云数仓复合节点缩容为纯gcware节点缩容后gcware服务起不来的问题：
#集群初始状态
[gbase@wf-openssh8-node-0 gcinstall]$ gcadmin
CLUSTER STATE: ACTIVE
====================================
| GBASE GCWARE CLUSTER INFORMATION |
====================================
| NodeName | IpAddress | gcware |
------------------------------------
| gcware1 | 10.10.11.146 | OPEN |
------------------------------------
| gcware2 | 10.10.11.231 | OPEN |
------------------------------------
======================================================
| GBASE COORDINATOR CLUSTER INFORMATION
|
======================================================
| NodeName | IpAddress |
gcluster | DataState |
------------------------------------------------------
| coordinator1 | 10.10.11.146 | OPEN |
0 |
------------------------------------------------------
| coordinator2 | 10.10.11.231 | OPEN |
0 |
------------------------------------------------------
================================================
| GBASE CLUSTER FREE DATA NODE INFORMATION |
================================================
| NodeName | IpAddress | gnode | DataState |
------------------------------------------------
| FreeNode1 | 10.10.11.231 | OPEN | 0
|
------------------------------------------------
| FreeNode2 | 10.10.11.146 | OPEN | 0
|
------------------------------------------------
#缩掉10.10.11.231节点的coor和gnode服务
[gbase@wf-openssh8-node-0 gcinstall]$ cat demo.options
installPrefix= /opt
coordinateHost = 10.10.11.231
#coordinateHostNodeID = 234,235,237
dataHost = 10.10.11.231
existCoordinateHost =10.10.11.146
existDataHost =10.10.11.146
existGcwareHost=10.10.11.146,10.10.11.231
gcwareHost =
#gcwareHostNodeID = 234,235,237
dbaUser = gbase
dbaGroup = gbase
dbaPwd = 'gbase'
gcluster_instance_name=liurui
instance_root_name=root
instance_root_password=111111
rootPwd = '111111'
#rootPwdFile = rootPwd.json
#characterSet = utf8
#dbPort = 5258
#sshPort = 22
# 's3' or 'hdfs'
GCDW_STORAGE_STYLE = s3
GCDW_S3_BUCKET=liurui3
GCDW_S3_ENDPOINT=10.10.15.103:9000
GCDW_S3_ACCESS_KEY_ID=CXP0NQCOJVGWRPMF6K3O
GCDW_S3_SECRET_KEY=dj+CwfwFVZPmjWWCqJ1+ua+coB1DAMLeLt4zueJg
GCDW_S3_REGION=minio
GCDW_HDFS_NAMENODES=192.168.151.100:50070,192.168.151.101:50070
GCDW_HDFS_URI=hdp://gbase@192.168.151.100:50070/
#执行缩容命令，启动10.10.11.231节点的gcware服务失败
[gbase@wf-openssh8-node-0 gcinstall]$ ./unInstall.py
--silent=demo.options
These GCluster nodes will be uninstalled.
CoordinateHost:
10.10.11.231
DataHost:
10.10.11.231
Are you sure to uninstall GCluster ([Y,y]/[N,n])? y
delete coordinator nodes message from foundationdb
delete coordinator nodes message from foundationdb successfully
unInstall 10.10.11.231 's coordinate successfully.
unInstall 10.10.11.231 's data successfully.
Update all cluster common conf.
10.10.11.146 update cluster common conf successfully.
Update all gcware conf.
10.10.11.231 update gcware conf successfully.
10.10.11.146 update gcware conf successfully.
Starting all gcluster nodes ...
start gcware service failed on host 10.10.11.231.
231013 14:25:52 [GCWARE] connect to 10.10.11.231 error:connect
10.10.11.231:5919 error, Connection refused
231013 14:25:54 [GCWARE] connect to 10.10.11.231
error:connect 10.10.11.231:5919 error, Connection refused
231013 14:25:56 [GCWARE] connect to 10.10.11.231
error:connect 10.10.11.231:5919 error, Connection refused
231013 14:25:58 [GCWARE] connect to 10.10.11.231
error:connect 10.10.11.231:5919 error, Connection refused
#10.10.11.231节点查看服务
[gbase@liurui-node-0 sbin]$ gcware_monit.sh status
+---------------------------------------------------------------------------------------------------------------------------+
|SEG_NAME
PROG_NAME
STATUS
PID |
+---------------------------------------------------------------------------------------------------------------------------+
|gcware
gcware
Stoped
|
|gcware_mmonit
gcware_mmonit
Running 8394
|
+---------------------------------------------------------------------------------------------------------------------------+
[gbase@liurui-node-0 sbin]$ pwd
/opt/10.10.11.231/gcware/sbin
[gbase@liurui-node-0 sbin]$ ll
总用量 6364
-rwxr-xr-x 1 gbase gbase 6352120 10月 12 18:08 gcware
-rwxr-xr-x 1 gbase gbase 56856 10月 12 18:08
gcware_mmonit
-rwxr-xr-x 1 gbase gbase 66752 10月 12 18:08
gcware_monit
-rwxr-xr-x 1 gbase gbase 11631 10月 12 18:08
gcware_monit.sh
-rwxr-xr-x 1 gbase gbase 16963 10月 12 18:08
gcware_services
-rwxr-xr-x 1 gbase gbase 188 10月 12 18:08 gcwexec
#执行./gcware start后，发现报错libstdc++.so.6文件不存在
[gbase@liurui-node-0 sbin]$ ./gcware start
./gcware: error while loading shared libraries:
libstdc++.so.6: cannot open shared object file: No such file or directory
[gbase@liurui-node-0 sbin]$ ldd gcware
linux-vdso.so.1 =>
(0x00007fff18f8f000)
libcrypto.so.10 =>
/lib64/libcrypto.so.10 (0x00007f25702ea000)
libz.so.1 => /lib64/libz.so.1 (0x00007f25700d4000)
librt.so.1 => /lib64/librt.so.1
(0x00007f256fecc000)
libpthread.so.0 =>
/lib64/libpthread.so.0 (0x00007f256fcb0000)
libdl.so.2 => /lib64/libdl.so.2
(0x00007f256faac000)
libstdc++.so.6
=> not found
libm.so.6 => /lib64/libm.so.6
(0x00007f256f7aa000)
libgcc_s.so.1 => /lib64/libgcc_s.so.1
(0x00007f256f594000)
libc.so.6 => /lib64/libc.so.6
(0x00007f256f1c6000)
/lib64/ld-linux-x86-64.so.2
(0x00007f257074d000)
[gbase@刘瑞-node-0 ~]$ env
|grep gcware
LD_LIBRARY_PATH=/opt/jarvis/lib:$:/opt/10.10.11.231/gcware/libexec:/opt/10.10.11.231/gcware/libexec
PATH=/opt/10.10.11.231/gcware/sbin:/opt/10.10.11.231/gcware/sbin:/opt/jarvis/bin:/usr/local/bin:/bin:/usr/bin:/usr/local/sbin:/usr/sbin:/home/gbase/.local/bin:/home/gbase/bin
GCWARE_BASE=/opt/10.10.11.231/gcware
#查看系统环境是否存在libstdc++.so.6库
[gbase@刘瑞-node-0 sbin]$ ll
/lib64/libstdc++.so.6
lrwxrwxrwx 1 root
root 55 11月 15 10:27 /lib64/libstdc++.so.6
->/opt/10.10.11.231/gnode/server/lib/gbase/libstdc++.so.6
#该节点环境之前安装云数仓asan版本时将系统的libstdc++.so.6库重新链接到/opt/10.10.11.231/gnode/server/lib/gbase/libstdc++.so.6，缩容后/opt/10.10.11.231/gnode/server/lib/gbase/libstdc++.so.6路径文件不存在导致报错
解决：将 /lib64/libstdc++.so.6 重新链接到libstdc++.so.6.0.19
ln -s /lib64/libstdc++.so.6.0.19 /lib64/libstdc++.so.6
