网卡mtu值不同导致rac2节点ASM不能同时启动ORA-27550:TargetIDprotocolcheckfailed.(一)

2015-07-24 09:09:49 · 作者: · 浏览: 0

在安装完10205 rac 后准备配置 asm dg ,发现节点二上的ASM 没法启动:

en12 为private 网卡

en13 为 public 网卡

---->> 查看 oracle asm alert log 报错显示:

LMD0 started with pid=6, OS id=565728
LMS0 started with pid=7, OS id=491812
MMAN started with pid=8, OS id=594200
DBW0 started with pid=9, OS id=651626
LGWR started with pid=10, OS id=692672
CKPT started with pid=11, OS id=668066
SMON started with pid=12, OS id=696756
RBAL started with pid=13, OS id=700854
GMON started with pid=14, OS id=688448
Fri Feb 13 16:07:38 BEIST 2015
Maximum Tranmission Unit (mtu) of the ether adapter is different --------MTU 节点间不一致
on the node running instance 1, and this node.
Ether adapters connecting the cluster nodes must be configured
with identical mtu on all the nodes, for Oracle.
Please ensure the mtu attribute of the ether adapter on all
nodes are identical, before running Oracle.
Fri Feb 13 16:07:38 BEIST 2015
Errors in file /oracle/app/oracle/admin/+ASM/bdump/+asm2_lmon_582048.trc:
ORA-27550: Target ID protocol check failed. tid vers=%d, type=%d, remote instance number=%d, local instance number=%d
LMON: terminating instance due to error 27550
Fri Feb 13 16:07:39 BEIST 2015
System state dump is made for local instance
Fri Feb 13 16:07:39 BEIST 2015
Errors in file /oracle/app/oracle/admin/+ASM/bdump/+asm2_diag_614754.trc:
ORA-27550: Target ID protocol check failed. tid vers=%d, type=%d, remote instance number=%d, local instance number=%d
Fri Feb 13 16:07:39 BEIST 2015
Trace dumping is performing id=[cdmp_20150213160739]
Fri Feb 13 16:07:40 BEIST 2015

----查看节点一网络信息:

[oracle@dg1:/oracle/app/oracle/admin/+ASM/bdump/]$netstat -in

Name Mtu Network Address Ipkts Ierrs Opkts Oerrs Coll
en12 9000 link#2 0.1a.64.bb.82.8 286059 0 57330 3 0
en12 9000 10 10.0.0.1 286059 0 57330 3 0
en13 9000 link#3 0.1a.64.bb.82.9 17267057 0 15456852 24 0
en13 9000 172.31.2 172.31.2.130 17267057 0 15456852 24 0
en13 9000 172.31.2 172.31.2.132 17267057 0 15456852 24 0
lo0 16896 link#1 380344 0 380439 0 0
lo0 16896 127 127.0.0.1 380344 0 380439 0 0
lo0 16896 ::1 380344 0 380439 0 0


----查看节点二网络信息:
[oracle@dg2:/oracle/app/oracle/product/10/db/dbs/]$netstat -in
Name Mtu Network Address Ipkts Ierrs Opkts Oerrs Coll
en12 1500 link#2 0.1a.64.bb.82.64 104657 0 55568 3 0
en12 1500 10 10.0.0.2 104657 0 55568 3 0
en13 9000 link#3 0.1a.64.bb.82.65 1538972 0 761507 21 0
en13 9000 172.31.2 172.31.2.131 1538972 0 761507 21 0
en13 9000 172.31.2 172.31.2.133 1538972 0 761507 21 0
lo0 16896 link#1 112862 0 112928 0 0
lo0 16896 127 127.0.0.1 112862 0 112928 0 0

lo0 16896 ::1 112862 0 112928 0 0

[oracle@dg2:/oracle/app/oracle/product/10/db/dbs/]$ifconfig -a
en12: flags=5e080863,c0
inet 10.0.0.2 netmask 0xffffff00 broadcast 10.0.0.255
tcp_sendspace 131072 tcp_recvspace 65536 rfc1323 0
en13: flags=5e080863,c0
inet 172.31.2.131 netmask 0xffffff00 broadcast 172.31.2.255
inet 172.31.2.133 netmask 0xffffff00 broadcast 172.31.2.255
tcp_sendspace 262144 tcp_recvspace 131072 rfc1323 1
lo0: flags=e08084b
inet 127.0.0.1 netmask 0xff000000 broadcast 127.255.255.255
inet6 ::1/0
tcp_sendspace 131072 tcp_recvspace 131072 rfc1323 1



解决方法:

调整网卡mtu 值。注意需要关闭集群。以及是否启用巨帧。

未完待续。。。。。

~

~

BTW:

MTU
用途: 限制网络上传输的包的大小.
值: 缺省值:configuration-dependent
显示: lsattr -E -l interface_name
更改: chdev -l interface_name -a mtu=NewValue
使用命令 chdev,接口在使用时不能被更改。重新引导后更改才生效。
一种备用的方法如下:
ifconfig interface_name mtu NewValue
这方法在一个运行的系统中更改 MTU 大小,但在系统重启后不保留这个值。


诊断: 包碎片统计.
调整: 增加网络接口的 MTU 的大小。对于 Gigabit Ethernet 适配器