Automic Workload Automation

 View Only
Back to discussions
Expand all | Collapse all

ASO Worker Processes wont stay up: error in WP log files: No primary work process is registered

  • 1.  ASO Worker Processes wont stay up: error in WP log files: No primary work process is registered

    Posted Mar 01, 2018 09:31 AM
    Oracle DB
    Automic ASO 12.1
    RH Linux AE

    Service Manager looks like this:

    yj4po0cozk7n.pnghttps://us.v-cdn.net/5019921/uploads/editor/y4/yj4po0cozk7n.png" width="640">

    full (but misleading) error that's easy to find in $HOME/AutomationEngine/temp: (WPsrv_log_***_00.txt)
    20180228/141754.574 - U00003367 No primary work process is registered in the database; next attempt in '10' seconds.

    This error (much more helpful) was only seen in the log after trying to start one WP:
    20180228/154530.977 - U00029108 UCUDB: SQL_ERROR    Database handles  DB-HENV: 19cf310  DB-HDBC: 1a45178
    20180228/154530.978 - U00003591 UCUDB - DB error info: OPC: 'OCIStmtExecute' Return code: 'ERROR'
    20180228/154530.978 - U00003592 UCUDB - Status: '' Native error: '1654' Msg: 'ORA-01654: unable to extend index AUTOMIC.NK_AH_REORG by 8 in tablespace UC4_INDEX'
    20180228/154530.978 - U00003536 UCUDB: FATAL DATA BASE ERROR: Re-connection will be attempted in 10 seconds.
    20180228/154531.040 - U00003538 UCUDB: Re-connection to database successful. Processing will continue.
    20180228/154534.041 - U00000006 DEADLOCK
    20180228/154534.041 - U00003594 UCUDB Ret: '6' opcode: 'EXEC' SQL Stmnt: 'UPDATE AH SET AH_TimeStamp4 = ?, AH_Status = ? WHERE AH_OH_Idnr    = ? AND AH_TimeStamp4 is Null'

    This is from not following the install 
    https://docs.automic.com/documentation/webhelp/english/ALL/components/DOCU/12.1/AWA%20Guides/help.htm#Installation_Common/PreparationSteps/Preparingforthemanualinstallation.htm#ORACLE

    y8g8wouf1vot.pnghttps://us.v-cdn.net/5019921/uploads/editor/wl/y8g8wouf1vot.png" width="1204">
    -------------------------------------
    How to fix the issue:

    SQL> @datafiles_all.sql
    15:58:06 SQL> set echo off
    Elapsed: 00:00:03.44
    15:58:09 SQL> alter database datafile '/apps/opt/oracle/oradata/automic/system01.dbf' resize 2G;

    Database altered.

    Elapsed: 00:00:03.18
    15:58:46 SQL> alter database datafile '/apps/opt/oracle/oradata/automic/sysaux01.dbf' resize 4G;

    Database altered.

    Elapsed: 00:00:06.64
    15:59:21 SQL> alter database datafile '/apps/opt/oracle/oradata/automic/UC4_INDEX.dbf' resize 2G;

    Also issued a increase on the DATA tablespace as that had about 3200 meg free...



  • 2.  ASO Worker Processes wont stay up: error in WP log files: No primary work process is registered

    Posted Mar 01, 2018 11:25 AM
    More notes from our DBA: "Root cause in this case appears to be that the oracle tablespaces got full and needed to be expanded.  They were initially set to around one gig in space for a couple of the tables.  I expanded them by about 100% here yesterday.

    Moving forward when I spoke to Mitch a determination of XXXX days/weeks of data will be stored (logs ect) in oracle.  We will track to see how large the tables generally will need to be for the timeframe and make the table spaces generally 20-30% larger than what is needed. just to account for some items.  While Atomic suggests allowing the tablespaces to be unlimited autoextents I suggest NOT to do this for the following reason

    In the event that the tablespaces grow to fill up the drive space the DB itself may crash.  Trying to recover a DB or getting it up when the drive is totally full is a chore and often very difficult to do vs if you set a limit as to how large they can get and you hit the max you can issue a command or two to make the tables larger, get back up and working pretty quickly and then figure out future actions vs its full I have to figure out how to get data out of it (yet cant get the DB up to get data out of it)"