Re: Filter Plugin
This is usually indicative that the plugin hasn't loaded / initialised properly, or the plugin alias isn't pointing to the plugin. If you check the plugin log in the usual logging directory (ARserver\Db\arjavaplugin.log) it should tell you if it managed to initialize the plugin and if not why not, there may also be something in arjavaplugin-stderr-date.log that might give you a clue if it's not starting. I'd double check the path to the jar file in pluginsvr_config.xml, also the plugin server port (that is set in pluginsvr_config.xml) has to be the same as the one used in the plugin alias in ar.conf/ar.cfg. Just looking at the plugin config in that setup document, I see plugin nameMII.ARF.REGEXSPLIT/name classnameid.co.mii.filterplugin.SplitRegexFilterPlugin/classname pathelement type=locationC:/Program Files/BMC Software/ARSystem/pluginsvr/regexfilterapi-3.0.jar/pathelement /plugin Which I would expect to have had a few more bits in e.g. plugin nameMII.ARF.REGEXSPLIT/name typeFilterAPI/type codeJAVA/code filenameC:/Program Files/BMC Software/ARSystem/pluginsvr/regexfilterapi-3.0.jar/filename classnameid.co.mii.filterplugin.SplitRegexFilterPlugin/classname /plugin although they may not make any difference. ___ UNSUBSCRIBE or access ARSlist Archives at www.arslist.org attend wwrug12 www.wwrug12.com ARSList: Where the Answers Are
Re: Java API connection pool question..
If the firewall is only closing the inactive connections, you might be able to persuade the mid tier to close them itself. Normally the mid tier keeps some unspecified number of idle connections in the pool for a certain time. In 7.6.4 there are options in the mid-tier config pages to set the number of idle connections that will remain in the pool and optionally the time they will remain there. I haven't tried them to see if they actually work in earlier versions and it's just the config pages that have changed, so YMMV, but the corresponding settings in config.properties for the number of idle connections is: arsystem.pooling_min_connections_per_server=0 And the time in minutes that an idle connection will remain in the pool: arsystem.proxy_idle_time_per_server=0 ___ UNSUBSCRIBE or access ARSlist Archives at www.arslist.org attend wwrug12 www.wwrug12.com ARSList: Where the Answers Are
Re: Remedy System comes to a standstill/unresonsive
I've seen similar symptoms on solaris before, it was resolved by setting RPC-Non-Blocking-IO: T in the ar.conf. I think you might need to ensure a couple of patches are installed as well. ___ UNSUBSCRIBE or access ARSlist Archives at www.arslist.org attend wwrug12 www.wwrug12.com ARSList: Where the Answers Are
Re: Sort of OT: Question for timezone experts out there..
You could use the database session info within your trigger to find out which
server process is generating the deletes. That will at least let you confirm
that it's an arserverd in your server group or not. e.g.
select sys_context('USERENV', 'HOST') from dual;
select process, program from v$session where sid = ( select sid from v$mystat
where rownum=1 );
You'll need access to v$ views for that last bit though.
___
UNSUBSCRIBE or access ARSlist Archives at www.arslist.org
attend wwrug12 www.wwrug12.com ARSList: Where the Answers Are
Re: arsystem.ehcache.maxElementsInMemory - what whould it be set to for full ITSM midtier - box has 12GB memory
There is quite a lot of information about the various parameters in the mid tier guide (in the configuring the mid-tier chapter, cache settings section) maxElementsInMemory is a total count that is actually ignored in favour of referenceMaxElementsInMemory, which seems to be set to 1250 by default and gets multiplied by the various factors to determine the count for each type of object. You'll run into an address space limit if you are on a 32 bit JVM regardless of the amount of RAM in the box (a little under 3GB on windows usually) so you probably aren't going to fit the full ITSM suite in memory unless you are running on a 64bit JVM. I read somewhere (in the BSM tuning guide I think) that as a rule of thumb for every 750MB of additional heap size configured over and above the recommended 1 GB you can increase this value by 1250, so that's probably a good starting point. ___ UNSUBSCRIBE or access ARSlist Archives at www.arslist.org attend wwrug12 www.wwrug12.com ARSList: Where the Answers Are
Re: Performance Tuning
BMC White Paper 85503 Performance Tuning for Business Service Management is a good starting point. ___ UNSUBSCRIBE or access ARSlist Archives at www.arslist.org attend wwrug12 www.wwrug12.com ARSList: Where the Answers Are
Re: SOAP API (Resolved)
SOAP-encoding refers to a specific way of encoding data into XML, so whether or not you can call a specific web service depends on whether or not the web services uses RPC-encoding of arrays and structures. If the web service is document-literal there is no issue. Well, there might be some different issues. If the worst comes to the worst you can insert some sort of adaptation layer to get around any limitations. ___ UNSUBSCRIBE or access ARSlist Archives at www.arslist.org attend wwrug12 www.wwrug12.com ARSList: Where the Answers Are
Re: Java API Help - StructItemInfo
I find a good strategy sometimes is to locate the corresponding item in the C API as it's usually better documented. In this case, it seems these are used when you are importing or exporting VUIs (type = VUI) or mail templates (type = MAIL), and they allow you to specify the specific VUI or mail template you want to import/export for the schema named in the name field. So unless you are dealing with VUIs or mail templates you can probably ignore them. ___ UNSUBSCRIBE or access ARSlist Archives at www.arslist.org attend wwrug12 www.wwrug12.com ARSList: Where the Answers Are
