|
|
|
Oracle 9i RAC & Linux
|
|||
|---|---|---|---|
|
#18+
Есть-ли у кого-нибудь опыт установки? ... |
|||
|
:
Нравится:
Не нравится:
|
|||
| 29.11.2002, 18:01 |
|
||
|
Oracle 9i RAC & Linux
|
|||
|---|---|---|---|
|
#18+
извиняюсь, что ответ не по сути вопроса, но просто интересно чем отличается Oracle 9i RAC от Oracle Parallel Server + OEM? ... |
|||
|
:
Нравится:
Не нравится:
|
|||
| 29.11.2002, 18:12 |
|
||
|
Oracle 9i RAC & Linux
|
|||
|---|---|---|---|
|
#18+
Как удостовериться что Oracle работает как кластер? При входе в OEM (через OMS) показывает только один экземпляр в кластере, а второй отображается вне кластера. :( SRVCTL показывает что все хорошо. Но Intelligent Agent как-то не так берет информацию о втором экземпляре. Подскажите чего-нибудь. Oracle 9.2.0.2.0 Linux RH7.3 ... |
|||
|
:
Нравится:
Не нравится:
|
|||
| 29.11.2002, 18:14 |
|
||
|
Oracle 9i RAC & Linux
|
|||
|---|---|---|---|
|
#18+
nastroy FALOFER-ON opciu na kliente i urony c pomoshu SRVCTL odin uzel posmotry na reakciu. ili propishi v init.ora oba listenera LOCAL_LISTENER= i REMOTE_LISTENER= posle zapuska LSNRCTL poluchi status i servisy > lsnrctl status > lsnrctl services krome togo utilita SRVCTL VSEGDA pokazivaet VSE usly klastera > srvctl status primechanie -- dly obslugi y ne ispolzoval OEM sovsem, tolko rukamy. ... |
|||
|
:
Нравится:
Не нравится:
|
|||
| 29.11.2002, 20:02 |
|
||
|
Oracle 9i RAC & Linux
|
|||
|---|---|---|---|
|
#18+
2 dba: >> извиняюсь, что ответ не по сути вопроса, но просто интересно чем отличается Oracle 9i RAC от Oracle Parallel Server + OEM? Nemnogo naril: =============================================================== BSPn - Block Server Processes (BSPn) Oracle 8i Parallel Server ------------------------------- - For 8i cache fusion, BSP rolls back any uncommitted transactions for any blocks that are being requested for consistent read by the remote instance. - For 8i cache fusion, BSP copies blocks directly from the holding instance's buffer cache and sends a read consistent copy of the block to the requesting foreground process on the requesting instance to be placed into the buffer cache. Oracle 9i Real Application Clusters --------------------------------------------- - Obsolete: In 9i The BSP process is eliminated and all of its previous functionality is now handled by the Lock Manager Server process (LMSn). =============================================================================== DIAG - Diagnosability Daemon (DIAG) Oracle 8i Parallel Server ------------------------------- - DIAG did not exist in 8i Parallel Server Oracle 9i Real Application Clusters --------------------------------------------- - Monitors the health of the instance. - Captures data for instance process failures. =============================================================================== LCKn - Lock Process (LCK0) Oracle 8i Parallel Server ------------------------------- - Parallel Cache Management (PCM) lock ownership is divided among available LCK processes. - Manages instance resource requests and cross-instance call operations for shared resources. - Builds a list of invalid lock elements and validates lock elements during recovery. - Able to use multiple LCK processes. Oracle 9i Real Application Clusters --------------------------------------------- - Much of the lock process functionality is now handled by the Lock Manager Server process (LMSn). - Manages instance resource requests and cross-instance call operations for shared resources. - Builds a list of invalid lock elements and validates lock elements during recovery. - Can only use a single LCK process since primary functionality is handled by LMSn. =============================================================================== LMD - Lock Manager Daemon Process (LMDn) Oracle 8i Parallel Server ------------------------------- - Manages global lock requests from remote instances. - Handles all remote resource requests from remote instances. - Handles global lock deadlock detection. - Monitors for lock conversion timeouts Oracle 9i Real Application Clusters --------------------------------------------- - Much of the lock manager daemon process functionality is now handled by the Lock Manager Server process (LMSn). - Manages lock manager service requests for global cache service (DLM in 8i) resources and sends them to a service queue to be handled by the LMSn process. - Handles global lock deadlock detection. - Monitors for lock conversion timeouts. =============================================================================== LMON - Lock Monitor Process (LMON) Oracle 8i Parallel Server ------------------------------- - Monitors the Clusters to manage - Maintains consistency of DLM memory in case of process death. - Checks for instance death. - Handles reconfiguration of locks and resources when an instance joins or leaves the Clusters. - Listens for local messages. - Generates a trace file that tracks instance reconfigurations. Oracle 9i Real Application Clusters --------------------------------------------- - Maintains consistency of global cache service (DLM in 8i) memory in case of process death. - Checks for instance death. - Handles reconfiguration of locks and resources when an instance joins or leaves the Clusters. - Listens for local messages. - Generates a trace file that tracks instance reconfigurations. =============================================================================== LMSn - Lock Manager Servers Process Oracle 8i Parallel Server ------------------------------- - LMSn did not exist in 8i Parallel Server Oracle 9i Real Application Clusters --------------------------------------------- - Retrieve requests from the server queue queued by LMD to perform requested lock operations. - Can have up to 10 LMSn processes, the number of LMSn processes varies according to the amount of messenging traffic between nodes. - Rolls back any uncommitted transactions for any blocks that are being requested for consistent read by the remote instance. - Copies blocks directly from the holding instance's buffer cache and sends a read consistent copy of the block to the requesting foreground process on the requesting instance to be placed into the buffer cache. - Controls the flow of messages between instances. =============================================================================== ... |
|||
|
:
Нравится:
Не нравится:
|
|||
| 28.12.2002, 00:40 |
|
||
|
Oracle 9i RAC & Linux
|
|||
|---|---|---|---|
|
#18+
>Nemnogo naril: спасибо, интересно. А на практике RAC действительно лучше (в смысле межинстансных блокировок и коллизий) чем 8.1.7 OPS? И какой вообще опыт от использования OPS (RAC) на линухе? Меня прежде всего пугает watchdog, который перезагружает кластер (или нод - не помню) в случае потери межинстанстной синхронизации. Я так и не понял почему только Оракл/Линух (в отличии от остальных Юнихов) требует такой архитектуры. ... |
|||
|
:
Нравится:
Не нравится:
|
|||
| 28.12.2002, 01:52 |
|
||
|
Oracle 9i RAC & Linux
|
|||
|---|---|---|---|
|
#18+
Прсоматривал старые топики и обнаружил сообщения которые не видел В патче 9.2.0.2 Oracle избавился от watchdog и поставил свой самописанный модуль hangcheck-timer (естественно без исходников :) Модуль запускается без ошибок только с ядром RH AS2 После этого все живет без перезагрузок по сбою и инстансы нормально переключаются между собой. ... |
|||
|
:
Нравится:
Не нравится:
|
|||
| 29.01.2003, 17:57 |
|
||
|
Oracle 9i RAC & Linux
|
|||
|---|---|---|---|
|
#18+
2 MW Так ты ж вроде на RH7.3 ставил? ... |
|||
|
:
Нравится:
Не нравится:
|
|||
| 30.01.2003, 14:37 |
|
||
|
|

start [/forum/topic.php?fid=52&msg=32085420&tid=1991939]: |
0ms |
get settings: |
5ms |
get forum list: |
9ms |
check forum access: |
2ms |
check topic access: |
2ms |
track hit: |
163ms |
get topic data: |
6ms |
get forum data: |
2ms |
get page messages: |
26ms |
get tp. blocked users: |
1ms |
| others: | 203ms |
| total: | 419ms |

| 0 / 0 |
