지송하지만.. linux 용 MCserviceGuard 질문 드립니다.

[유택남]

장비 - rx4620
OS - Redhat Enterprise AS 3
cluster pkg - serviceguard-A.11.16.090.product.redhat.ia64.rpm

우선 cluster volume으로 VolGroup02,03,04,05를 사용하고 있었습니다.
시스템 재부팅 후 VolGroup05/LogVol00 mount 되지 않습니다. 당연히 pkg fail 나구요.. fail시 log입니다.

##### Node "ocs01": Starting package at Wed Feb 24 20:53:56 KST 2010 ######
Incorrect metadata area header checksum
Incorrect metadata area header checksum
Found duplicate PV I9iKOJKw36h6CfnHlK6L9B4k7czW7cQA: using /dev/emcpowerc1 not /dev/sda1
Incorrect metadata area header checksum
Found duplicate PV cY8vjhYnEvjmYBrK6UxGF1pn1uEqPyJs: using /dev/sdd1 not /dev/emcpowerb1
Found duplicate PV wZzpXbyPmA0l3VpNR7g2Y6V95It9CMQM: using /dev/emcpowerd1 not /dev/sdb1
Found duplicate PV 5mbXQyXxwM20qy955hWSgCshqpHDOh4q: using /dev/sde1 not /dev/emcpowera1
Incorrect metadata area header checksum
Found duplicate PV i1avX8CMz9munXaBUpFaDqTVFaAxX5wI: using /dev/emcpowere1 not /dev/sdc1
Found duplicate PV I9iKOJKw36h6CfnHlK6L9B4k7czW7cQA: using /dev/sdf1 not /dev/emcpowerc1
Incorrect metadata area header checksum
Found duplicate PV wZzpXbyPmA0l3VpNR7g2Y6V95It9CMQM: using /dev/sdg1 not /dev/emcpowerd1
Found duplicate PV i1avX8CMz9munXaBUpFaDqTVFaAxX5wI: using /dev/sdh1 not /dev/emcpowere1
Found duplicate PV cY8vjhYnEvjmYBrK6UxGF1pn1uEqPyJs: using /dev/sdi1 not /dev/sdd1
Found duplicate PV 5mbXQyXxwM20qy955hWSgCshqpHDOh4q: using /dev/sdj1 not /dev/sde1
Incorrect metadata area header checksum
Feb 24 20:53:56 - Node "ocs01": Activating volume group VolGroup02 .
Incorrect metadata area header checksum
Incorrect metadata area header checksum
Found duplicate PV I9iKOJKw36h6CfnHlK6L9B4k7czW7cQA: using /dev/emcpowerc1 not /dev/sda1
Incorrect metadata area header checksum
Found duplicate PV cY8vjhYnEvjmYBrK6UxGF1pn1uEqPyJs: using /dev/sdd1 not /dev/emcpowerb1
Found duplicate PV wZzpXbyPmA0l3VpNR7g2Y6V95It9CMQM: using /dev/emcpowerd1 not /dev/sdb1
Found duplicate PV 5mbXQyXxwM20qy955hWSgCshqpHDOh4q: using /dev/sde1 not /dev/emcpowera1
Incorrect metadata area header checksum
Found duplicate PV i1avX8CMz9munXaBUpFaDqTVFaAxX5wI: using /dev/emcpowere1 not /dev/sdc1
Found duplicate PV I9iKOJKw36h6CfnHlK6L9B4k7czW7cQA: using /dev/sdf1 not /dev/emcpowerc1
Incorrect metadata area header checksum
Found duplicate PV wZzpXbyPmA0l3VpNR7g2Y6V95It9CMQM: using /dev/sdg1 not /dev/emcpowerd1
Found duplicate PV i1avX8CMz9munXaBUpFaDqTVFaAxX5wI: using /dev/sdh1 not /dev/emcpowere1
Found duplicate PV cY8vjhYnEvjmYBrK6UxGF1pn1uEqPyJs: using /dev/sdi1 not /dev/sdd1
Found duplicate PV 5mbXQyXxwM20qy955hWSgCshqpHDOh4q: using /dev/sdj1 not /dev/sde1
Incorrect metadata area header checksum
Feb 24 20:53:56 - Node "ocs01": Activating volume group VolGroup03 .
Incorrect metadata area header checksum
Incorrect metadata area header checksum
Found duplicate PV I9iKOJKw36h6CfnHlK6L9B4k7czW7cQA: using /dev/emcpowerc1 not /dev/sda1
Incorrect metadata area header checksum
Found duplicate PV cY8vjhYnEvjmYBrK6UxGF1pn1uEqPyJs: using /dev/sdd1 not /dev/emcpowerb1
Found duplicate PV wZzpXbyPmA0l3VpNR7g2Y6V95It9CMQM: using /dev/emcpowerd1 not /dev/sdb1
Found duplicate PV 5mbXQyXxwM20qy955hWSgCshqpHDOh4q: using /dev/sde1 not /dev/emcpowera1
Incorrect metadata area header checksum
Found duplicate PV i1avX8CMz9munXaBUpFaDqTVFaAxX5wI: using /dev/emcpowere1 not /dev/sdc1
Found duplicate PV I9iKOJKw36h6CfnHlK6L9B4k7czW7cQA: using /dev/sdf1 not /dev/emcpowerc1
Incorrect metadata area header checksum
Found duplicate PV wZzpXbyPmA0l3VpNR7g2Y6V95It9CMQM: using /dev/sdg1 not /dev/emcpowerd1
Found duplicate PV i1avX8CMz9munXaBUpFaDqTVFaAxX5wI: using /dev/sdh1 not /dev/emcpowere1
Found duplicate PV cY8vjhYnEvjmYBrK6UxGF1pn1uEqPyJs: using /dev/sdi1 not /dev/sdd1
Found duplicate PV 5mbXQyXxwM20qy955hWSgCshqpHDOh4q: using /dev/sdj1 not /dev/sde1
Incorrect metadata area header checksum
Feb 24 20:53:56 - Node "ocs01": Activating volume group VolGroup04 .
Incorrect metadata area header checksum
Incorrect metadata area header checksum
Found duplicate PV I9iKOJKw36h6CfnHlK6L9B4k7czW7cQA: using /dev/emcpowerc1 not /dev/sda1
Incorrect metadata area header checksum
Found duplicate PV cY8vjhYnEvjmYBrK6UxGF1pn1uEqPyJs: using /dev/sdd1 not /dev/emcpowerb1
Found duplicate PV wZzpXbyPmA0l3VpNR7g2Y6V95It9CMQM: using /dev/emcpowerd1 not /dev/sdb1
Found duplicate PV 5mbXQyXxwM20qy955hWSgCshqpHDOh4q: using /dev/sde1 not /dev/emcpowera1
Incorrect metadata area header checksum
Found duplicate PV i1avX8CMz9munXaBUpFaDqTVFaAxX5wI: using /dev/emcpowere1 not /dev/sdc1
Found duplicate PV I9iKOJKw36h6CfnHlK6L9B4k7czW7cQA: using /dev/sdf1 not /dev/emcpowerc1
Incorrect metadata area header checksum
Found duplicate PV wZzpXbyPmA0l3VpNR7g2Y6V95It9CMQM: using /dev/sdg1 not /dev/emcpowerd1
Found duplicate PV i1avX8CMz9munXaBUpFaDqTVFaAxX5wI: using /dev/sdh1 not /dev/emcpowere1
Found duplicate PV cY8vjhYnEvjmYBrK6UxGF1pn1uEqPyJs: using /dev/sdi1 not /dev/sdd1
Found duplicate PV 5mbXQyXxwM20qy955hWSgCshqpHDOh4q: using /dev/sdj1 not /dev/sde1
Incorrect metadata area header checksum
ERROR: Function get_md; Failed to RUN vgcfgrestore
###### Node "ocs01": Package start failed at Wed Feb 24 20:53:56 KST 2010 ######

VolGroup05/LogVol00 이 다행히도 backup이라서 우선
/usr/local/cmcluster/conf/pkg1/pkg1.sh에서
VolGroup05부분 #으로 처리해서 서비스 가동 중입니다.
/backup을 다시 구성해서 mount 해보려 했으나 똑같은 error message 가 떨어지네요..

다시 구성 시 작업 내용입니다. (node1)
fdisk /dev/sdj로 partition 삭제 후 재부팅.
재부팅 후 fdisk /dev/sdj partition 생성 후 type를
8e (LVM)로 설정
pvcreate /dev/sdj1
vgcreate VolGroup05 /dev/sdj1
lvcreate -l size -n LogVol00 VolGroup05
mkfs -t ext3 /dev/VolGroup05/LogVol00
mount /dev/mapper/VolGroup05-Logvol00 /backup
mount 확인 및 umount
vgchange -a n VolGroup05

node2 reboot
node2 /dev/VolGroup05생성 확인
cmruncl -v
node 들은 run되지만 pkg가 되지 않음.
VolGroup05빼면 잘됨.

어떻게 하면 좋을까요??? 도움 부탁 드립니다.

[양계전]

안녕하세요~

리눅스에서는 약간 서비스가드 방식이 다릅니다

volumn을 생성하신후에 추가로 해줘야할 작업이 있습니다

보니깐 양쪽에 만들어놓은 /dev/VolGroup05/Logvol00 이 잘보이는거 같은데요

node A 에서

vgchange --deltag $(uname -n) VolGroup05

하신후

cluster pkg에 VolGroup05 와 Logvol 주석을 제거시킨후에

cmruncl 함 해보십시요.

[유택남]

답변 감사 드립니다.
우선 어제 작업을 다시 해봤습니다.
아래는 작업 내용입니다.
----------------------------------------------
node 1
vcreate -f /dev/sdj1
vgcreate /dev/VolGroup05 /dev/sdj1
lvcreate -l 32673 -n LogVol00 VolGroup05
mke2fs -j /dev/VolGroup05/LogVol00
mount -t ext3 /dev/VolGroup05/LogVol00 /backup
umount /backup
vgchange -a n VolGroup05
vgchange --deltag $(uname -n) VolGroup05

node2 reboot
vgscan
vgchange --addtag $(uname -n) VolGroup05
vgchange -a y VolGroup05
mount /dev/VolGroup05/LogVol00 /backup
umount /backup
vgchange -a n VolGroup05
vgchange --deltag $(uname -n) VolGroup05

node1
cmruncl -v

-------------------------------------------------
위와 같이 작업을 하였으나 같은 error message가 발생됩
니다. 우선 최초 장애시 pkg1관련 backup 볼륨이 보이지
않았습니다.
정상적으로 vgremove를 할 수 없었기에 reboot 후
fdisk로 새로 구성 후 위와 같은 작업을 한고 있는 상황입니다. 기존의 VolGroup05 정보가 남아있어서 그런건지
아니면 제가 cluster VolGroup 작업을 잘못 한건지 모르겠
습니다.

그럼 좋은 하루 되세요.. ^^