/0.8.0.108.0.0.3 sdisk CLAIMED DEVICE HP A6189B

/dev/dsk/c6t0d3 /dev/rdsk/c6t0d3

disk 12 0/9/0/0.8.0.108.0.0.4 sdisk CLAIMED DEVICE HP A6189B

/dev/dsk/c6t0d4 /dev/rdsk/c6t0d4

disk 14 0/9/0/0.8.0.108.0.0.5 sdisk CLAIMED DEVICE HP A6189B

/dev/dsk/c6t0d5 /dev/rdsk/c6t0d5

disk 16 0/9/0/0.8.0.108.0.0.6 sdisk CLAIMED DEVICE HP A6189B

/dev/dsk/c6t0d6 /dev/rdsk/c6t0d6

disk 18 0/9/0/0.8.0.108.0.1.0 sdisk CLAIMED DEVICE HP A6189B

/dev/dsk/c6t1d0 /dev/rdsk/c6t1d0



NULL:>strings /etc/lvmtab

/dev/vg00

/dev/dsk/c2t0d0

/dev/dsk/c1t0d0

/dev/vg01

/dev/dsk/c4t0d1

/dev/dsk/c6t0d1

/dev/vg03

/dev/dsk/c6t0d3

/dev/dsk/c4t0d3

/dev/vg04

/dev/dsk/c6t0d4

/dev/dsk/c4t0d4

/dev/vg05

/dev/dsk/c4t0d5

/dev/dsk/c6t0d5

/dev/vg07

/dev/dsk/c4t1d0

/dev/dsk/c6t1d0

/dev/vg02

/dev/dsk/c4t0d2

/dev/dsk/c6t0d2

/dev/vg06

/dev/dsk/c6t0d6

/dev/dsk/c4t0d6

---------------------------------------------------------

위 메시지가 glance로 봤을때 dsik i/o 가 100%에 도달하면 주시적으로 뿌려집니다.

가르침 좀 부탁드리겠습니다.

장비는 Rp5470 OS는 11.11 array는 va7400 입니다.

[monoworld]

안녕하세요??

이 문제는 OS 상으로 봤을때는 병목 현상 관련된 문제 인거 같고요.

VG7400성능을 초과하는 I/O가 발생해서 disk 쪽에서 응답을 줄수 없어 발생한것 같습니다.



에러 로그 발생시간대에

1. 해당 VG(vg7)사용 App확인

2. glance등을 이용한 에러 발생시 사용률체크('u'누른후 Qlen 확인)--> Qlen이 5개 이상이면 병목 의심.

3. 채널 분산등을 통한 병목 해소...

이렇게 문제 해결에 접근하면 될거 같고요.



추가적으로 아래 정보 확인 하여 VA상태 확인해보세요.

갑자기 성능 저하 생겼다면 va disk등 H/W fail 발생 했을

가능성 있습니다. 로그 확인 후 이상있으면 올려 주세요.



1. va 정보 확인 (VA7410 의 VA_ID 및 va_alias name 확인)

#armdsp -i

2. va 구성 확인 (현재 va7410 LUN 정보 및 리던던시 그룹 확인)

#armdsp -a va_alias_name

[이태곤]

안녕하십니까



VA74x0은 lvol구성시 stripe을 하는것이 성능이슈가 발생하지 않습니다.

stripe을 하지않으면 전체디스크의 반만 I/O을 발생시켜 위와 같은 문제가 발생할수 있습니다.

기회가 되시면 stripe구성을 점검해보십시오



그런 사항이 안되시면 임시 방편으로 아래 명령어를 사용하여 PATH failover을 조금이나바 줄이시면 됩니다.



# pvchange -t 180 /dev/dsk/c4t1d0

[김봉현_1]

깔끔한 답변 감사합니다. 보면 LUN이 300G 400G 이상씩 나눠져 있고 LUN 통채로 Lvol 하나로 구성 되어있습니다..

[이태곤]

VA74x0에 대해 추가 설명을 드리면



vg07이 300g의 용량이 필요하면 RG1에서 150G RG2에서 150G을 만들어 합치고 lvol을 만들때 stripe을 하시면 됩니다.