Unable to Start OE Management after 11.6.1 Patch

Posted by Jeff Ledbetter on 11-Mar-2016 10:06

Hi.

We are beating our heads against the wall over here after applying the 11.6.1 patch on Linux 64-bit.

Fathom refuses to start so we cannot access via OpenEdge Explorer. 

Any ideas on where to start looking? It usually "just works" and was working until this morning.

All Replies

Posted by James Palmer on 11-Mar-2016 10:21

Was this morning when you installed 11.6.1? Is there anything unusual with the way you've installed 11.6 or is it all default settings?

Posted by James Palmer on 11-Mar-2016 10:22

Oh and I'd be contacting tech support as well.

Posted by Jeff Ledbetter on 11-Mar-2016 10:44

Yes, this morning was the patch installation. No problems starting prior to that.

When trying to start fathom manually, getting an error creating OpenEdge Management configuration database..

Posted by Brian K. Maher on 11-Mar-2016 10:50

Jeff,
 
For my own interest, could you post the [State Info] section from your $DLC/installd.ini file?
 
Brian

Posted by Jeff Ledbetter on 11-Mar-2016 10:54

[State Info]

nPlatformId=43

targetDir=/dlc/11.6

workDir=/dlcwrk/11.6

oemtargetDir=/dlc/oemgmt_11.6

oemworkDir=/dlcwrk/oemgmt_11.6

archiveDir=/dlc/install/11.6

docRootDir=

scriptsDir=

javaHome=/dlc/11.6/jdk

jfcHome=/dlc/11.6/jfc

cpinternal=ISO8859-1

cpcoll=Basic

cpcase=Basic

dateFormat=mdy

numberFormat=

numsep=44

numdec=46

folderName=

defaultLanguage=English - American

defaultCountry=

Error in my admserv.log:

[2016/03/11@10:53:22.607-0600] [0] [Fathom]              * Error starting Fathom probe: "Error loading the activity database: Error opening OpenEdge Management activity database: /dlcwrk/oemgmt_11.6/cachedata/cachedb, Reason: Cluster with id 39 already belongs to class PartitionSummary" (10148)

Posted by Brian K. Maher on 11-Mar-2016 11:02

Thanks.  It looks okay.  I was wondering if the original install was corrupted.

Posted by Libor Laubacher on 11-Mar-2016 11:14

proenv

proadsv -stop

cd /dlcwrk/oemgmt_11.6/cachedata/cachedb/

rm -rf *

proadsv -start

Posted by Jeff Ledbetter on 11-Mar-2016 11:49

Thank you. That seemed to be the magic trick.  What was the deal?

Posted by Matt Baker on 11-Mar-2016 11:58

It is an orientdb bug: github.com/.../3434.

Supposed to have already been fixed a while ago prior to the version we use.

I suppose if you carried over your cache database from pre 11.6, it may have already been triggered?  Impossible to know at this point since you wiped out the database :)

Posted by Brian Bowman on 11-Mar-2016 12:03

FYI - to add to Matt’s comment:
 
The Orient DB is the caching db that we use under the covers to save cache data to for the graphing portion of OE Management and Explorer.
 
Brian
 
Brian L. Bowman
 
Senior Principal Product Manager
Progress Software Corporation
14 Oak Park, Bedford, MA, USA 01730
 
Phone: +1 (603) 801-8259
Email: bowman@progress.com
 
 

This thread is closed