Details

    • Type: Improvement Improvement
    • Status: Closed
    • Priority: Major Major
    • Resolution: Fixed
    • Affects Version/s: 4.0-ALPHA
    • Fix Version/s: 4.0-ALPHA
    • Labels:
      None

      Description

      There have been more improvements to how the new UI works, but the current open bugs are getting hard to keep straight. This is the new catch-all JIRA for continued improvements.

      1. SOLR-3162.patch
        631 kB
        Stefan Matheis (steffkes)
      2. SOLR-3162.patch
        631 kB
        Stefan Matheis (steffkes)
      3. SOLR-3162.patch
        227 kB
        Stefan Matheis (steffkes)
      4. SOLR-3162.patch
        216 kB
        Stefan Matheis (steffkes)
      5. SOLR-3162.patch
        186 kB
        Stefan Matheis (steffkes)
      6. SOLR-3162.patch
        186 kB
        Stefan Matheis (steffkes)
      7. SOLR-3162-index.png
        139 kB
        Stefan Matheis (steffkes)
      8. SOLR-3162-schema-browser.png
        111 kB
        Stefan Matheis (steffkes)

        Issue Links

          Activity

          Hide
          Erick Erickson added a comment -

          Any work here should be applied after the other patches, which are all committed to trunk as of 24-Feb-2012

          Show
          Erick Erickson added a comment - Any work here should be applied after the other patches, which are all committed to trunk as of 24-Feb-2012
          Hide
          Bill Bell added a comment -

          We need to add "edismax" in the query section. "dismax" is there... Also, all the edismax parameters. ps,pf,qf, etc.

          Show
          Bill Bell added a comment - We need to add "edismax" in the query section. "dismax" is there... Also, all the edismax parameters. ps,pf,qf, etc.
          Hide
          Bill Bell added a comment -

          Should we put on the front-page "HTTP caching is OFF" ?

          Show
          Bill Bell added a comment - Should we put on the front-page "HTTP caching is OFF" ?
          Hide
          Stefan Matheis (steffkes) added a comment -

          Quick Path for the Zookeeper Servlet, we didn't need the Content-'Preview' if we are browsing the tree and i removed the escaped Quotes, Mark mentioned them on SOLR-3116

          Show
          Stefan Matheis (steffkes) added a comment - Quick Path for the Zookeeper Servlet, we didn't need the Content-'Preview' if we are browsing the tree and i removed the escaped Quotes, Mark mentioned them on SOLR-3116
          Hide
          Stefan Matheis (steffkes) added a comment -

          Bill,

          We need to add "edismax" in the query section. "dismax" is there... Also, all the edismax parameters. ps,pf,qf, etc.

          Will add them to the List .. i never used edismax myself, so, is there Documentation about the available params (and their allowed values)?

          Show
          Stefan Matheis (steffkes) added a comment - Bill, We need to add "edismax" in the query section. "dismax" is there... Also, all the edismax parameters. ps,pf,qf, etc. Will add them to the List .. i never used edismax myself, so, is there Documentation about the available params (and their allowed values)?
          Hide
          Erick Erickson added a comment -

          Stefan:

          For the time being, edismax can have the same parameters as dismax, the differences are mostly in the syntax that's parsed. There's some talk of pf2 and pf3 parameters, but that hasn't been committed (SOLR-2058).

          BTW, I'm sooooo glad you're working on this stuff, I really hope that the new UI pretty much replaces the old one for 4.0 and my js skills...er...could be better...

          Show
          Erick Erickson added a comment - Stefan: For the time being, edismax can have the same parameters as dismax, the differences are mostly in the syntax that's parsed. There's some talk of pf2 and pf3 parameters, but that hasn't been committed ( SOLR-2058 ). BTW, I'm sooooo glad you're working on this stuff, I really hope that the new UI pretty much replaces the old one for 4.0 and my js skills...er...could be better...
          Hide
          Erick Erickson added a comment -

          Ahhh, misread SOLR-2058. pf2 and pf3 have been in for a long time, that JIRA was about adding phrase slop to them....

          Show
          Erick Erickson added a comment - Ahhh, misread SOLR-2058 . pf2 and pf3 have been in for a long time, that JIRA was about adding phrase slop to them....
          Hide
          Stefan Matheis (steffkes) added a comment -

          For the time being, edismax can have the same parameters as dismax, the differences are mostly in the syntax that's parsed.

          I had a look at the Query-Interface .. should we just duplicate the "dismax" Section, or it is more usefull, that we extend the section to "(e)dismax" and include a selection between "dismax" & "edismax"? Don't know how ppl use these two .. Using the same Param-Values with different parsers? Or? Tell me how

          Show
          Stefan Matheis (steffkes) added a comment - For the time being, edismax can have the same parameters as dismax, the differences are mostly in the syntax that's parsed. I had a look at the Query-Interface .. should we just duplicate the "dismax" Section, or it is more usefull, that we extend the section to "(e)dismax" and include a selection between "dismax" & "edismax"? Don't know how ppl use these two .. Using the same Param-Values with different parsers? Or? Tell me how
          Hide
          Erick Erickson added a comment -

          By and large, they use one or the other, so it seems that having them be siblings makes the most sense. But I don't have strong feelings about it if anyone else wants to chime in...

          Show
          Erick Erickson added a comment - By and large, they use one or the other, so it seems that having them be siblings makes the most sense. But I don't have strong feelings about it if anyone else wants to chime in...
          Hide
          Erick Erickson added a comment -

          I'm not sure whether then information is there and we're not showing it or whether the info just isn't there. But it would be helpful if we knew which node was the leader in collections/collection###/leader_elect/shard#. I think the raw data looks something like:
          15130063773499394-MACINTOSH.LOCAL:7574_SOLR_-N_0000000001 (ephemeral v=0)
          and the -N000000000001 indicates that this is a leader, but it might just be the count. I only have two elements/shard, so since I just have 000000000 and 00000001 they might just be shard ids....

          Show
          Erick Erickson added a comment - I'm not sure whether then information is there and we're not showing it or whether the info just isn't there. But it would be helpful if we knew which node was the leader in collections/collection###/leader_elect/shard#. I think the raw data looks something like: 15130063773499394-MACINTOSH.LOCAL:7574_SOLR_-N_0000000001 (ephemeral v=0) and the -N000000000001 indicates that this is a leader, but it might just be the count. I only have two elements/shard, so since I just have 000000000 and 00000001 they might just be shard ids....
          Hide
          Stefan Matheis (steffkes) added a comment - - edited

          Combined Patch, contains:

          • ZookeeperServlet
            • removed Content-Preview
            • escaped Quotes
          • Cloud-Tab
            • First node is automatically expanded
          • Logging
          • Schema-Browser
            • Button for loading Info on Demand
            • Form to decide how many Terms to load
            • Type/Name shown in Plain-Text
          • Index
            • Check for activated Admin-Handlers, display error message otherwise
          Show
          Stefan Matheis (steffkes) added a comment - - edited Combined Patch, contains: ZookeeperServlet removed Content-Preview escaped Quotes Cloud-Tab First node is automatically expanded Logging Now functional, using SOLR-2459 Schema-Browser Button for loading Info on Demand Form to decide how many Terms to load Type/Name shown in Plain-Text Index Check for activated Admin-Handlers, display error message otherwise
          Hide
          Stefan Matheis (steffkes) added a comment -

          Should we put on the front-page "HTTP caching is OFF" ?

          Yes, good Question .. don't know? Actually the following code in admin/_info.jsp is used:

          boolean cachingEnabled = !solrConfig.getHttpCachingConfig().isNever304(); 

          I could try to get the <httpCaching />-Element directly from the solrconfig, but this is a bit error prone

          Show
          Stefan Matheis (steffkes) added a comment - Should we put on the front-page "HTTP caching is OFF" ? Yes, good Question .. don't know? Actually the following code in admin/_info.jsp is used: boolean cachingEnabled = !solrConfig.getHttpCachingConfig().isNever304(); I could try to get the <httpCaching /> -Element directly from the solrconfig, but this is a bit error prone
          Hide
          Stefan Matheis (steffkes) added a comment -

          Neil created an Issue on my github repo, regarding the following:

          You can see either an "[ ENABLE ]" or "[ DISABLE ]" link.
          This link is missing from the new admin page.

          Never used this myself .. any ideas how to integrate this in the admin ui? Additionally we would need the functionality as a handler/servlet-thingy - should not be that completed, but perhaps we can combine the tasks? Enable/Disable as before + possibility to check the current state?

          Show
          Stefan Matheis (steffkes) added a comment - Neil created an Issue on my github repo, regarding the following: You can see either an "[ ENABLE ]" or "[ DISABLE ]" link. This link is missing from the new admin page. Never used this myself .. any ideas how to integrate this in the admin ui? Additionally we would need the functionality as a handler/servlet-thingy - should not be that completed, but perhaps we can combine the tasks? Enable/Disable as before + possibility to check the current state?
          Hide
          Sami Siren added a comment -

          Thanks Stefan for working on this one.

          I'd like to throw a suggestion into the soup... There's some useful system info that would make a good addition in the ui:

          http://localhost:8983/solr/collection1/admin/system?wt=json

          Show
          Sami Siren added a comment - Thanks Stefan for working on this one. I'd like to throw a suggestion into the soup... There's some useful system info that would make a good addition in the ui: http://localhost:8983/solr/collection1/admin/system?wt=json
          Hide
          Erick Erickson added a comment - - edited

          At least I think this requires SOLR-2459.

          Show
          Erick Erickson added a comment - - edited At least I think this requires SOLR-2459 .
          Hide
          Erick Erickson added a comment -

          Stefan:
          Yep, that form for a patch file looks good, thanks!

          On a quick test I'm getting a 404 error when clicking on the Solr Cloud link, don't know if I've got something messed up or not...

          Show
          Erick Erickson added a comment - Stefan: Yep, that form for a patch file looks good, thanks! On a quick test I'm getting a 404 error when clicking on the Solr Cloud link, don't know if I've got something messed up or not...
          Hide
          Stefan Matheis (steffkes) added a comment -

          Sami, which properties would you like to see there? Mainly those from the system-Property, or things like the (boot)classpath from jvm/jmx ?

          Show
          Stefan Matheis (steffkes) added a comment - Sami, which properties would you like to see there? Mainly those from the system -Property, or things like the (boot)classpath from jvm/jmx ?
          Hide
          Sami Siren added a comment -

          Mainly I would be interested in seeing all the memory/swap related stuff + numbers about current open files + max open files

          Show
          Sami Siren added a comment - Mainly I would be interested in seeing all the memory/swap related stuff + numbers about current open files + max open files
          Hide
          Stefan Matheis (steffkes) added a comment -

          At least I think this requires SOLR-2459.

          Yepp, Ryan posted that it was committed in Rev 1292617, don't know why it's still open

          Show
          Stefan Matheis (steffkes) added a comment - At least I think this requires SOLR-2459 . Yepp, Ryan posted that it was committed in Rev 1292617, don't know why it's still open
          Hide
          Stefan Matheis (steffkes) added a comment -

          On a quick test I'm getting a 404 error when clicking on the Solr Cloud link, don't know if I've got something messed up or not...

          Uh oh, thanks for the hint .. my fault, used the wrong path for the zookeeper-Servlet

          Show
          Stefan Matheis (steffkes) added a comment - On a quick test I'm getting a 404 error when clicking on the Solr Cloud link, don't know if I've got something messed up or not... Uh oh, thanks for the hint .. my fault, used the wrong path for the zookeeper-Servlet
          Hide
          Mark Miller added a comment -

          Wishlist item:

          The zookeeper info has information about the current cluster state, but its not really in a very user friendly consumable form.

          It would be really great if we had a more graphical representation of cluster using the information from zookeeper.

          All of the information to build such a view is in the zookeeper servlet.

          The clusterstate.json file has the layout and state about each node, and the /live_nodes list gives which nodes are considered up and down by zookeeper. Together, this gives the major status of the cluster.

          So for a cluster that had a clusterstate.json with one collection, 2 shards, and 2 nodes for each shard, you might imagine seeing 4 circles arranged in a square under the heading "collection1". There would be a row of circles for each shard and a column for each replica. One of the circles would represent a shard entry that did not have a node_name under /live_nodes, so it would be black. Another circle would represent a shard entry that had a node_name that was under/live_nodes and had a state of active so it would be green. Another circle would be good with /live_nodes but have a state of recoverying, so it would be yellow. The final circle would be good with /live_nodes but have a state of recovery_failed, so it would be red.

          At a glance, you could roughly see what your cluster looked like, and what state it was in. Then of course you could also add more, but this would be the basics. Just an idea

          Show
          Mark Miller added a comment - Wishlist item: The zookeeper info has information about the current cluster state, but its not really in a very user friendly consumable form. It would be really great if we had a more graphical representation of cluster using the information from zookeeper. All of the information to build such a view is in the zookeeper servlet. The clusterstate.json file has the layout and state about each node, and the /live_nodes list gives which nodes are considered up and down by zookeeper. Together, this gives the major status of the cluster. So for a cluster that had a clusterstate.json with one collection, 2 shards, and 2 nodes for each shard, you might imagine seeing 4 circles arranged in a square under the heading "collection1". There would be a row of circles for each shard and a column for each replica. One of the circles would represent a shard entry that did not have a node_name under /live_nodes, so it would be black. Another circle would represent a shard entry that had a node_name that was under/live_nodes and had a state of active so it would be green. Another circle would be good with /live_nodes but have a state of recoverying, so it would be yellow. The final circle would be good with /live_nodes but have a state of recovery_failed, so it would be red. At a glance, you could roughly see what your cluster looked like, and what state it was in. Then of course you could also add more, but this would be the basics. Just an idea
          Hide
          Mark Miller added a comment -

          Wishlist item:

          The zookeeper info has information about the current cluster state, but its not really in a very user friendly consumable form.

          It would be really great if we had a more graphical representation of cluster using the information from zookeeper.

          All of the information to build such a view is in the zookeeper servlet.

          The clusterstate.json file has the layout and state about each node, and the /live_nodes list gives which nodes are considered up and down by zookeeper. Together, this gives the major status of the cluster.

          So for a cluster that had a clusterstate.json with one collection, 2 shards, and 2 nodes for each shard, you might imagine seeing 4 circles arranged in a square under the heading "collection1". There would be a row of circles for each shard and a column for each replica. One of the circles would represent a shard entry that did not have a node_name under /live_nodes, so it would be black. Another circle would represent a shard entry that had a node_name that was under/live_nodes and had a state of active so it would be green. Another circle would be good with /live_nodes but have a state of recoverying, so it would be yellow. The final circle would be good with /live_nodes but have a state of recovery_failed, so it would be red.

          At a glance, you could roughly see what your cluster looked like, and what state it was in. Then of course you could also add more, but this would be the basics. Just an idea

          Show
          Mark Miller added a comment - Wishlist item: The zookeeper info has information about the current cluster state, but its not really in a very user friendly consumable form. It would be really great if we had a more graphical representation of cluster using the information from zookeeper. All of the information to build such a view is in the zookeeper servlet. The clusterstate.json file has the layout and state about each node, and the /live_nodes list gives which nodes are considered up and down by zookeeper. Together, this gives the major status of the cluster. So for a cluster that had a clusterstate.json with one collection, 2 shards, and 2 nodes for each shard, you might imagine seeing 4 circles arranged in a square under the heading "collection1". There would be a row of circles for each shard and a column for each replica. One of the circles would represent a shard entry that did not have a node_name under /live_nodes, so it would be black. Another circle would represent a shard entry that had a node_name that was under/live_nodes and had a state of active so it would be green. Another circle would be good with /live_nodes but have a state of recoverying, so it would be yellow. The final circle would be good with /live_nodes but have a state of recovery_failed, so it would be red. At a glance, you could roughly see what your cluster looked like, and what state it was in. Then of course you could also add more, but this would be the basics. Just an idea
          Hide
          Ryan McKinley added a comment -

          Mark... I made SOLR-3174 to track the possibility of cluster state visualization

          Is this what you were thinking? or something else?

          Show
          Ryan McKinley added a comment - Mark... I made SOLR-3174 to track the possibility of cluster state visualization Is this what you were thinking? or something else?
          Hide
          Erik Hatcher added a comment - - edited

          Quoting from a comment above:

          ... Additionally we would need the functionality as a handler/servlet-thingy ...

          This is where the, tada, VelocityResponseWriter could really help here. Rather than raw data having to come from Ajax calls, through a Velocity template you can get at pretty much anything from the SolrCore and configuration, and then use that to generate a response (even, say, an "x = '$whatever'" kinda variable into JavaScript. For example:

          never304 = $request.core.solrConfig.httpCachingConfig.never304
          

          You could get this using an out of the box request like this: http://localhost:8983/solr/select?q=*:*&wt=velocity&v.template=foo&v.template.foo=never304%20=%20$request.core.solrConfig.httpCachingConfig.never304 (though of course I'd recommend templates be created from this under conf/velocity rather than passed in via the request). The point is, everything can be gathered when you're "inside" Solr, but requires explicit exposing of these inner details to make the data available cleanly for this Ajax-the-data-generate-UI-in-JavaScript approach.

          Show
          Erik Hatcher added a comment - - edited Quoting from a comment above: ... Additionally we would need the functionality as a handler/servlet-thingy ... This is where the, tada, VelocityResponseWriter could really help here. Rather than raw data having to come from Ajax calls, through a Velocity template you can get at pretty much anything from the SolrCore and configuration, and then use that to generate a response (even, say, an "x = '$whatever'" kinda variable into JavaScript. For example: never304 = $request.core.solrConfig.httpCachingConfig.never304 You could get this using an out of the box request like this: http://localhost:8983/solr/select?q=*:*&wt=velocity&v.template=foo&v.template.foo=never304%20=%20$request.core.solrConfig.httpCachingConfig.never304 (though of course I'd recommend templates be created from this under conf/velocity rather than passed in via the request). The point is, everything can be gathered when you're "inside" Solr, but requires explicit exposing of these inner details to make the data available cleanly for this Ajax-the-data-generate-UI-in-JavaScript approach.
          Hide
          Stefan Matheis (steffkes) added a comment -

          Updated Patch includes redesigned Index Screen

          Show
          Stefan Matheis (steffkes) added a comment - Updated Patch includes redesigned Index Screen
          Hide
          Stefan Matheis (steffkes) added a comment -

          Attached Screenshots shows the new Index Screen and the Schema-Browser .. if anyone would have a look w/o applying the Patch

          Show
          Stefan Matheis (steffkes) added a comment - Attached Screenshots shows the new Index Screen and the Schema-Browser .. if anyone would have a look w/o applying the Patch
          Hide
          Sami Siren added a comment -

          Attached Screenshots shows the new Index Screen

          Looks nice! I noticed there's a typo in the index screen "File Descripter Count" -> "File Descriptor Count"

          Show
          Sami Siren added a comment - Attached Screenshots shows the new Index Screen Looks nice! I noticed there's a typo in the index screen "File Descripter Count" -> "File Descriptor Count"
          Hide
          Erick Erickson added a comment -

          Cool!

          Minor nit: When looking at the schema file in the cloud tree (cloud>>config>>myconf>>schema.xml) all the double quotes are escaped. This is not true when examining the schema.xml file in the "singlecore" bit. Don't know whether the raw data comes in like that or not.

          Also, the cloud tree is not expanded on startup, (Chrome 17.0.963.56)

          But the 404 error is gone which is a very good thing <G>....

          Show
          Erick Erickson added a comment - Cool! Minor nit: When looking at the schema file in the cloud tree (cloud>>config>>myconf>>schema.xml) all the double quotes are escaped. This is not true when examining the schema.xml file in the "singlecore" bit. Don't know whether the raw data comes in like that or not. Also, the cloud tree is not expanded on startup, (Chrome 17.0.963.56) But the 404 error is gone which is a very good thing <G>....
          Hide
          Mark Miller added a comment -

          Is this what you were thinking?

          +1

          Show
          Mark Miller added a comment - Is this what you were thinking? +1
          Hide
          Stefan Matheis (steffkes) added a comment -

          Updated Patch, contains:

          • edismax Options on Query-Tab
          • Check if System-Information on Dashboard is available
          • Fixed Param-Handling on Dataimport
          • Autoload™ Functionality on Schema-Browser
          • Dummy Debug-Option on Cloud-Tab
          Show
          Stefan Matheis (steffkes) added a comment - Updated Patch, contains: edismax Options on Query-Tab Check if System-Information on Dashboard is available Fixed Param-Handling on Dataimport Autoload™ Functionality on Schema-Browser Dummy Debug-Option on Cloud-Tab
          Hide
          Stefan Matheis (steffkes) added a comment - - edited

          Sami, yepp also noticed that .. already fixed after taking the screenshot

          Erick, Are the double Quotes still there? The Patch should remove all replaceAll usages, so the raw content should be visible right know. I'm not completely sure which sources are used for the cloud-tab and the /admin/file Handler, so it maybe give you different output ;o

          Cloud-Tree Expanding still not working? Even w/ the latest Patch? Just to be sure, cleared the Browser-Cache?

          Show
          Stefan Matheis (steffkes) added a comment - - edited Sami, yepp also noticed that .. already fixed after taking the screenshot Erick, Are the double Quotes still there? The Patch should remove all replaceAll usages, so the raw content should be visible right know. I'm not completely sure which sources are used for the cloud-tab and the /admin/file Handler, so it maybe give you different output ;o Cloud-Tree Expanding still not working? Even w/ the latest Patch? Just to be sure, cleared the Browser-Cache?
          Hide
          Erick Erickson added a comment -

          Stefan:

          The escapes are still there, like this. I'm starting to think there's something up with how ZK returns the data, see below...
          <fieldType class=\"solr.TrieIntField\" name=\"tint\" omitNorms=\"true\" positionIncrementGap=\"0\" precisionStep=\"8\"/>

          Hmmm, ZK savvy people. I'm guessing that when you ask the cloud for something like the schema.xml file, it really comes from ZK rather than any of the conf directories. If so, is there anything special about how that comes back to the caller? Here's the path used in the new Admin UI:

          http://localhost:8983/solr/zookeeper?detail=true&path=%2Fconfigs%2Fmyconf%2Fschema.xml

          and just plugging that into the browser returns stuff like
          <fieldType class=\"solr.TrieIntField\" .....

          the top-level node is now expanded on startup.

          Show
          Erick Erickson added a comment - Stefan: The escapes are still there, like this. I'm starting to think there's something up with how ZK returns the data, see below... <fieldType class=\"solr.TrieIntField\" name=\"tint\" omitNorms=\"true\" positionIncrementGap=\"0\" precisionStep=\"8\"/> Hmmm, ZK savvy people. I'm guessing that when you ask the cloud for something like the schema.xml file, it really comes from ZK rather than any of the conf directories. If so, is there anything special about how that comes back to the caller? Here's the path used in the new Admin UI: http://localhost:8983/solr/zookeeper?detail=true&path=%2Fconfigs%2Fmyconf%2Fschema.xml and just plugging that into the browser returns stuff like <fieldType class=\"solr.TrieIntField\" ..... the top-level node is now expanded on startup.
          Hide
          Stefan Matheis (steffkes) added a comment - - edited

          That's pretty odd :< After removing this line (occurs two times):

          -            str = str.replaceAll("\\\"", "\\\\\"");

          the Problem should not longer exists. The raw Data which is stored in ZK is fine, i added these replaces at the beginning to generate valid output - which is no longer required through the use of noggit. So, the output of

          <fieldType name=\"int\" class=\"solr.TrieIntField\" precisionStep=\"0\" omitNorms=\"true\" positionIncrementGap=\"0\"/>

          should change to that

          <fieldType name="int" class="solr.TrieIntField" precisionStep="0" omitNorms="true" positionIncrementGap="0"/>

          if the escaped quotes are still there .. could you double-check if you still have some replace*() Calls inside solr/core/src/java/org/apache/solr/servlet/ZookeeperInfoServlet.java Erick? Maybe i missed some of them, which shouldn't be - but i don't know what's wrong on your machine :/

          Show
          Stefan Matheis (steffkes) added a comment - - edited That's pretty odd :< After removing this line (occurs two times): - str = str.replaceAll( "\\\" ", " \\\\\""); the Problem should not longer exists. The raw Data which is stored in ZK is fine, i added these replaces at the beginning to generate valid output - which is no longer required through the use of noggit. So, the output of <fieldType name=\ " int \" class=\ "solr.TrieIntField\" precisionStep=\ "0\" omitNorms=\ " true \" positionIncrementGap=\ "0\" /> should change to that <fieldType name= " int " class= "solr.TrieIntField" precisionStep= "0" omitNorms= " true " positionIncrementGap= "0" /> if the escaped quotes are still there .. could you double-check if you still have some replace*() Calls inside solr/core/src/java/org/apache/solr/servlet/ZookeeperInfoServlet.java Erick? Maybe i missed some of them, which shouldn't be - but i don't know what's wrong on your machine :/
          Hide
          Mark Miller added a comment -

          Awesome stuff. Can't say it enough.

          Show
          Mark Miller added a comment - Awesome stuff. Can't say it enough.
          Hide
          Stefan Matheis (steffkes) added a comment -

          Updated Patch

          • Lucene Guidelines, indentation using two whitespaces
          • 'ZeroClipboard' Functionality for Cloud-Tab (output depends on SOLR-3181)
          • Basic Timer on Replication Interface
          • Extended Error-Handling on Analysis-Interface
          Show
          Stefan Matheis (steffkes) added a comment - Updated Patch Lucene Guidelines, indentation using two whitespaces 'ZeroClipboard' Functionality for Cloud-Tab (output depends on SOLR-3181 ) Basic Timer on Replication Interface Extended Error-Handling on Analysis-Interface
          Hide
          Ryan McKinley added a comment -

          Rather then make this issue huge... could we make new issues for the outstanding tasks? It is difficult to tell what is left to do.

          We can use the JIRA component "web gui"

          Show
          Ryan McKinley added a comment - Rather then make this issue huge... could we make new issues for the outstanding tasks? It is difficult to tell what is left to do. We can use the JIRA component "web gui"
          Hide
          Stefan Matheis (steffkes) added a comment -

          Updated Patch, Changes regarding SOLR-3202

          Show
          Stefan Matheis (steffkes) added a comment - Updated Patch, Changes regarding SOLR-3202
          Hide
          Erick Erickson added a comment -

          r: 1298010

          I folded SOLR-3181 into this commit.

          Show
          Erick Erickson added a comment - r: 1298010 I folded SOLR-3181 into this commit.
          Hide
          Erick Erickson added a comment -

          OK, let's use different/new JIRAs to continue UI development....

          Show
          Erick Erickson added a comment - OK, let's use different/new JIRAs to continue UI development....
          Hide
          Aliaksandr Zhuhrou added a comment -

          Guys, does the new admin interface works with the war deployment for solr?
          I got NullPointerException on the org.apache.solr.servlet.LoadAdminUiServlet#doGet
          File f = new File(getServletContext().getRealPath("admin.html"));
          I am using tomcat 7.0.23 for deployment.
          Also I may mess something with solr.war

          Show
          Aliaksandr Zhuhrou added a comment - Guys, does the new admin interface works with the war deployment for solr? I got NullPointerException on the org.apache.solr.servlet.LoadAdminUiServlet#doGet File f = new File(getServletContext().getRealPath("admin.html")); I am using tomcat 7.0.23 for deployment. Also I may mess something with solr.war
          Hide
          Thomas Weidman added a comment -

          Hi, I've made some adjustments to Dataimport admin handler UI that others may like:

          1) js\scripts\dataimport.js -> add new var in dataimport.html template loading:

          var buttons = $( 'button.actions', dataimport_element );

          2) Add event for button.actions

          buttons.on('click', function () {
          $.ajax({
          url : handler_url + '?command=' + $(this).attr('command'),
          dataType : 'xml',
          beforeSend : function( xhr, settings )
          {
          },
          success : function( response, text_status, xhr )

          { console.debug( response ); dataimport_fetch_status(); }

          ,
          error : function( xhr, text_status, error_thrown )

          { console.debug( arguments ); }

          ,
          complete : function( xhr, text_status )
          {
          }
          });
          });

          3) Add buttons to template

          <fieldset>
          <legend style="padding:10px 5px; 2px 5px">commands</legend>
          <button class="actions" command="full-import">Full Import</button>
            
          <button class="actions" command="delta-import">Delta Import</button>
          </fieldset>

          Show
          Thomas Weidman added a comment - Hi, I've made some adjustments to Dataimport admin handler UI that others may like: 1) js\scripts\dataimport.js -> add new var in dataimport.html template loading: var buttons = $( 'button.actions', dataimport_element ); 2) Add event for button.actions buttons.on('click', function () { $.ajax({ url : handler_url + '?command=' + $(this).attr('command'), dataType : 'xml', beforeSend : function( xhr, settings ) { }, success : function( response, text_status, xhr ) { console.debug( response ); dataimport_fetch_status(); } , error : function( xhr, text_status, error_thrown ) { console.debug( arguments ); } , complete : function( xhr, text_status ) { } }); }); 3) Add buttons to template <fieldset> <legend style="padding:10px 5px; 2px 5px">commands</legend> <button class="actions" command="full-import">Full Import</button>    <button class="actions" command="delta-import">Delta Import</button> </fieldset>
          Hide
          Stefan Matheis (steffkes) added a comment -

          Thomas, did you think that Command-Dropdown (which is already in there, first of all Options in the Form) is not visible enough so that we need two seperate Buttons?

          Show
          Stefan Matheis (steffkes) added a comment - Thomas, did you think that Command-Dropdown (which is already in there, first of all Options in the Form) is not visible enough so that we need two seperate Buttons?
          Hide
          Stefan Matheis (steffkes) added a comment -

          Aliaksandr, i never worked with Tomcat myself, so don't know :/ But could you open a separate Issue with a complete Stacktrace and link it to SOLR-3238?

          Show
          Stefan Matheis (steffkes) added a comment - Aliaksandr, i never worked with Tomcat myself, so don't know :/ But could you open a separate Issue with a complete Stacktrace and link it to SOLR-3238 ?
          Hide
          Aliaksandr Zhuhrou added a comment -

          Sure, I can do this.

          Show
          Aliaksandr Zhuhrou added a comment - Sure, I can do this.

            People

            • Assignee:
              Erick Erickson
              Reporter:
              Erick Erickson
            • Votes:
              1 Vote for this issue
              Watchers:
              2 Start watching this issue

              Dates

              • Created:
                Updated:
                Resolved:

                Development