Uploaded image for project: 'Data Management'
  1. Data Management
  2. DM-21175

Try using a single DDS sample cache shared by the script queue and scripts

    XMLWordPrintable

    Details

    • Story Points:
      1
    • Sprint:
      TSSW Sprint - Sep 2 - Sep 14, TSSW Sprint - Sep 16 - Sep 28, TSSW Sprint - Sep 29 - Oct 13, TSSW Sprint - Oct 14 - Oct 27
    • Team:
      Telescope and Site

      Description

      The ADLink consultant we hired suggested that we use a single DDS cache shared between the script queue and scripts in order to speed up starting up scripts (apparently it will greatly reduce the time required to wait for historical data).

      Dave Mills suggests adding the following to the ospl.xml file pointed to by OSPL_URI
      (ie the global configuration):

      <DurabilityService name="durability">
            <Network>
               <InitialDiscoveryPeriod>0.1</InitialDiscoveryPeriod>
               <Alignment>
                  <TimeAlignment>false</TimeAlignment>
                  <RequestCombinePeriod>
                     <Initial>1.0</Initial>
                     <Operational>0.1</Operational>
                  </RequestCombinePeriod>
               </Alignment>
               <WaitForAttachment maxWaitCount="1">
                  <ServiceName>ddsi2</ServiceName>
               </WaitForAttachment>
            </Network>
            <Persistent>
               <StoreDirectory>/data/persist/</StoreDirectory>
               <StoreMode>XML</StoreMode>
             </Persistent>
            <Tracing>
               <Verbosity>NONE</Verbosity>
      <!--          <OutputFile>durability.log</OutputFile> -->
            </Tracing>
            <NameSpaces>
               <NameSpace name="defaultNamespace">
                  <Partition>*</Partition>
               </NameSpace>
               <Policy alignee="Initial" aligner="true" durability="Durable" nameSpace="defaultNamespace"/>
            </NameSpaces>
         </DurabilityService>
      

      I think the relevant bit is:

          <Persistent>
              <StoreDirectory>/data/persist/</StoreDirectory>
              <StoreMode>XML</StoreMode>
          </Persistent>
      

        Attachments

          Issue Links

            Activity

            Hide
            rowen Russell Owen added a comment - - edited

            For the record: release notes for 6.10 are here: https://istkb.adlinktech.com/article/fixed-bugs-and-changes-in-vortex-opensplice-6-10-x/
            unfortunately I don't see the text Dave Mills quoted, so I have no idea which version contains the possible fix.

            The licensed version I have a copy of is some flavor of 6.10.2. But the Python is too old so I'll need a new Docker container before continuing these tests.

            Show
            rowen Russell Owen added a comment - - edited For the record: release notes for 6.10 are here: https://istkb.adlinktech.com/article/fixed-bugs-and-changes-in-vortex-opensplice-6-10-x/ unfortunately I don't see the text Dave Mills quoted, so I have no idea which version contains the possible fix. The licensed version I have a copy of is some flavor of 6.10.2. But the Python is too old so I'll need a new Docker container before continuing these tests.
            Show
            dmills Dave Mills added a comment - https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=5&ved=2ahUKEwitovy82OnkAhUMs54KHa-xBYEQFjAEegQIAhAB&url=https%3A%2F%2Fistkb.adlinktech.com%2Farticle%2Fnew-features-vortex-opensplice%2F&usg=AOvVaw3B2k3VxWfOraUL4nDit5u5 The referenced text is in the release notes for 6.6 section
            Hide
            rowen Russell Owen added a comment - - edited

            Thank you for the reference Dave Mills. Since we are using OpenSplice 6.9 (community) and 6.10 (licensed) this fix should have been present for my tests.
            I strongly suspect my approach is flawed. Perhaps the scripts aren't using the cache file, but I don't see how that could be; subprocesses inherit the environment.

            It is also worth trying the very latest release (6.10.2p4), but I will need a licensed version of that.

            Show
            rowen Russell Owen added a comment - - edited Thank you for the reference Dave Mills . Since we are using OpenSplice 6.9 (community) and 6.10 (licensed) this fix should have been present for my tests. I strongly suspect my approach is flawed. Perhaps the scripts aren't using the cache file, but I don't see how that could be; subprocesses inherit the environment. It is also worth trying the very latest release (6.10.2p4), but I will need a licensed version of that.
            Hide
            rowen Russell Owen added a comment -

            For the next step I propose to the latest OpenSplice: 6.10.3 as I write this. This is blocked on DM-21750.

            I do not expect it to work and I have no idea what else we can do. It may be time to try to get more help from ADLink at this point.

            Show
            rowen Russell Owen added a comment - For the next step I propose to the latest OpenSplice: 6.10.3 as I write this. This is blocked on DM-21750 . I do not expect it to work and I have no idea what else we can do. It may be time to try to get more help from ADLink at this point.
            Hide
            rowen Russell Owen added a comment -

            I have spent these story points already and will put future work on a new ticket.

            Show
            rowen Russell Owen added a comment - I have spent these story points already and will put future work on a new ticket.

              People

              Assignee:
              rowen Russell Owen
              Reporter:
              rowen Russell Owen
              Watchers:
              Dave Mills, Russell Owen
              Votes:
              0 Vote for this issue
              Watchers:
              2 Start watching this issue

                Dates

                Created:
                Updated:
                Resolved:

                  Jenkins

                  No builds found.