[jira] [Commented] (SPARK-27169) number of active tasks is negative on executors page
[ https://issues.apache.org/jira/browse/SPARK-27169?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16934424#comment-16934424 ] Kevin Appel commented on SPARK-27169: - I have run into this similar issue before on jobs with over many thousand tasks, the events are getting dropped somewhere and the UI is showing gaps or anomalies in the metrics, such as the stages don't appear to be completed or the executors are showing negative metrics. Through trial and error using the following is giving reliable metrics now: --conf spark.scheduler.listenerbus.eventqueue.size=20 > number of active tasks is negative on executors page > > > Key: SPARK-27169 > URL: https://issues.apache.org/jira/browse/SPARK-27169 > Project: Spark > Issue Type: Bug > Components: Web UI >Affects Versions: 2.3.2 >Reporter: acupple >Priority: Minor > Attachments: QQ20190315-102215.png, QQ20190315-102235.png, > image-2019-03-19-15-17-25-522.png, image-2019-03-19-15-21-03-766.png, > job_1924.log, stage_3511.log > > > I use spark to process some data in HDFS and HBASE, I use one thread consume > message from a queue, and then submit to a thread pool（16 fix size）for spark > processor. > But when run for some time, the active jobs will be thousands, and number of > active tasks are negative. > Actually, these jobs are already done when I check driver logs。 > -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Commented] (SPARK-27169) number of active tasks is negative on executors page
[ https://issues.apache.org/jira/browse/SPARK-27169?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16797070#comment-16797070 ] acupple commented on SPARK-27169: - Thanks for you suggestion, and I will try increment the queue size and reproduce the case。 > number of active tasks is negative on executors page > > > Key: SPARK-27169 > URL: https://issues.apache.org/jira/browse/SPARK-27169 > Project: Spark > Issue Type: Bug > Components: Web UI >Affects Versions: 2.3.2 >Reporter: acupple >Priority: Minor > Attachments: QQ20190315-102215.png, QQ20190315-102235.png, > image-2019-03-19-15-17-25-522.png, image-2019-03-19-15-21-03-766.png, > job_1924.log, stage_3511.log > > > I use spark to process some data in HDFS and HBASE, I use one thread consume > message from a queue, and then submit to a thread pool（16 fix size）for spark > processor. > But when run for some time, the active jobs will be thousands, and number of > active tasks are negative. > Actually, these jobs are already done when I check driver logs。 > -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Commented] (SPARK-27169) number of active tasks is negative on executors page
[ https://issues.apache.org/jira/browse/SPARK-27169?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16797059#comment-16797059 ] acupple commented on SPARK-27169: - Can not find any "Dropping event" log, but some warn that "Dropped events from appStatus" > number of active tasks is negative on executors page > > > Key: SPARK-27169 > URL: https://issues.apache.org/jira/browse/SPARK-27169 > Project: Spark > Issue Type: Bug > Components: Web UI >Affects Versions: 2.3.2 >Reporter: acupple >Priority: Minor > Attachments: QQ20190315-102215.png, QQ20190315-102235.png, > image-2019-03-19-15-17-25-522.png, image-2019-03-19-15-21-03-766.png, > job_1924.log, stage_3511.log > > > I use spark to process some data in HDFS and HBASE, I use one thread consume > message from a queue, and then submit to a thread pool（16 fix size）for spark > processor. > But when run for some time, the active jobs will be thousands, and number of > active tasks are negative. > Actually, these jobs are already done when I check driver logs。 > -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Commented] (SPARK-27169) number of active tasks is negative on executors page
[ https://issues.apache.org/jira/browse/SPARK-27169?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16797066#comment-16797066 ] shahid commented on SPARK-27169: Yes. that means many event drops happens. Can you try increasing the queue size, "spark.scheduler.listenerbus.eventqueue.capacity" (default 1) might helps. If event drop happens, then UI display weirdly only, I'm not sure, from the UI side we can do anything. Do you have any reproducible steps for that, so that I can try? > number of active tasks is negative on executors page > > > Key: SPARK-27169 > URL: https://issues.apache.org/jira/browse/SPARK-27169 > Project: Spark > Issue Type: Bug > Components: Web UI >Affects Versions: 2.3.2 >Reporter: acupple >Priority: Minor > Attachments: QQ20190315-102215.png, QQ20190315-102235.png, > image-2019-03-19-15-17-25-522.png, image-2019-03-19-15-21-03-766.png, > job_1924.log, stage_3511.log > > > I use spark to process some data in HDFS and HBASE, I use one thread consume > message from a queue, and then submit to a thread pool（16 fix size）for spark > processor. > But when run for some time, the active jobs will be thousands, and number of > active tasks are negative. > Actually, these jobs are already done when I check driver logs。 > -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Commented] (SPARK-27169) number of active tasks is negative on executors page
[ https://issues.apache.org/jira/browse/SPARK-27169?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16797040#comment-16797040 ] shahid commented on SPARK-27169: Hi, It seems, from the above log we can't say that event drop has happened or not. Could you please check in the driver log that "Dropping event from queue" phrase is there or not? > number of active tasks is negative on executors page > > > Key: SPARK-27169 > URL: https://issues.apache.org/jira/browse/SPARK-27169 > Project: Spark > Issue Type: Bug > Components: Web UI >Affects Versions: 2.3.2 >Reporter: acupple >Priority: Minor > Attachments: QQ20190315-102215.png, QQ20190315-102235.png, > image-2019-03-19-15-17-25-522.png, image-2019-03-19-15-21-03-766.png, > job_1924.log, stage_3511.log > > > I use spark to process some data in HDFS and HBASE, I use one thread consume > message from a queue, and then submit to a thread pool（16 fix size）for spark > processor. > But when run for some time, the active jobs will be thousands, and number of > active tasks are negative. > Actually, these jobs are already done when I check driver logs。 > -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Commented] (SPARK-27169) number of active tasks is negative on executors page
[ https://issues.apache.org/jira/browse/SPARK-27169?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16796039#comment-16796039 ] acupple commented on SPARK-27169: - The full event log is too large. unknown stage log： [^stage_3511.log] > number of active tasks is negative on executors page > > > Key: SPARK-27169 > URL: https://issues.apache.org/jira/browse/SPARK-27169 > Project: Spark > Issue Type: Bug > Components: Web UI >Affects Versions: 2.3.2 >Reporter: acupple >Priority: Minor > Attachments: QQ20190315-102215.png, QQ20190315-102235.png, > image-2019-03-19-15-17-25-522.png, image-2019-03-19-15-21-03-766.png, > job_1924.log, stage_3511.log > > > I use spark to process some data in HDFS and HBASE, I use one thread consume > message from a queue, and then submit to a thread pool（16 fix size）for spark > processor. > But when run for some time, the active jobs will be thousands, and number of > active tasks are negative. > Actually, these jobs are already done when I check driver logs。 > -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Commented] (SPARK-27169) number of active tasks is negative on executors page
[ https://issues.apache.org/jira/browse/SPARK-27169?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16795878#comment-16795878 ] shahid commented on SPARK-27169: Thank you. Could you please provide full event log if possible? I think, some task events are missed in the eventlog > number of active tasks is negative on executors page > > > Key: SPARK-27169 > URL: https://issues.apache.org/jira/browse/SPARK-27169 > Project: Spark > Issue Type: Bug > Components: Web UI >Affects Versions: 2.3.2 >Reporter: acupple >Priority: Minor > Attachments: QQ20190315-102215.png, QQ20190315-102235.png, > image-2019-03-19-15-17-25-522.png, image-2019-03-19-15-21-03-766.png, > job_1924.log > > > I use spark to process some data in HDFS and HBASE, I use one thread consume > message from a queue, and then submit to a thread pool（16 fix size）for spark > processor. > But when run for some time, the active jobs will be thousands, and number of > active tasks are negative. > Actually, these jobs are already done when I check driver logs。 > -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Commented] (SPARK-27169) number of active tasks is negative on executors page
[ https://issues.apache.org/jira/browse/SPARK-27169?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16795774#comment-16795774 ] acupple commented on SPARK-27169: - I have enabled event log, on the spark UI, the job has not completed, but according to event log, the job has completed. !image-2019-03-19-15-17-25-522.png|width=872,height=48! the attachment is event log [^job_1924.log] And there is one unknown stage > number of active tasks is negative on executors page > > > Key: SPARK-27169 > URL: https://issues.apache.org/jira/browse/SPARK-27169 > Project: Spark > Issue Type: Bug > Components: Web UI >Affects Versions: 2.3.2 >Reporter: acupple >Priority: Minor > Attachments: QQ20190315-102215.png, QQ20190315-102235.png, > image-2019-03-19-15-17-25-522.png, job_1924.log > > > I use spark to process some data in HDFS and HBASE, I use one thread consume > message from a queue, and then submit to a thread pool（16 fix size）for spark > processor. > But when run for some time, the active jobs will be thousands, and number of > active tasks are negative. > Actually, these jobs are already done when I check driver logs。 > -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Commented] (SPARK-27169) number of active tasks is negative on executors page
[ https://issues.apache.org/jira/browse/SPARK-27169?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16794794#comment-16794794 ] shahid commented on SPARK-27169: Do you have eventlog corresponding to the application? > number of active tasks is negative on executors page > > > Key: SPARK-27169 > URL: https://issues.apache.org/jira/browse/SPARK-27169 > Project: Spark > Issue Type: Bug > Components: Web UI >Affects Versions: 2.3.2 >Reporter: acupple >Priority: Minor > Attachments: QQ20190315-102215.png, QQ20190315-102235.png > > > I use spark to process some data in HDFS and HBASE, I use one thread consume > message from a queue, and then submit to a thread pool（16 fix size）for spark > processor. > But when run for some time, the active jobs will be thousands, and number of > active tasks are negative. > Actually, these jobs are already done when I check driver logs。 > -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Commented] (SPARK-27169) number of active tasks is negative on executors page
[ https://issues.apache.org/jira/browse/SPARK-27169?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16794773#comment-16794773 ] acupple commented on SPARK-27169: - driver or executor logs? there are many logs, no errors, only spark UI display abnormally > number of active tasks is negative on executors page > > > Key: SPARK-27169 > URL: https://issues.apache.org/jira/browse/SPARK-27169 > Project: Spark > Issue Type: Bug > Components: Web UI >Affects Versions: 2.3.2 >Reporter: acupple >Priority: Minor > Attachments: QQ20190315-102215.png, QQ20190315-102235.png > > > I use spark to process some data in HDFS and HBASE, I use one thread consume > message from a queue, and then submit to a thread pool（16 fix size）for spark > processor. > But when run for some time, the active jobs will be thousands, and number of > active tasks are negative. > Actually, these jobs are already done when I check driver logs。 > -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Commented] (SPARK-27169) number of active tasks is negative on executors page
[ https://issues.apache.org/jira/browse/SPARK-27169?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16793675#comment-16793675 ] shahid commented on SPARK-27169: Seems event drop has happened. > number of active tasks is negative on executors page > > > Key: SPARK-27169 > URL: https://issues.apache.org/jira/browse/SPARK-27169 > Project: Spark > Issue Type: Bug > Components: Web UI >Affects Versions: 2.3.2 >Reporter: acupple >Priority: Minor > Attachments: QQ20190315-102215.png, QQ20190315-102235.png > > > I use spark to process some data in HDFS and HBASE, I use one thread consume > message from a queue, and then submit to a thread pool（16 fix size）for spark > processor. > But when run for some time, the active jobs will be thousands, and number of > active tasks are negative. > Actually, these jobs are already done when I check driver logs。 > -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
