Re: Archiver Appliance PVs never connect to be archived, always "Initial Sampling"

["Manoussakis, Adamandios via Tech-talk" <tech-talk at aps.anl.gov>]

Great thanks for your help Jeong.  Currently I am just trying to set it up to be able to see how the archiver works, not for production at this moment.

The status returns that all Tomcat instances have started and I am able to access the web app front end

    16 Oct 28 11:31:44 pc-pc systemd[1]: Starting EPICS Archiver Appliance for als...

    17 Oct 28 11:31:44 pc-pc bash[9241]: Tomcat started.

    18 Oct 28 11:31:44 pc-pc bash[9241]: >>> Archiver Appliance Service mgmt started

    19 Oct 28 11:31:44 pc-pc bash[9241]: Tomcat started.

    20 Oct 28 11:31:44 pc-pc bash[9241]: >>> Archiver Appliance Service engine started

    21 Oct 28 11:31:44 pc-pc bash[9241]: Tomcat started.

    22 Oct 28 11:31:44 pc-pc bash[9241]: >>> Archiver Appliance Service etl started

    23 Oct 28 11:31:44 pc-pc bash[9241]: Tomcat started.

    24 Oct 28 11:31:44 pc-pc bash[9241]: >>> Archiver Appliance Service retrieval started

    25 Oct 28 11:31:44 pc-pc systemd[1]: Started EPICS Archiver Appliance for als.

Its all localhost so there should be no issue with firewall I dont think. 

I do not have selinux installed on my ubuntu 18.04

the mysql db seems to be working properly see below, archappl has all privilieges

MariaDB [mysql]> SELECT user FROM user;

+------------------+

| user             |

+------------------+

| admin            |

| archappl         |

| debian-sys-maint |

| root             |

+------------------+

make sql.show

>>    1/   1/   4<<                        ArchivePVRequests

>>    2/   2/   4<<                      ExternalDataServers

>>    3/   3/   4<<                                PVAliases

>>    4/   4/   4<<                               PVTypeInfo

I did look in the log files in /opt/epicsarchiverap/mgmt/logs/catalina.out catalina 2020-10-27.log archapply_service.log localhost_access_log but I did not see any errors in the log just debug/info messages.  One thing that I was wondering was these statements about the PVs found in the retrieval logs but not sure what they mean exactly.

41878 [hz.client_0.event-5] DEBUG org.epics.archiverappliance.config.DefaultConfigService  - Skipping publishing event into this JVM ComputeMetaInfo/appliance0/appliance0_ENGINE/ca://calc:a as destination is not me RETRIEVAL

41878 [hz.client_0.event-5] DEBUG org.epics.archiverappliance.config.DefaultConfigService  - Skipping publishing event into this JVM ComputeMetaInfo/appliance0/appliance0_ENGINE/calc:a as destination is not me RETRIEVAL

---

From: Jeong Han Lee <citadel.lee at gmail.com>
Sent: Wednesday, October 28, 2020 11:10 AM
To: Manoussakis, Adamandios <manoussakis1 at llnl.gov>
Cc: tech-talk <tech-talk at aps.anl.gov>
Subject: Re: Archiver Appliance PVs never connect to be archived, always "Initial Sampling"

 

Hi,

I have an obligation to reply to this email. I will come back to you
in more detail later.

Meanwhile, you may check the following things:
1) firewall
2) selinux
3) mysql db
4) several log files will be located in mgmt,etl,engine,retrieval/logs

Ant this repository was designed to help any users to set it easily,
and it is not for production. Technically, it doesn't matter, but for
the production you should consider many configurations in great details.

HTH,
Han


On Tue, Oct 27, 2020 at 11:45 PM Manoussakis, Adamandios via Tech-talk
<tech-talk at aps.anl.gov> wrote:
>
> Hello All,
>
> I have started to build the epics archiver appliance following Hans repo at https://github.com/jeonghanlee/epicsarchiverap-env.  Everything seems to be working great with the application and I am able to access the archiver web application just fine.  I start up a basic softIoc with a couple PVs and my pyDM widget is able to control and access them just fine but I cannot get them to be archived by the appliance.  I can also use caget/camonitor/caput without any issue to manipulate the softIoc.   I saw another post about checking the ports in the appliances.xml but that seems to be correct from what I can tell.  I tried to look through the logs in /opt/epicsarchiverap/engine mgmt etl retrieval but I didnt see any errors that stood out.  I am not sure if its a connection issue with the ca and the archiver or a record thing I am missing but any advice would be helpful.  I have tried using ca://calc:a , calc:a but no luck getting it to archive.
>
> appliances.xml
>
> <appliances>
>    <appliance>
>      <identity>appliance0</identity>
>      <cluster_inetport>localhost:16670</cluster_inetport>
>      <mgmt_url>http://localhost:17665/mgmt/bpl</mgmt_url>
>      <engine_url>http://localhost:17666/engine/bpl</engine_url>
>      <etl_url>http://localhost:17667/etl/bpl</etl_url>
>      <retrieval_url>http://localhost:17668/retrieval/bpl</retrieval_url>
>      <data_retrieval_url>http://localhost:17668/retrieval</data_retrieval_url>
>    </appliance>
> </appliances>
>
> also checked the conf file for archappl
>
> JAVA_HOME="/usr/lib/jvm/java-11-openjdk-amd64"
> JAVA_OPTS="-Djava.security.egd=file:/dev/./urandom"
> CATALINA_HOME="/usr/share/tomcat9"
> CATALINA_OPTS="-XX:MaxMetaspaceSize=256m -Xms512m -Xmx512m -XX:+UseG1GC -ea"
>
> ARCHAPPL_APPLIANCES="/opt/epicsarchiverap/appliances.xml"
> ARCHAPPL_POLICIES="/opt/epicsarchiverap/policies.py"
> ARCHAPPL_PROPERTIES_FILENAME="/opt/epicsarchiverap/archappl.properties"
> ARCHAPPL_MYIDENTITY="appliance0"
> ARCHAPPL_STORAGE_TOP="/arch"
> ARCHAPPL_SHORT_TERM_FOLDER="/arch/sts/ArchiverStore"
> ARCHAPPL_MEDIUM_TERM_FOLDER="/arch/mts/ArchiverStore"
> ARCHAPPL_LONG_TERM_FOLDER="/arch/lts/ArchiverStore"
>
> Example Record for one of the PVs
>
> record(ao,"$(INST):a"){
> field(DESC,"Input A")
> field(DTYP,"Soft Channel")
> field(PREC,1)
> field(DRVH,10)
> field(HOPR,10)
> field(VAL,0)
> field(LOPR,-10)
> field(DRVL,-10)
> field(UDF,1)
> field(FLNK,"$(INST):sum")
> field(ADEL,0)
> field(MDEL,0)
> }
>
> Thanks