導致 Scan VIP 和 Scan Listener(監聽程式)出現故障的最常見的 5 個問題

mosdoc發表於2016-12-13

導致 Scan VIP 和 Scan Listener(監聽程式)出現故障的最常見的 5 個問題 (文件 ID 1602038.1)
文件內容


用途

適用範圍

詳細資訊
  問題 1:安裝Patch 23273629後SCAN Listener 或 local listener無法啟動
 
  問題 2:間歇性VIP, SCAN VIP failover和VIP failover後Listener停止
  問題 3:SCAN 連線有關問題
  問題 4:在 SCAN listener 發生故障切換(failover)後,服務未註冊到 SCAN listener
  問題 5:SCAN有關的安裝問題
  問題 6:SCAN VIP 或 SCAN LISTENER是OFFLINE / UNKNOWN狀態

參考


適用於:

Oracle Database - Enterprise Edition - 版本 11.2.0.1 和更高版本
本文件所含資訊適用於所有平臺

用途

本文簡要總結了與 SCAN VIP 和 SCAN LISTENERS 有關的常見問題。

有關SCAN的更多資訊,請參考Note 887522.1 Grid Infrastructure Single Client Access Name (SCAN) Explained

適用範圍

適用於11gR2 Grid Infrastructure.

詳細資訊

問題 1:安裝Patch 23273629後SCAN Listener 或 local listener無法啟動

請具體參考:

note 2166451.1 - SCAN Listener or local listener fails to start after applying Patch 23273629 - Oracle Grid Infrastructure Patch Set Update 12.1.0.2.160719 (Jul2016)

問題 2:間歇性VIP, SCAN VIP failover和VIP failover後Listener停止

間歇性VIP, SCAN VIP failover和Listener停止導致的連線問題.

$ crsctl stat res -t
--------------------------------------------------------------------------------
NAME           TARGET  STATE        SERVER                   STATE_DETAILS      
--------------------------------------------------------------------------------

ora.LISTENER.lsnr
               ONLINE  ONLINE          rac2
ora.rac1.vip 
            1 ONLINE INTERMEDIATE rac2          FAILED OVER
ora.rac2.vip
            1 ONLINE  ONLINE          rac2           
ora.scan2.vip 
            1 ONLINE   ONLINE         rac2
ora.scan2.vip 
            1 ONLINE   ONLINE         rac2
ora.scan3.vip 
            1 ONLINE   ONLINE         rac2

原因:

1. Unpublished Bug 12680491 Intermittent hiccup in network CHECK action can fail over vip, bring listener offline briefly, 請參考 Note 12680491.8
2. Bug 13582411 RAC ONE NODE: SCAN_LISTENER GO OFF LINE AFTER NETWORK HICCUP

 

解決方案:
補丁安裝後,需要再修改VIP 和 SCAN VIP 的STOP_DEPENDENCIES屬性,使補丁生效。

問題 3:SCAN 連線有關問題

透過SCAN 連線資料庫時可能遇到下面的錯誤

Intermittent ORA-12541 TNS: No Listener
ORA-12514: TNS:listener does not currently know of service requested in connect descriptor
ORA-12516: TNS:listener could not find available handler with matching protocol Stack
TNS-12561: TNS:unknown error
TNS-01101: Could not find service name for LISTENER_SCANx
ORA-12537: TNS:connection closed
TNS-12518: TNS:listener could not hand off client connection

原因:

1. 本地LISTENER沒有啟動。
2. 叢集以外的伺服器設定remote_listener指向叢集SCAN(有可能出現在DataGuard 備庫)
3. tnsnames.ora沒有正確定義"<SCAN>:<PORT>"
4. 本地LISTENER在RDBMS ORACLE_HOME/network/admin/tnsnames.ora中定義,並且手動啟動。
5. COST問題
6. lsnrctl status <listener name> 在錯誤的ORACLE_HOME(需要在GRID_HOME下執行)或者錯誤的節點下執行。
7. 錯誤ORACLE_HOME設定。
8. oracle許可權問題,丟失了setuid 和 setgid許可權。
9. 由於Bug 15874565

 

解決方案:

請參考下面的Note的解決方案:
1. How to Troubleshoot Connectivity Issue with 11gR2 SCAN Name Note 975457.1
2. Intermittent ORA-12541 when Connecting via SCAN listener in RAC Note 1547757.1
3. TNS-12514 Unable to connect to RAC database using SCAN name Note 1459768.1
4. Connect Via SCAN Intermittently Fails With ORA-12516: TNS:listener Could Not Find Available Handler With Matching Protocol Stack Note 1638175.1
5. SCAN: Following (COST) Implementation Services Fail to Register to SCAN Listeners TNS-12561 Note 1624154.1
6. SCAN: Using Lsnrctl To Check Scan Listener Status Yields Error TNS-12541 TNS: No Listener Note 1666358.1
7. TNS-01101: Could not find service name for LISTENER_SCANx Note 1064628.1
8. TNS-12518 Can Not Connect To RAC 11gR2 Database Note 1088754.1
9. Duplicate VIP or SCAN VIP or SCAN Listener Found on More than 1 Node in a RAC Cluster Note 1575272.1

問題 4:在 SCAN listener 發生故障切換(failover)後,服務未註冊到 SCAN listener

在執行 SCAN VIP 和 SCAN listener故障切換後,例項未註冊到 SCAN listener。這種情況只會發生在其中1 個 scan listener上。客戶機連線間歇性出現“ORA-12514 TNS:listener does not currently know of service requested in connect descriptor”。

原因:

1. 未釋出的 Bug 12659561:在執行 scan listener故障切換後,資料庫例項可能未註冊到 scan listener(請參閱 Note 12659561.8),這一問題已在 11.2.0.3.2 中修復,針對 11.2.0.2 的 Merge patch13354057 適用於特定平臺。
2. 未釋出的 Bug 13066936:在執行 scan 故障切換時,例項未註冊服務(請參閱 Note 13066936.8)。

 

解決方案:

1) 對於以上兩個 Bug,解決方法是執行以下步驟,在未註冊到 SCAN listener的資料庫例項上登出並重新註冊remote_listener。
show parameter remote_listener
alter system set remote_listener='';
alter system register;
alter system set remote_listener='<scan>:<port>';
alter system register;

2) 服務未註冊到 SCAN listener(監聽程式)時要檢查的其他要點:
a. 正確定義了 remote_listener 和 local_listener
b. sqlnet.ora 中定義了 EZCONNECT,示例:NAMES.DIRECTORY_PATH= (TNSNAMES, EZCONNECT)
c. 定義3個SCAN IP,那麼不能在/etc/hosts中定義,應該在DNS中定義。
d. 多次執行nslookup <scan> 應以 round-robin (迴圈)方式顯示 SCAN VIP
e. 如果未配置 Secure transports (COST) 的類,則不要在 listener.ora 中設定 SECURE_REGITER_<listener>。

問題 5:SCAN有關的安裝問題

安裝Grid Infrastructure時,有時會出現下面的錯誤:


INS-40719 Scan Already Assigned To Another System
INS-41105: subnet chosen as public does not match subnet of SCAN VIP
PRVG-1101 : SCAN name "cluscan.us.oracle.com" failed to resolve
INFO: PRVF-4664 : Found inconsistent name resolution entries for SCAN name "cluscan.us.oracle.com"
INFO: PRVF-4657 : Name resolution setup check for "scanclunm" (IP address: 10.4.0.202) failed
PRCS-1016 : Failed to resolve Single Client Access Name cluscan.us.oracle.com

原因
1. 分配不正確的SCAN IP.
2. 伺服器網路配置問題.
3. DNS SCAN 名字定義問題.

 

解決方案:

請參考下面的Note的解決方案:
1. Installing 11gR2 Grid Infrastructure Fails with INS-40719 Scan Already Assigned To Another SystemNote 955636.1
2. INS-41105: subnet chosen as public does not match subnet of SCAN VIP Note 1540546.1
3. PRVF-4664 PRVF-4657: Found inconsistent name resolution entries for SCAN name Note 887471.1

問題 6:SCAN VIP 或 SCAN LISTENER是OFFLINE / UNKNOWN狀態

crsctl stat res -t shows

ora.scan1.vip  1     ONLINE      UNKNOWN      racnode2          CHECK TIMED OUT 
or 
ora.scan1.vip  1     ONLINE         OFFLINE       racnode2

原因
1. 網路配置問題
2. Bug 14179376 - SCAN VIP GOES OFFLINE INTERMITTENTLY
3. Solaris IPMP 有關問題
4. AIX LHEA 有關問題
5. Windows Bug 13371153 - Windows: SCAN VIP fail ping after relocate even with new arp driver (patch 10124639)
6. Bug 14259185 - WINDOWS: vip and scan vips fail back not pingable in win2008r2

 

解決方案:

請參考下面的Note的解決方案:
1. SCAN VIP Showing Status UNKNOWN - CHECK TIMED OUT On One Node Note 1341882.1
2. SCAN VIP GOES OFFLINE INTERMITTENTLY Note 1544475.1
3. Solaris: 11gR2 VIP / SCAN VIP and Dependent Resources Offline after Active Public NIC in IPMP Group Fails Note 1069254.1
4. AIX: VIP and SCAN VIP fails to failover to other node after pulled cable on public network if LHEA is being used Note 1530943.1
5. RAC on Windows: Relocate SCAN (and node VIP) Not Working In Timely Manner (Bug:13371153/10124639) Note 1423313.1
6. WINDOWS: vip and scan vips fail back not pingable in win2008r2 Note 14259185.8

 

Database - RAC/Scalability 社群
為了與 Oracle 專家和業內同行進一步討論這個話題,我們建議您加入 My Oracle Support 的 Database - RAC/Scalability 社群參與討論。

參考

NOTE:1379498.1 - 11.2.0.3 VIP/SCAN VIP is Not Pingable After Failover Leads to Connection Issue
NOTE:1423313.1 - RAC on Windows: Relocate SCAN (and node VIP) Not Working In Timely Manner (Bug:13371153/10124639)
NOTE:1056713.1 - TOP Note: Solutions for Typical Grid Infrastructure/RAC Database runInstaller Issues
NOTE:1496510.1 - SCAN listener port changed to default 1521 After upgrade to 11.2.0.3 GI from lower release
NOTE:1107295.1 - Linux: How to Configure the DNS Server for 11gR2 SCAN
NOTE:462933.1 - What is TCP_VALIDNODE_CHECKING and How to Use It
- OCE: MULTIPLE PUBLIC NETWORK, PULL PRIMARY PUBLIC CABLE, VIP DOES NOT FAILOVER

來自 “ ITPUB部落格 ” ,連結:http://blog.itpub.net/31393455/viewspace-2130377/,如需轉載,請註明出處,否則將追究法律責任。

相關文章