When migrating the database, there was a strange problem.
Run start Mgr in OGG
The interface indicates that MGR is already running, but viewing the MGR process or stop state through the info all command
Then go to the Dirrpt directory to view the MGR process log, found Ogg error message is as follows:
CD Dirrpt
Cat Mgr.rpt
Source Context:
Sourcemodule: [Ggdb.ora.sess]
SourceID: [/SCRATCH/AIME1/ADESTORE/VIEWS/AIME1_ADC4150256/OGGCORE/OPENSYS/SRC/GGLIB/GGDBORA/OCISESS.C]
Sourcefunction: [Oci_try]
Sourceline: [642]
Threadbacktrace: [] Elements
: [/opt/ggs/libgglog.so (Cmessagecontext::addthreadcontext () +0x1e) [0x2adda13be06e]]
: [/opt/ggs/libgglog.so (Cmessagefactory::createmessage (csourcecontext*, unsigned int, ...) +0X2CC) [0x2adda13ba44c]]
: [/opt/ggs/libgglog.so (_msg_err_oracle_oci_error_with_desc (csourcecontext*, int, char const*, Char const*, Cmessagefactory::messa
gedisposition) +0x41) [0x2adda13a68b7]
: [./mgr (Ocisess_context_def::oci_try (int, char const*, ...) +0X4E0) [0x4f2070]]
: [./mgr (Ocisess_context_def::logon (Ggs::gglib::ggapp::cloginname const&, ggs::gglib::ggapp::cdbobjname< ( Dbobjtype) 12> Const&, GGs
:: gglib::ggunicode::ustring const&, int, bool, int, int, int) +0x86f) [0x4f346f]]
: [./mgr (Dboci_init_connection_logon (Ggs::gglib::ggapp::cloginname const&, ggs::gglib::ggapp::cdbobjname< ( Dbobjtype) 12> const&, ch
AR const*, int, int, int, char*) +0xd6) [0x4d8b26]
: [./mgr [0x4ebbf5]]
: [./mgr (Gl_odbc_param (char const*, char const*, char*) +0x4c) [0X4EBF2C]]
: [./mgr (Param_file_args (char*) +0x12d3) [0x4479e3]]
: [./mgr (Init_functions (int, char**) +0x5f0) [0x448030]]
: [./mgr (Main_loop (int, char**) +0x4c) [0x44e13c]]
: [./mgr (MAIN+0XDC) [0X44E8BC]]
: [/lib64/libc.so.6 (__libc_start_main+0xf4) [0x33a801d994]]
: [./mgr (__GXX_PERSONALITY_V0+0X14A) [0x43a49a]]
2016-03-16 00:04:47 Error OGG-00664 OCI error beginning session (status = 1034-ora-01034:oracle not available
Ora-27101:shared Memory realm does not exist
linux-x86_64 error:2: No such file or directory).
2016-03-16 00:04:47 ERROR OGG-01668 PROCESS abending.
It is strange to see this error, the Oracle instance does not start when the error is reported, but my instance has been started, why still report this error.
The search for information through the web was not resolved, and then it was studied by itself.
Check if the Ogg user is unable to identify the instance because of the Ogg user group:
[[email protected] dirrpt]$ ID GGs
uid=1008 (GGs) gid=1000 (oinstall) groups=1000 (Oinstall), 1001 (DBA), 1004 (ASMDBA)
Check the Ogg directory permissions found also no problem.
View MGR Process configuration
Ggsci (QFANGSH2) 2> view params Mgr
Dynamicportlist 7840-7914
PORT 7809
purgeoldextracts/opt/ggs/dirdat/*****/*, usecheckpoints, Minkeepfiles 30
AutoRestart ER *, retries 3, Waitminutes 5
USERID * * *, PASSWORD *******
Purgeddlhistory minkeepdays 3, maxkeepdays 5, Frequencyminutes 30
Purgemarkerhistory minkeepdays 3, maxkeepdays 5, Frequencyminutes 30
Lagreporthours 1
Laginfominutes 30
Lagcriticalminutes 45
There was no other configuration of the database other than the user, and the idea that the Oracle_home variable was incorrectly configured was also out of the question.
So the configuration file does not have the configuration of the relevant parameters, what is Ogg's MGR process to determine the Oracle instance?
Later thought is not through ORACLE_SID environment variable, go to check some environment configuration,
The result is that ORACLE_SID is configured as a SID for another node, modified correctly, and made effective.
New into Ogg client
./ggsci
Start Mgr
Start *
Ggsci (******) 3> info all
Program Status Group lags at chkpt time Since chkpt
MANAGER RUNNING
EXTRACT RUNNING ****** 00:00:09 00:00:05
EXTRACT RUNNING ****** 00:00:00 00:00:02
EXTRACT RUNNING ****** 00:00:00 00:00:00
Replicat RUNNING ****** 00:00:00 00:00:06
Replicat RUNNING ****** 00:00:00 00:00:03
Mgr Process normal start, problem solving
If you have a problem, you can always solve the problem by checking it more.
OGG Mgr Process failed to start problem resolution