[jira] [Created] (ZEPPELIN-5323) Interpreter Recovery Does Not Preserve Running Spark Jobs
Paul Brenner created ZEPPELIN-5323:
--
Summary: Interpreter Recovery Does Not Preserve Running Spark Jobs
Key: ZEPPELIN-5323
URL: https://issues.apache.org/jira/browse/ZEPPELIN-5323
Project: Zeppelin
Issue Type: Bug
Reporter: Paul Brenner
We are using zeppelin 0.10 built from from master on March 26th, looks like the
most recent commit was 85ed8e2e51e1ea10df38d4710216343efe218d60. We tried to
enable interpreter recovery by adding the following to zeppelin-site.xml:
zeppelin.recovery.storage.class
org.apache.zeppelin.interpreter.recovery.FileSystemRecoveryStorage
ReoveryStorage implementation based on hadoop
FileSystem
zeppelin.recovery.dir
/user/zeppelin/recovery
Location where recovery metadata is stored
when we start up zeppelin we get no errors, I can start a job running and I see
in {{/user/zeppelin/recovery/spark_paul.recovery}} that it lists
{{spark_paul-anonymous-2G3KV92PG 10.16.41.212:34374}} so that look
promising
when we stop zeppelin the interpreter process keeps running, but I see the
following happens to the spark job
21/04/08 13:42:09 INFO yarn.YarnAllocator: Canceling requests for 262 executor
container(s) to have a new desired total 0 executors.
21/04/08 13:42:09 INFO yarn.ApplicationMaster$AMEndpoint: Driver terminated or
disconnected! Shutting down. zeppelin-212.sec.placeiq.net:36733
21/04/08 13:42:09 INFO yarn.ApplicationMaster$AMEndpoint: Driver terminated or
disconnected! Shutting down. zeppelin-212.sec.placeiq.net:36733
21/04/08 13:42:09 INFO yarn.ApplicationMaster: Final app status: SUCCEEDED,
exitCode: 0
21/04/08 13:42:09 INFO yarn.ApplicationMaster: Unregistering ApplicationMaster
with SUCCEEDED
21/04/08 13:42:09 INFO impl.AMRMClientImpl: Waiting for application to be
successfully unregistered.
21/04/08 13:42:09 INFO yarn.ApplicationMaster: Deleting staging directory
hdfs://nameservice1/user/pbrenner/.sparkStaging/application_1617808481394_4478
21/04/08 13:42:09 INFO util.ShutdownHookManager: Shutdown hook called
then when we start zeppelin back up I see the following on the paragraph that
was running:
java.lang.RuntimeException: Interpreter instance
org.apache.zeppelin.spark.SparkInterpreter not created
at
org.apache.zeppelin.interpreter.remote.PooledRemoteClient.callRemoteFunction(PooledRemoteClient.java:114)
at
org.apache.zeppelin.interpreter.remote.RemoteInterpreterProcess.callRemoteFunction(RemoteInterpreterProcess.java:99)
at
org.apache.zeppelin.interpreter.remote.RemoteInterpreter.getFormType(RemoteInterpreter.java:281)
at org.apache.zeppelin.notebook.Paragraph.jobRun(Paragraph.java:442)
at org.apache.zeppelin.notebook.Paragraph.jobRun(Paragraph.java:71)
at org.apache.zeppelin.scheduler.Job.run(Job.java:172)
at
org.apache.zeppelin.scheduler.AbstractScheduler.runJob(AbstractScheduler.java:132)
at
org.apache.zeppelin.scheduler.RemoteScheduler$JobRunner.run(RemoteScheduler.java:182)
at
java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
at java.util.concurrent.FutureTask.run(FutureTask.java:266)
at
java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$201(ScheduledThreadPoolExecutor.java:180)
at
java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:293)
at
java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
at
java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
at java.lang.Thread.run(Thread.java:748)
It looks VERY close to working, but somehow spark jobs are still getting
shutdown when we shutdown zepplin. Any ideas?
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Created] (ZEPPELIN-5310) Cluster mode is broken on latest build from source
Paul Brenner created ZEPPELIN-5310:
--
Summary: Cluster mode is broken on latest build from source
Key: ZEPPELIN-5310
URL: https://issues.apache.org/jira/browse/ZEPPELIN-5310
Project: Zeppelin
Issue Type: Bug
Affects Versions: 0.10.0
Environment: Interpreter settings are as follows:
"spark_paul": {
"id": "spark_paul",
"name": "spark_paul",
"group": "spark",
"properties": {
"SPARK_HOME": {
"name": "SPARK_HOME",
"value": "",
"type": "string",
"description": "Location of spark distribution"
},
"spark.master": {
"name": "spark.master",
"value": "yarn",
"type": "string",
"description": "Spark master uri. local | yarn-client | yarn-cluster | spark
master address of standalone mode, ex) spark://master_host:7077"
},
"spark.submit.deployMode": {
"name": "spark.submit.deployMode",
"value": "client",
"type": "string",
"description": "The deploy mode of Spark driver program, either \"client\" or
\"cluster\", Which means to launch driver program locally (\"client\") or
remotely (\"cluster\") on one of the nodes inside the cluster."
},
"spark.app.name": {
"name": "spark.app.name",
"value": "zeppelin_dev_paul",
"type": "string",
"description": "The name of spark application."
},
"spark.driver.cores": {
"name": "spark.driver.cores",
"value": "1",
"type": "number",
"description": "Number of cores to use for the driver process, only in cluster
mode."
},
"spark.driver.memory": {
"name": "spark.driver.memory",
"value": "5g",
"type": "string",
"description": "Amount of memory to use for the driver process, i.e. where
SparkContext is initialized, in the same format as JVM memory strings with a
size unit suffix (\"k\", \"m\", \"g\" or \"t\") (e.g. 512m, 2g)."
},
"spark.executor.cores": {
"name": "spark.executor.cores",
"value": "1",
"type": "number",
"description": "The number of cores to use on each executor"
},
"spark.executor.memory": {
"name": "spark.executor.memory",
"value": "3g",
"type": "string",
"description": "Executor memory per worker instance. ex) 512m, 32g"
},
"spark.executor.instances": {
"name": "spark.executor.instances",
"value": "2",
"type": "number",
"description": "The number of executors for static allocation."
},
"spark.files": {
"name": "spark.files",
"value": "",
"type": "string",
"description": "Comma-separated list of files to be placed in the working
directory of each executor. Globs are allowed."
},
"spark.jars": {
"name": "spark.jars",
"value":
"http://nexus.placeiq.net:8081/nexus/content/repositories/releases/com/placeiq/lap/4.1.25/lap-4.1.25.jar,hdfs://gandalf-nn.placeiq.net/lib/dap/0.1.0/dap-jar-assembled.jar";;,
"type": "string",
"description": "Comma-separated list of jars to include on the driver and
executor classpaths. Globs are allowed."
},
"spark.jars.packages": {
"name": "spark.jars.packages",
"value": "ds-commons:ds-commons_2.11:0.1-SNAPSHOT",
"type": "string",
"description": "Comma-separated list of Maven coordinates of jars to include
on the driver and executor classpaths. The coordinates should be
groupId:artifactId:version. If spark.jars.ivySettings is given artifacts will
be resolved according to the configuration in the file, otherwise artifacts
will be searched for in the local maven repo, then maven central and finally
any additional remote repositories given by the command-line option
--repositories."
},
"zeppelin.spark.useHiveContext": {
"name": "zeppelin.spark.useHiveContext",
"value": true,
"type": "checkbox",
"description": "Use HiveContext instead of SQLContext if it is true. Enable
hive for SparkSession."
},
"zeppelin.spark.printREPLOutput": {
"name": "zeppelin.spark.printREPLOutput",
"value": true,
"type": "checkbox",
"description": "Print REPL output"
},
"zeppelin.spark.maxResult": {
"name": "zeppelin.spark.maxResult",
"value": "1000",
"type": "number",
"description": "Max number of result to display."
},
"zeppelin.spark.enableSupportedVersionCheck": {
"name": "zeppelin.spark.enableSupportedVersionCheck",
"value": true,
"type": "checkbox",
"description": "Whether checking supported spark version. Developer only
setting, not for production use"
},
"zeppelin.spark.uiWebUrl": {
"name": "zeppelin.spark.uiWebUrl",
"value": "",
"type": "string",
"description": "Override Spark UI default URL. In Kubernetes mode, value can
be Jinja template string with 3 template variables \u0027PORT\u0027,
\u0027SERVICE_NAME\u0027 and \u0027SERVICE_DOMAIN\u0027. (ex:
http://\{{PORT}}-\{{SERVICE_NAME}}.\{{SERVICE_DOMAIN}})"
},
"zeppelin.spark.ui.hidden": {
"name": "zeppelin.spark.ui.hidden",
"value": false,
"type": "checkbox",
"description": "Whether hide spark ui in zeppelin ui"
},
"spark.webui.yarn.useProxy": {
"name": "spark.webui.yarn.useProxy",
"value": false,
"type": "checkbox",
"description": "whether use yarn proxy url as spark weburl
[jira] [Created] (ZEPPELIN-4754) Helium - No Packages showing up in 0.9
Paul Brenner created ZEPPELIN-4754: -- Summary: Helium - No Packages showing up in 0.9 Key: ZEPPELIN-4754 URL: https://issues.apache.org/jira/browse/ZEPPELIN-4754 Project: Zeppelin Issue Type: Bug Reporter: Paul Brenner Attachments: image-2020-04-13-10-53-22-142.png In zeppelin 0.9 builds created over the past few weeks we no longer see any helium plugins. !image-2020-04-13-10-53-22-142.png! They still show up fine in the copy of 0.82 that we have running on a different server. Is this a known issue related to work on https://issues.apache.org/jira/browse/ZEPPELIN-4699 ? -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Created] (ZEPPELIN-4751) Colons in notebook names cause zeppelin not to start
Paul Brenner created ZEPPELIN-4751: -- Summary: Colons in notebook names cause zeppelin not to start Key: ZEPPELIN-4751 URL: https://issues.apache.org/jira/browse/ZEPPELIN-4751 Project: Zeppelin Issue Type: Bug Affects Versions: 0.9.0 Reporter: Paul Brenner Using zeppelin 0.9 built 4/9/20, some users had created notebooks with colons in the name of the notebook. The build we had of 0.9 from about a week earlier happily created and used those notebooks. But when we tried to restart with this week's build it wouldn't start until we went in and manually renamed the notebooks from the command line to not have colons in the names. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Created] (ZEPPELIN-4738) All output disappears in zeppelin 0.9
Paul Brenner created ZEPPELIN-4738: -- Summary: All output disappears in zeppelin 0.9 Key: ZEPPELIN-4738 URL: https://issues.apache.org/jira/browse/ZEPPELIN-4738 Project: Zeppelin Issue Type: Bug Reporter: Paul Brenner Randomly, notebooks will stop showing output in our recent build of Zeppelin 0.9 (built April 1,2020) . The only solution is to reload the webpage. I have not seen anything related in logs, but happy to help diagnose. I have also not figured out any trigger of this behavior. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Created] (ZEPPELIN-4737) Paragraphs Restart After Completing in Zeppelin 0.9
Paul Brenner created ZEPPELIN-4737: -- Summary: Paragraphs Restart After Completing in Zeppelin 0.9 Key: ZEPPELIN-4737 URL: https://issues.apache.org/jira/browse/ZEPPELIN-4737 Project: Zeppelin Issue Type: Bug Reporter: Paul Brenner We have encountered paragraphs randomly restarting after completion. When they restart themselves the paragraph gets stuck in pending and never completes. This is on a build of 0.9 created April 1 2020. I haven't seen anything in logs that seems helpful for diagnosing. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Created] (ZEPPELIN-4676) Run All Below Deletes All Code Below
Paul Brenner created ZEPPELIN-4676: -- Summary: Run All Below Deletes All Code Below Key: ZEPPELIN-4676 URL: https://issues.apache.org/jira/browse/ZEPPELIN-4676 Project: Zeppelin Issue Type: Bug Affects Versions: 0.9.0 Reporter: Paul Brenner Enter code into multiple paragraphs. Use "Run All Below" and it will run the paragraphs but delete all the code you entered. This is a show stopper. Same thing happens with run all above -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Created] (ZEPPELIN-4648) Attempt to enable interpreter recovery causes errors
Paul Brenner created ZEPPELIN-4648:
--
Summary: Attempt to enable interpreter recovery causes errors
Key: ZEPPELIN-4648
URL: https://issues.apache.org/jira/browse/ZEPPELIN-4648
Project: Zeppelin
Issue Type: Bug
Reporter: Paul Brenner
Zeppelin is extremely unstable for our use cases (ZEPPELIN-4599). I'd like to
try interpreter recovery as described here:
[https://medium.com/@zjffdu/zeppelin-0-8-0-new-features-ea53e8810235].
I added
{code:java}
zeppelin.recovery.storage.class
org.apache.zeppelin.interpreter.recovery.FileSystemRecoveryStorage
ReoveryStorage implementation
zeppelin.recovery.dir
/opt/zeppelin/recovery
Location where recovery metadata is stored
{code}
to zeppelin-site.xml then attempted to start zeppelin (0.9 built from source
and 0.82 packaged version). When zeppelin starts up I can no longer log in and
get the errors shown in the following logs (these are clean and only show the
messages since starting with recovery enabled). Any idea what is wrong? I
haven't seen any posts from people using recovery.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Created] (ZEPPELIN-4608) Login looses original URL
Paul Brenner created ZEPPELIN-4608: -- Summary: Login looses original URL Key: ZEPPELIN-4608 URL: https://issues.apache.org/jira/browse/ZEPPELIN-4608 Project: Zeppelin Issue Type: Bug Affects Versions: 0.8.2, 0.9.0 Reporter: Paul Brenner This is minor, but when I am not logged into zeppelin (happens daily because we have to restart daily due to zeppelin instability) and I click a link directly to notebook such as [https://zeppelin.companyname.net/#/notebook/2EYBD6A86|https://zeppelin.placeiq.net/#/notebook/2EYBD6A86] I get redirected to the login screen, which is good, but then once I login I am redirected to [https://zeppelin.companyname.net/#/|https://zeppelin.placeiq.net/#/notebook/2EYBD6A86] . After the login it should remember where I was originally going and bring me to [https://zeppelin.companyname.net/#/notebook/2EYBD6A86|https://zeppelin.placeiq.net/#/notebook/2EYBD6A86]. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Created] (ZEPPELIN-4603) No longer possible to change interpreters for a notebook in recent 0.9
Paul Brenner created ZEPPELIN-4603: -- Summary: No longer possible to change interpreters for a notebook in recent 0.9 Key: ZEPPELIN-4603 URL: https://issues.apache.org/jira/browse/ZEPPELIN-4603 Project: Zeppelin Issue Type: Bug Affects Versions: 0.9.0 Reporter: Paul Brenner Building 0.9 from source (based on a build using the live code on 1/27/2020 and again on 2/3/2020 ) no longer lets us select interpreters for a notebook. You can define an interpreter when you first create the notebook and that is your only option. If you clone a notebook you can't select an appropriate interpreter. If I create a notebook and initially set it to a spark interpreter, I have no option to later select any other interpreter. Clicking the interpreter binding gear shows only 1 interpreter available, none of the other interpreters defined in the interpreter page show up. This isn't isolated to a single interpreter. If I create a notebook and initially set it to a different interpreter the same problem happens. Even worse, cloning a notebook results in 0 options for interpreters in the interpreter binding gear making cloned notebooks useless. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Created] (ZEPPELIN-4599) Zeppelin becomes unresponsive and can only be recovered by restart
Paul Brenner created ZEPPELIN-4599: -- Summary: Zeppelin becomes unresponsive and can only be recovered by restart Key: ZEPPELIN-4599 URL: https://issues.apache.org/jira/browse/ZEPPELIN-4599 Project: Zeppelin Issue Type: Bug Affects Versions: 0.8.2 Reporter: Paul Brenner Attachments: zeppelin-yarn-zeppelin-210.sec.placeiq.net.log, zeppelin-yarn-zeppelin-210.sec.placeiq.net.out We use zeppelin with 10-20 users working primarily in spark. Every few days, and sometimes multiple times per day, the zeppelin webui becomes unresponsive and the only solution we have found is to restart zeppelin. This is extremely disruptive. "Unresponsive" usually takes the form of no longer being able to create new paragraphs, clicking run no longer doing anything or being stuck forever in pending, inability to create new notebooks, or the inability to load notebooks. We have tried adding monitoring to the box zeppelin runs on and see nothing out of the ordinary with: GC rates, CPU utilizations, Memory usage, and heap utilization We also don't see anything unusual in the logs. Is there any other way we can diagnose this issue to help find the root cause. 0.9 is currently too broken to use (based on a build using the live code on 1/27/2020 and again on 2/3/2020 ) Attaching a copy of logs JIC. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Created] (ZEPPELIN-3692) Pressing Tab Deletes Selected Editor Text After Notebook Has Been Open For Awhile
Paul Brenner created ZEPPELIN-3692: -- Summary: Pressing Tab Deletes Selected Editor Text After Notebook Has Been Open For Awhile Key: ZEPPELIN-3692 URL: https://issues.apache.org/jira/browse/ZEPPELIN-3692 Project: Zeppelin Issue Type: Bug Affects Versions: 0.8.0 Reporter: Paul Brenner Sometimes on notebooks that have been open for a while, selecting a block of text and pressing tab just deletes all the text and replaces it with a tab. This is different from the expected and usual behavior of indenting the selected text. I see plenty of errors in the java script console but don't know if any of them are relevant. Here they are JIC vendor.49d751b0c72342f6.js:37 Error: [ngRepeat:dupes] Duplicates in a repeater are not allowed. Use 'track by' expression to specify unique keys. Repeater: url in paragraph.runtimeInfos.jobUrl.values, Duplicate key: string:[http://zeppelin-211.nym1.placeiq.net:4040/jobs,] Duplicate value: [http://zeppelin-211.nym1.placeiq.net:4040/jobs] [http://errors.angularjs.org/1.5.7/ngRepeat/dupes?p0=url%20in%20paragraph.runtimeInfos.jobUrl.values&p1=string%3Ahttp%3A%2F%2Fzeppelin-211.nym1.placeiq.net%3A4040%2Fjobs&p2=http%3A%2F%2Fzeppelin-211.nym1.placeiq.net%3A4040%2Fjobs] at vendor.49d751b0c72342f6.js:36 at vendor.49d751b0c72342f6.js:40 at e (vendor.49d751b0c72342f6.js:38) at o.$digest (vendor.49d751b0c72342f6.js:38) at n.safeDigest (vendor.49d751b0c72342f6.js:76) at b (vendor.49d751b0c72342f6.js:76) at n._onMessageHandler (vendor.49d751b0c72342f6.js:76) at WebSocket. (vendor.49d751b0c72342f6.js:36) (anonymous) @ vendor.49d751b0c72342f6.js:37 app.5a2c363a867d674675d7.js:34 Uncaught TypeError: Cannot read property 'p20180731_170134_1204161345_1_table_gridApi' of undefined at t.value (app.5a2c363a867d674675d7.js:34) at t.value (app.5a2c363a867d674675d7.js:34) at app.5a2c363a867d674675d7.js:47 at dispatch (vendor.49d751b0c72342f6.js:30) at q.handle (vendor.49d751b0c72342f6.js:30) 27vendor.49d751b0c72342f6.js:37 Error: [ngRepeat:dupes] Duplicates in a repeater are not allowed. Use 'track by' expression to specify unique keys. Repeater: url in paragraph.runtimeInfos.jobUrl.values, Duplicate key: string:[http://zeppelin-211.nym1.placeiq.net:4040/jobs,] Duplicate value: [http://zeppelin-211.nym1.placeiq.net:4040/jobs] [http://errors.angularjs.org/1.5.7/ngRepeat/dupes?p0=url%20in%20paragraph.runtimeInfos.jobUrl.values&p1=string%3Ahttp%3A%2F%2Fzeppelin-211.nym1.placeiq.net%3A4040%2Fjobs&p2=http%3A%2F%2Fzeppelin-211.nym1.placeiq.net%3A4040%2Fjobs] at vendor.49d751b0c72342f6.js:36 at vendor.49d751b0c72342f6.js:40 at e (vendor.49d751b0c72342f6.js:38) at o.$digest (vendor.49d751b0c72342f6.js:38) at n.safeDigest (vendor.49d751b0c72342f6.js:76) at b (vendor.49d751b0c72342f6.js:76) at n._onMessageHandler (vendor.49d751b0c72342f6.js:76) at WebSocket. (vendor.49d751b0c72342f6.js:36) (anonymous) @ vendor.49d751b0c72342f6.js:37 app.5a2c363a867d674675d7.js:34 Uncaught TypeError: Cannot read property 'p20180731_170134_1204161345_1_table_gridApi' of undefined at t.value (app.5a2c363a867d674675d7.js:34) at t.value (app.5a2c363a867d674675d7.js:34) at app.5a2c363a867d674675d7.js:47 at dispatch (vendor.49d751b0c72342f6.js:30) at q.handle (vendor.49d751b0c72342f6.js:30) 2vendor.49d751b0c72342f6.js:37 Error: [ngRepeat:dupes] Duplicates in a repeater are not allowed. Use 'track by' expression to specify unique keys. Repeater: url in paragraph.runtimeInfos.jobUrl.values, Duplicate key: string:[http://zeppelin-211.nym1.placeiq.net:4040/jobs,] Duplicate value: [http://zeppelin-211.nym1.placeiq.net:4040/jobs] [http://errors.angularjs.org/1.5.7/ngRepeat/dupes?p0=url%20in%20paragraph.runtimeInfos.jobUrl.values&p1=string%3Ahttp%3A%2F%2Fzeppelin-211.nym1.placeiq.net%3A4040%2Fjobs&p2=http%3A%2F%2Fzeppelin-211.nym1.placeiq.net%3A4040%2Fjobs] at vendor.49d751b0c72342f6.js:36 at vendor.49d751b0c72342f6.js:40 at e (vendor.49d751b0c72342f6.js:38) at o.$digest (vendor.49d751b0c72342f6.js:38) at n.safeDigest (vendor.49d751b0c72342f6.js:76) at b (vendor.49d751b0c72342f6.js:76) at n._onMessageHandler (vendor.49d751b0c72342f6.js:76) at WebSocket. (vendor.49d751b0c72342f6.js:36) (anonymous) @ vendor.49d751b0c72342f6.js:37 4vendor.49d751b0c72342f6.js:37 Error: [ngRepeat:dupes] Duplicates in a repeater are not allowed. Use 'track by' expression to specify unique keys. Repeater: url in paragraph.runtimeInfos.jobUrl.values, Duplicate key: string:[http://zeppelin-211.nym1.placeiq.net:4040/jobs,] Duplicate value: [http://zeppelin-211.nym1.placeiq.net:4040/jobs] [http://errors.angularjs.org/1.5.7/ngRepeat/dupes?p0=url%20in%20paragraph.runtimeInfos.jobUrl.values&p1=string%3Ahttp%3A%2F%2Fzeppelin-211.nym1.placeiq.net%3A4040%2Fjobs&p2=http%3A%2F%2Fzeppelin-211.nym
[jira] [Created] (ZEPPELIN-3616) Paragraph editor sections auto-collapse
Paul Brenner created ZEPPELIN-3616: -- Summary: Paragraph editor sections auto-collapse Key: ZEPPELIN-3616 URL: https://issues.apache.org/jira/browse/ZEPPELIN-3616 Project: Zeppelin Issue Type: Bug Affects Versions: 0.8.0 Reporter: Paul Brenner We are using 0.8 release and noticed that the editor section of paragraphs will randomly collapse when you leave a notebook open for a while. Clicking "hide editor" followed by "show editor" will bring them back but it is quite annoying and wasn't there in the pre-release version we were using. Currently we have not figured out a specific process to reproduce this. Will update this ticket if we can figure out how to reliably reproduce. -- This message was sent by Atlassian JIRA (v7.6.3#76005)
[jira] [Created] (ZEPPELIN-2487) Enabling downloading more rows with Download Data as CSV
Paul Brenner created ZEPPELIN-2487: -- Summary: Enabling downloading more rows with Download Data as CSV Key: ZEPPELIN-2487 URL: https://issues.apache.org/jira/browse/ZEPPELIN-2487 Project: Zeppelin Issue Type: Improvement Affects Versions: 0.7.1 Reporter: Paul Brenner There are limits to how much data the download to csv button will download which limit Zeppelin’s usefulness for our BI teams. In Chrome the limit is either 1.5MB or 3500 rows. In Firefox it is possible to download 100k rows, however the web interface slows to a crawl. Although most users fluent in whatever language they are working with would just export 100k rows directly to csv (e.g. via df.write.csv() in spark), if Zeppelin is going to be a BI tool useful to analysts it is important to include a way to export more data to csv via the web app. Would it be possible to include a way to download more data (perhaps up to 100k rows if we need a limit) without showing all the rows on screen? Hue is able to do this by only loading data as it is scrolled through. If that is too complex perhaps just add an option to download all of the data sent to z.show even though z.show is limited to showing 1000 rows. Discussion on the users list is here: https://lists.apache.org/thread.html/fb9d3f95bb525fcb05baea06b303c192f393e7570eecd643f3cb3057@%3Cusers.zeppelin.apache.org%3E -- This message was sent by Atlassian JIRA (v6.3.15#6346)
[jira] [Created] (ZEPPELIN-2415) Auto-suggestion of interpreter permissions should list roles as well
Paul Brenner created ZEPPELIN-2415: -- Summary: Auto-suggestion of interpreter permissions should list roles as well Key: ZEPPELIN-2415 URL: https://issues.apache.org/jira/browse/ZEPPELIN-2415 Project: Zeppelin Issue Type: Improvement Reporter: Paul Brenner Auto-suggestion of interpreter permissions should list roles as well. This has already been implemented for notebook permissions in: https://issues.apache.org/jira/browse/ZEPPELIN-1237 If there is anywhere else that permissions are used similarly to how they are implemented in interpreter or notebook permissions we should probably identify those and implement groups there for consistency as well -- This message was sent by Atlassian JIRA (v6.3.15#6346)
[jira] [Created] (ZEPPELIN-2354) Toolbar blocks majority of screen on mobile devices
Paul Brenner created ZEPPELIN-2354: -- Summary: Toolbar blocks majority of screen on mobile devices Key: ZEPPELIN-2354 URL: https://issues.apache.org/jira/browse/ZEPPELIN-2354 Project: Zeppelin Issue Type: Improvement Components: GUI Affects Versions: 0.7.1 Environment: iOS 10.3 - Safari Browser Reporter: Paul Brenner Priority: Minor Zeppelin works on mobile, which is amazing, but a few tweaks would really help improve the experience. 1. When working on a notebook the white toolbar (with,e.g., run all, show/hide code, and run scheduler buttons) takes up almost the entire screen. Is there a way to address this? Perhaps would it be possible collapse all toolbars (both the white toolbar and the blue menu bar) with a button or when entering text in a paragraph? 2. Paragraphs with long names can interfere with the buttons on the top right of the paragraph (e.g., run paragraph, collapse code) so that when trying to tap the run button the paragraph name will be selected instead. Is it possible to address this? Perhaps have the paragraph name wrap before it hits the buttons? -- This message was sent by Atlassian JIRA (v6.3.15#6346)
