这篇文章主要介绍了如何解决修改/dev/shm大小造成Oracle 12c集群启动故障，具有一定借鉴价值，感兴趣的朋友可以参考下，希望大家阅读完这篇文章之后大有收获，下面让小编带着大家一起了解一下。

由于维护人员修改Oracle Linux 7中的/dev/shm大小造成其大小小于Oracle实例的MEMORY_TARGET或者SGA_TARGET而导致集群不能启动(CRS-4535,CRS-4000)

[grid@jtp1~]$crsctlstatres-tCRS-4535:CannotcommunicatewithClusterReadyServicesCRS-4000:CommandStatusfailed,orcompletedwitherrors.

检查asm磁盘的权限是否问题，发现磁盘权限正常

[root@jtp3~]#ls-lrt/dev/asm*brw-rw----.1gridoinstall8,128Apr32018/dev/asmdisk07brw-rw----.1gridoinstall8,48Apr32018/dev/asmdisk02brw-rw----.1gridoinstall8,96Apr32018/dev/asmdisk05brw-rw----.1gridoinstall8,112Apr32018/dev/asmdisk06brw-rw----.1gridoinstall8,64Apr32018/dev/asmdisk03brw-rw----.1gridoinstall8,80Apr32018/dev/asmdisk04brw-rw----.1gridoinstall8,32Apr32018/dev/asmdisk01

重启crs

[root@jtp1bin]#./crsctlstopcrs-fCRS-2791:StartingshutdownofOracleHighAvailabilityServices-managedresourceson'jtp1'CRS-2673:Attemptingtostop'ora.mdnsd'on'jtp1'CRS-2673:Attemptingtostop'ora.gpnpd'on'jtp1'CRS-2677:Stopof'ora.mdnsd'on'jtp1'succeededCRS-2677:Stopof'ora.gpnpd'on'jtp1'succeededCRS-2673:Attemptingtostop'ora.cluster_interconnect.haip'on'jtp1'CRS-2673:Attemptingtostop'ora.drivers.acfs'on'jtp1'CRS-2677:Stopof'ora.drivers.acfs'on'jtp1'succeededCRS-2677:Stopof'ora.cluster_interconnect.haip'on'jtp1'succeededCRS-2673:Attemptingtostop'ora.ctssd'on'jtp1'CRS-2673:Attemptingtostop'ora.evmd'on'jtp1'CRS-2677:Stopof'ora.ctssd'on'jtp1'succeededCRS-2677:Stopof'ora.evmd'on'jtp1'succeededCRS-2673:Attemptingtostop'ora.cssd'on'jtp1'CRS-2677:Stopof'ora.cssd'on'jtp1'succeededCRS-2673:Attemptingtostop'ora.gipcd'on'jtp1'CRS-2673:Attemptingtostop'ora.driver.afd'on'jtp1'CRS-2677:Stopof'ora.driver.afd'on'jtp1'succeededCRS-2677:Stopof'ora.gipcd'on'jtp1'succeededCRS-2793:ShutdownofOracleHighAvailabilityServices-managedresourceson'jtp1'hascompletedCRS-4133:OracleHighAvailabilityServiceshasbeenstopped.[root@jtp1bin]#./crsctlstartcrsCRS-4123:OracleHighAvailabilityServiceshasbeenstarted.

查看crs的alert.log发现磁盘组不能加载

[root@jtp1~]#tail-f/u01/app/grid/diag/crs/jtp1/crs/trace/alert.log2018-04-0218:30:21.227[OHASD(8143)]CRS-8500:OracleClusterwareOHASDprocessisstartingwithoperatingsystemprocessID81432018-04-0218:30:21.230[OHASD(8143)]CRS-0714:OracleClusterwareRelease12.2.0.1.0.2018-04-0218:30:21.245[OHASD(8143)]CRS-2112:TheOLRservicestartedonnodejtp1.2018-04-0218:30:21.262[OHASD(8143)]CRS-8017:location:/etc/oracle/lastgasphas2rebootadvisorylogfiles,0wereannouncedand0errorsoccurred2018-04-0218:30:21.262[OHASD(8143)]CRS-1301:OracleHighAvailabilityServicestartedonnodejtp1.2018-04-0218:30:21.567[ORAROOTAGENT(8214)]CRS-8500:OracleClusterwareORAROOTAGENTprocessisstartingwithoperatingsystemprocessID82142018-04-0218:30:21.600[CSSDAGENT(8231)]CRS-8500:OracleClusterwareCSSDAGENTprocessisstartingwithoperatingsystemprocessID82312018-04-0218:30:21.607[CSSDMONITOR(8241)]CRS-8500:OracleClusterwareCSSDMONITORprocessisstartingwithoperatingsystemprocessID82412018-04-0218:30:21.620[ORAAGENT(8225)]CRS-8500:OracleClusterwareORAAGENTprocessisstartingwithoperatingsystemprocessID82252018-04-0218:30:22.146[ORAAGENT(8316)]CRS-8500:OracleClusterwareORAAGENTprocessisstartingwithoperatingsystemprocessID83162018-04-0218:30:22.211[MDNSD(8335)]CRS-8500:OracleClusterwareMDNSDprocessisstartingwithoperatingsystemprocessID83352018-04-0218:30:22.215[EVMD(8337)]CRS-8500:OracleClusterwareEVMDprocessisstartingwithoperatingsystemprocessID83372018-04-0218:30:23.259[GPNPD(8369)]CRS-8500:OracleClusterwareGPNPDprocessisstartingwithoperatingsystemprocessID83692018-04-0218:30:24.275[GPNPD(8369)]CRS-2328:GPNPDstartedonnodejtp1.2018-04-0218:30:24.283[GIPCD(8433)]CRS-8500:OracleClusterwareGIPCDprocessisstartingwithoperatingsystemprocessID84332018-04-0218:30:26.296[CSSDMONITOR(8464)]CRS-8500:OracleClusterwareCSSDMONITORprocessisstartingwithoperatingsystemprocessID84642018-04-0218:30:28.299[CSSDAGENT(8482)]CRS-8500:OracleClusterwareCSSDAGENTprocessisstartingwithoperatingsystemprocessID84822018-04-0218:30:28.496[OCSSD(8497)]CRS-8500:OracleClusterwareOCSSDprocessisstartingwithoperatingsystemprocessID84972018-04-0218:30:29.538[OCSSD(8497)]CRS-1713:CSSDdaemonisstartedinhubmode2018-04-0218:30:36.015[OCSSD(8497)]CRS-1707:Leaseacquisitionfornodejtp1number1completed2018-04-0218:30:37.087[OCSSD(8497)]CRS-1605:CSSDvotingfileisonline:AFD:CRS1;detailsin/u01/app/grid/diag/crs/jtp1/crs/trace/ocssd.trc.2018-04-0218:30:37.103[OCSSD(8497)]CRS-1672:Thenumberofvotingfilescurrentlyavailable1hasfallentotheminimumnumberofvotingfilesrequired1.2018-04-0218:30:46.237[OCSSD(8497)]CRS-1601:CSSDReconfigurationcomplete.Activenodesarejtp1.2018-04-0218:30:48.514[OCTSSD(9302)]CRS-8500:OracleClusterwareOCTSSDprocessisstartingwithoperatingsystemprocessID93022018-04-0218:30:48.535[OCSSD(8497)]CRS-1720:ClusterSynchronizationServicesdaemon(CSSD)isreadyforoperation.2018-04-0218:30:50.626[OCTSSD(9302)]CRS-2407:ThenewClusterTimeSynchronizationServicereferencenodeishostjtp1.2018-04-0218:30:50.627[OCTSSD(9302)]CRS-2401:TheClusterTimeSynchronizationServicestartedonhostjtp1.2018-04-0218:31:04.202[ORAROOTAGENT(8214)]CRS-5019:AllOCRlocationsareonASMdiskgroups[CRS],andnoneofthesediskgroupsaremounted.Detailsareat"(:CLSN00140:)"in"/u01/app/grid/diag/crs/jtp1/crs/trace/ohasd_orarootagent_root.trc".2018-04-0218:41:00.225[ORAROOTAGENT(8214)]CRS-5818:Abortedcommand'start'forresource'ora.storage'.Detailsat(:CRSAGF00113:){0:9:3}in/u01/app/grid/diag/crs/jtp1/crs/trace/ohasd_orarootagent_root.trc.2018-04-0218:41:03.757[ORAROOTAGENT(8214)]CRS-5017:Theresourceaction"ora.storagestart"encounteredthefollowingerror:2018-04-0218:41:03.757+Storageagentstartactionaborted.Fordetailsreferto"(:CLSN00107:)"in"/u01/app/grid/diag/crs/jtp1/crs/trace/ohasd_orarootagent_root.trc".2018-04-0218:41:03.760[OHASD(8143)]CRS-2757:Command'Start'timedoutwaitingforresponsefromtheresource'ora.storage'.Detailsat(:CRSPE00221:){0:9:3}in/u01/app/grid/diag/crs/jtp1/crs/trace/ohasd.trc.2018-04-0218:42:09.921[ORAROOTAGENT(8214)]CRS-5019:AllOCRlocationsareonASMdiskgroups[CRS],andnoneofthesediskgroupsaremounted.Detailsareat"(:CLSN00140:)"in"/u01/app/grid/diag/crs/jtp1/crs/trace/ohasd_orarootagent_root.trc".

检查跟踪文件，发现查询ASM_DISCOVERY_ADDRESS与ASM_DISCOVERY_ADDRESS属性时出现

[root@jtp1~]#more/u01/app/grid/diag/crs/jtp1/crs/trace/ohasd_orarootagent_root.trcTracefile/u01/app/grid/diag/crs/jtp1/crs/trace/ohasd_orarootagent_root.trcOracleDatabase12cClusterwareRelease12.2.0.1.0-ProductionCopyright1996,2016Oracle.Allrightsreserved.***TRACECONTINUEDFROMFILE/u01/app/grid/diag/crs/jtp1/crs/trace/ohasd_orarootagent_root_93.trc***2018-04-0218:42:09.165:CSSCLNT:3554666240:clsssterm:terminatingcontext(0x7f03c0229390)2018-04-0218:42:09.165:default:3554666240:clsCredDomClose:Credctxdeleted0x7f03c04594702018-04-0218:42:09.166:GPNP:3554666240:clsgpnp_dbmsGetItem_profile:[atclsgpnp_dbms.c:399]Result:(0)CLSGPNP_OK.(:GPNP00401:)gotASM-Profile.Mode='remote'2018-04-0218:42:09.253:CSSCLNT:3554666240:clsssinit:initializedcontext:(0x7f03c045c2c0)flags0x1152018-04-0218:42:09.253:CSSCLNT:3554666240:clsssterm:terminatingcontext(0x7f03c045c2c0)2018-04-0218:42:09.254:CLSNS:3554666240:clsns_SetTraceLevel:tracelevelsetto1.2018-04-0218:42:09.254:GPNP:3554666240:clsgpnp_dbmsGetItem_profile:[atclsgpnp_dbms.c:399]Result:(0)CLSGPNP_OK.(:GPNP00401:)gotASM-Profile.Mode='remote'2018-04-0218:42:09.257:default:3554666240:InitedLSFcontext:0x7f03c04f04202018-04-0218:42:09.260:CLSCRED:3554666240:clsCredCommonInit:Initedsingletoncredctx.2018-04-0218:42:09.260:CLSCRED:3554666240:(:CLSCRED0101:)clsCredDomInitRootDom:Usingusergivenstoragecontextforrepositoryaccess.2018-04-0218:42:09.294:USRTHRD:3554666240:{0:9:3}8033Error4queryinglengthofattrASM_DISCOVERY_ADDRESS2018-04-0218:42:09.300:USRTHRD:3554666240:{0:9:3}8033Error4queryinglengthofattrASM_DISCOVERY_ADDRESS2018-04-0218:42:09.356:CLSCRED:3554666240:(:CLSCRED1079:)clsCredOcrKeyExists:Objdom:SYSTEM.credentials.domains.root.ASM.Self.5c82286a084bcf37ffa014144074e5dd.rootnotfound2018-04-0218:42:09.356:USRTHRD:3554666240:{0:9:3}7755Error4openingdomrootin0x7f03c064c980

检查ASM的alert.log 发现/dev/shm大小小于MEMORY_TARGET大小，并且给出了/dev/shm应该被设置的最小值

[root@jtp1~]#tail-f/u01/app/grid/diag/asm/+asm/+ASM1/trace/alert_+ASM1.logWARNING:ASMdoesnotsupportipclw.SwitchingtoskgxpWARNING:ASMdoesnotsupportipclw.SwitchingtoskgxpWARNING:ASMdoesnotsupportipclw.Switchingtoskgxp*instance_numberobtainedfromCSS=1,checkingfortheexistenceofnode0...*node0doesnotexist.instance_number=1StartingORACLEinstance(normal)(OSid:9343)2018-04-02T18:31:00.187055+08:00CLInotifiernumLatches:7maxDescs:23012018-04-02T18:31:00.193961+08:00WARNING:YouaretryingtousetheMEMORY_TARGETfeature.Thisfeaturerequiresthe/dev/shmfilesystemtobemountedforatleast1140850688bytes./dev/shmiseithernotmountedorismountedwithavailablespacelessthanthissize.PleasefixthissothatMEMORY_TARGETcanworkasexpected.Currentavailableis1073573888andusedis167936bytes.Ensurethatthemountpointis/dev/shmforthisdirectory.

修改/dev/shm的大小可以通过修改/etc/fstab来实现，将/dev/shm的大小修改为12G

[root@jtp1bin]#df-hFilesystemSizeUsedAvailUse%Mountedon/dev/mapper/ol-root49G42G7.9G85%/devtmpfs12G28K12G1%/devtmpfs1.0G164K1.0G1%/dev/shmtmpfs1.0G9.3M1015M1%/runtmpfs1.0G01.0G0%/sys/fs/cgroup/dev/sda11014M141M874M14%/boot[root@jtp1bin]#vi/etc/fstab##/etc/fstab#CreatedbyanacondaonSatMar1815:27:132017##Accessiblefilesystems,byreference,aremaintainedunder'/dev/disk'#Seemanpagesfstab(5),findfs(8),mount(8)and/orblkid(8)formoreinfo#/dev/mapper/ol-root/xfsdefaults00UUID=ca5854cd-0125-4954-a5c4-1ac42c9a0f70/bootxfsdefaults00/dev/mapper/ol-swapswapswapdefaults00tmpfs/dev/shmtmpfsdefaults,size=12G00tmpfs/runtmpfsdefaults,size=12G00tmpfs/sys/fs/cgrouptmpfsdefaults,size=12G00

重启集群后，再次检查集群资源状态恢复正常

--------------------------------------------------------------------------------[grid@jtp1~]$crsctlstatres-t--------------------------------------------------------------------------------NameTargetStateServerStatedetails--------------------------------------------------------------------------------LocalResources--------------------------------------------------------------------------------ora.ASMNET1LSNR_ASM.lsnrONLINEONLINEjtp1STABLEONLINEONLINEjtp2STABLEora.CRS.dgONLINEONLINEjtp1STABLEONLINEONLINEjtp2STABLEora.DATA.dgONLINEONLINEjtp1STABLEONLINEONLINEjtp2STABLEora.FRA.dgONLINEONLINEjtp1STABLEONLINEONLINEjtp2STABLEora.LISTENER.lsnrONLINEONLINEjtp1STABLEONLINEONLINEjtp2STABLEora.TEST.dgONLINEONLINEjtp1STABLEONLINEONLINEjtp2STABLEora.chadONLINEONLINEjtp1STABLEONLINEONLINEjtp2STABLEora.net1.networkONLINEONLINEjtp1STABLEONLINEONLINEjtp2STABLEora.onsONLINEONLINEjtp1STABLEONLINEONLINEjtp2STABLEora.proxy_advmOFFLINEOFFLINEjtp1STABLEOFFLINEOFFLINEjtp2STABLE--------------------------------------------------------------------------------ClusterResources--------------------------------------------------------------------------------ora.LISTENER_SCAN1.lsnr1ONLINEONLINEjtp1STABLEora.LISTENER_SCAN2.lsnr1ONLINEONLINEjtp2STABLEora.LISTENER_SCAN3.lsnr1ONLINEONLINEjtp2STABLEora.MGMTLSNR1ONLINEONLINEjtp2169.254.237.25088.88.88.2,STABLEora.asm1ONLINEONLINEjtp1Started,STABLE2ONLINEONLINEjtp2Started,STABLE3OFFLINEOFFLINESTABLEora.cvu1ONLINEONLINEjtp2STABLEora.jy.db1ONLINEOFFLINESTABLE2ONLINEOFFLINESTABLEora.jtp1.vip1ONLINEONLINEjtp1STABLEora.jtp2.vip1ONLINEONLINEjtp2STABLEora.mgmtdb1ONLINEONLINEjtp2Open,STABLEora.qosmserver1ONLINEONLINEjtp2STABLEora.scan1.vip1ONLINEONLINEjtp1STABLEora.scan2.vip1ONLINEONLINEjtp2STABLEora.scan3.vip1ONLINEONLINEjtp2STABLE--------------------------------------------------------------------------------

到此集群恢复正常

感谢你能够认真阅读完这篇文章，希望小编分享的“如何解决修改/dev/shm大小造成Oracle 12c集群启动故障”这篇文章对大家有帮助，同时也希望大家多多支持亿速云，关注亿速云行业资讯频道，更多相关知识等着你来学习!