From: zunbeltz <zun...@es...> - 2014-09-09 12:15:19
|
Just to say that I find the nice web browser interface! Finally, I was able to configure everything and it seems running fine. Thanks, Zunbeltz al., 2014.eko iraren 08a 12:57(e)an, zun...@es... igorleak idatzi zuen: > Dear Kay, > > Thanks for the explanation. I configured the broadcast addresses but it > still seems that the engine is not getting data or that is not writting > anything to our DB. Is there any easy way to know which to the two is the > problem? (like redirecting to a text file the data) > > I attach the conifguration file and the log with the debug detail. > > Thanks > > The output on the console is: > kontrol@zboson eclipse]$ ./ArchiveEngine -port 4811 -rdb_url > jdbc:mysql://localhost:3306/archive -rdb_user **** -rdb_password ****** > -engine ****** -pluginCustomization > /home/kontrol/workspace/bin/engine/eclipse/plugin_customization.ini > * 2014-09-08 12:51:50,636 INFO [main] archive.engine.Application (?:?) - > Archive Engine 1.2.3 > 2014-09-08 12:51:50.893:INFO::Logging to STDERR via org.mortbay.log.StdErrLog > 2014-09-08 12:51:50.936:INFO::jetty-6.1.x > 2014-09-08 12:51:50.968:INFO::Started SelectChannelConnector@0.0.0.0:4811 > * 2014-09-08 12:51:50,980 INFO [main] engine.server.EngineServer (?:?) - > Engine HTTP Server port 4811 > * 2014-09-08 12:51:50,981 INFO [main] archive.engine.Application (?:?) - > Reading configuration 'DEMO' > * 2014-09-08 12:51:51,002 INFO [main] > engine.model.MonitoredArchiveChannel (?:?) - BP:LEBT:BD:ACCT1:BeamCurr: No > known last value > * 2014-09-08 12:51:51,038 INFO [main] archive.engine.Application (?:?) - > Read configuration: 1 channels in 0.056 seconds > * 2014-09-08 12:51:51,039 INFO [main] archive.engine.Application (?:?) - > Running, CA addr list: 127.0.0.1 10.20.0.255 10.30.0.255 > * 2014-09-08 12:51:51,041 INFO [WriteThread] engine.model.WriteThread > (?:?) - WriteThread starts > 0 > time for write no wile 0 sample:11 > time method write 0.010581904 > delay: 4990 > * 2014-09-08 12:51:51,207 INFO [ScanThread] engine.scanner.ScanThread > (?:?) - Scan Thread runs > 0 > time for write no wile 0 sample:1 > time method write 6.04778E-4 > delay: 5000 > 0 > time for write no wile 0 sample:1 > time method write 6.78464E-4 > delay: 5000 > 0 > time for write no wile 0 sample:1 > time method write 6.22518E-4 > delay: 5000 > 0 > time for write no wile 0 sample:0 > time method write 6.48291E-4 > delay: 5000 > 0 > time for write no wile 0 sample:0 > time method write 6.65963E-4 > delay: 5000 > > > >> On Sep 4, 2014, at 9:54 AM, zunbeltz <zun...@es...> >> wrote: >>> Thanks for the help Kay, >>> >>> Sorry for my ignorance in network setups but I have a question. >>> In the EPICS_CA_ADDR_LIST i have the IP addresses of S1 and S2. >>> Are addr-broadcast-address-of-subnet1, ... equal to these IPs? >> They're different. >> On a Linux computer, check /sbin/ifconfig. On Windows, check ipconfig. >> Look for the IP address, the net mask, and the broadcast address listed in >> there. >> >> Your IP addresses might be something like >> >> S1=123.45.10.1 >> S2=123.45.20.1 >> >> If your net masks are 255.255.255.0, their subnets would be >> >> Subnet where S1 resides=123.45.10.0 >> Subnet where S2 resides=123.45.20.0 >> >> and the associated broadcast addresses would be >> >> Broadcast for S1's subnet=123.45.10.255 >> Broadcast for S2's subnet=123.45.20.255 >> >> >> Why you would use the IP addresses >> export EPICS_CA_ADDR_LIST="123.45.10.1 123.45.20.1" >> respectively >> org.csstudio.platform.libs.epics/addr_list=123.45.10.1 123.45.20.1 >> ? >> >> IP addresses will work through most routers, and you get PVs from just >> those IOCs, in case you have the same PVs on multiple IOCs. >> You should not have the same channel name on multiple IOCs in the first >> place, so that's not a good reason to list individual IOC IP addresses. >> >> So this might look like the easiest way to get going, but broadcast >> addresses are better in the long run, i.e. >> export EPICS_CA_ADDR_LIST="123.45.10.255 123.45.20.255" >> respectively >> org.csstudio.platform.libs.epics/addr_list=123.45.10.255 123.45.20.255 >> >> In the real world, you don't have 2 IOCs. You might have 500, and you >> can't list the IP address of each single IOC. >> Even if you only have 5 IOCs. One day you'll add one more. Do you then >> want to track down every CA client and update its CA address list? >> And if you run more than one IOC on the same computer, you MUST use >> broadcast addresses, because otherwise you will only see the PVs from the >> IOC that was started _last_, see again >> https://wiki-ext.aps.anl.gov/epics/index.php/How_to_Configure_Channel_Access >> The only downside: You may need to configure the affected routers to pass >> those broadcasts. >> >> ===> In the real world, your clients will eventually either use broadcast >> addresses, or the single IP address of a CA gateway (and that gateway then >> uses broadcast addresses). >> >> -Kay >> >> >> ------------------------------------------------------------------------------ >> Want excitement? >> Manually upgrade your production database. >> When you want reliability, choose Perforce >> Perforce version control. Predictably reliable. >> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk >> >> >> _______________________________________________ >> Cs-studio-users mailing list >> Cs-...@li... >> https://lists.sourceforge.net/lists/listinfo/cs-studio-users -- Dr. Zunbeltz Izaola Accelerator Physics Group ESS-Bilbao |