<h3><u>#general</u></h3><br><strong>@eran.amrami: </strong>@eran.amrami has joined the channel<br><strong>@venkatram.bondugula: </strong>@venkatram.bondugula has joined the channel<br><strong>@mayanks: </strong>Hello, LinkedIn is hosting a virtual meetup for Pinot on Sept 2, 6pm. Please join the meetup page <https://u17000708.ct.sendgrid.net/ls/click?upn=1BiFF0-2FtVRazUn1cLzaiMfkD15963Wy9daymeW9L-2Bh3w7-2Bn3gGS4tWRaxy3iDlc8hNRHvWvLwvoPmKvDBo8MCw-3D-3Dmdv__vGLQYiKGfBLXsUt3KGBrxeq6BCTMpPOLROqAvDqBeTwLoGbWXlcReQrdqNl5iOfIMeywaJt6ZJxAJ15aF9CL0MNA0resGSKWs8p64x7XEo7qe08Gp-2Bxm-2FcVhOPd9xwCaTfcTaXIVvkiVIp2IgoZF0H6dMpoO22JsT97aljnUWkZn86NmadaoMWVw8salixcUmqQwrDhrZZD2nb-2B7X7KCLR3yEZhVuUlfdhfwWIV08ns-3D>. Registration link in the `Details` section.<br><strong>@mayanks: </strong>Note, in order to get the zoom link, please register at <https://u17000708.ct.sendgrid.net/ls/click?upn=1BiFF0-2FtVRazUn1cLzaiMUHWLijFWt2bRMjF3IjDMnO1Cs7Q9Cn51P3DCI-2B9Pbi5oC-2FfjN80gS-2Fka5BAkKC2iQ-3D-3DgKwE_vGLQYiKGfBLXsUt3KGBrxeq6BCTMpPOLROqAvDqBeTwLoGbWXlcReQrdqNl5iOfI1fiNSAIjZVz1m-2F0aw0NcLrP9pRPmBFsZuCHflJbKMk6YuCvHH3l3ygpLDY09YdCQjzcIkVusphRqq70DjM-2FKmdZ7399OqFerZ-2BoI1Hs7IeMmcASjv1YBviB-2BkI8dlLcRQlp-2BHCcIGdaGKsRv-2F8VzvmRsIbdkD3-2BkXPABa4oGt0o-3D><br><strong>@samuel: </strong>@samuel has joined the channel<br><strong>@samuel: </strong>hey everyone, seems like images are missing for the indexing section of the documenation: <https://u17000708.ct.sendgrid.net/ls/click?upn=1BiFF0-2FtVRazUn1cLzaiMdTeAXadp8BL3QinSdRtJdp7hEkgJimgu1J1h82HsMQqT6nVAGJ7v5KEfb9cxDSusXoY56WlEnDWhqn8kkuVirs-3DjBPj_vGLQYiKGfBLXsUt3KGBrxeq6BCTMpPOLROqAvDqBeTwLoGbWXlcReQrdqNl5iOfIl4EGh4290LA7nWpV0-2FnQ4CEEi1rdFOlEIka1Kee6UmZk3-2BiGzbSwBfTshuwaJsaMqJPx688P9rKokwTuM0zQhZdX6gtMAHnGP1jlXAP8WkMG6UKfUvHJFtTnE8dLY9gC270viKvjcNpANSC-2BdqJNmVJcex-2BMyf0zycpvtze-2BGRw-3D><br><h3><u>#random</u></h3><br><strong>@eran.amrami: </strong>@eran.amrami has joined the channel<br><strong>@venkatram.bondugula: </strong>@venkatram.bondugula has joined the channel<br><strong>@samuel: </strong>@samuel has joined the channel<br><h3><u>#troubleshooting</u></h3><br><strong>@quietgolfer: </strong>Does Pinot have a way to create an alias to a specific table? I'm thinking about the situation where I want to make a large change to a table and I'll need to recreate it. Can I use an alias and do the swap inside Pinot? Or would I want a layer outside of Pinot to convert this alias table name to a specific Pinot table?<br><strong>@samuel: </strong>@samuel has joined the channel<br><h3><u>#presto-pinot-streaming</u></h3><br><strong>@surendra: </strong>@surendra has joined the channel<br><h3><u>#pinot-docs</u></h3><br><strong>@kharekartik: </strong>Added this CLI reference doc <https://u17000708.ct.sendgrid.net/ls/click?upn=1BiFF0-2FtVRazUn1cLzaiMdTeAXadp8BL3QinSdRtJdoH1omTMY3qwbAUC5IutlmIVpWJt2TfLgpcnnOUu5u7bQ-3D-3D9F4U_vGLQYiKGfBLXsUt3KGBrxeq6BCTMpPOLROqAvDqBeTwLoGbWXlcReQrdqNl5iOfI1Xl8U48YfJayCxYvmeBhMmUPv4yJv4FSMyv7pRo-2BFTuHsNdwXdqNB-2B3mIrNzk7wPxUB9ts-2BH5igB1eZzX97-2BTlorzaHkfIYYNAcHQkADD00I4rz17aS6bvWs4TZfzT-2B68GG-2BVoQSaTHycXb0DPm8R1crG2jB8ALfhw615L8xqQk-3D> Let me know if anything is missing or needs to be added some commands which are not present ``` @SubCommand(name = "CollectMetadataForIndexTuning", impl = CollectMetadataForIndexTuning.class), @SubCommand(name = "EntriesScannedQuantileReport", impl = EntriesScannedQuantileReport.class), @SubCommand(name = "IndexTuner", impl = IndexTunerCommand.class), @SubCommand(name = "AnonymizeData", impl = AnonymizeDataCommand.class),```<br><strong>@kharekartik: </strong>I need some context when and where these commands are used<br><strong>@kharekartik: </strong><!here><br><strong>@mayanks: </strong>@steotia ^^<br><strong>@g.kishore: </strong>@kharekartik this is awesome<br><strong>@npawar: </strong>wow this is really well done!<br><strong>@mayanks: </strong>Yes :heavy_plus_sign:<br><strong>@fx19880617: </strong>super!<br><h3><u>#aggregate-metrics-change</u></h3><br><strong>@steotia: </strong>@steotia has joined the channel<br><strong>@jackie.jxt: </strong>@jackie.jxt has joined the channel<br><strong>@mayanks: </strong>@mayanks has joined the channel<br><strong>@steotia: </strong>Hey @jackie.jxt, it looks like this PR <https://u17000708.ct.sendgrid.net/ls/click?upn=1BiFF0-2FtVRazUn1cLzaiMSfW2QiSG4bkQpnpkSL7FiK3MHb8libOHmhAW89nP5XKCGn0A80vwG8LNt3DR3OvdyAsBA90ApTnwRj1wTM6R3M-3DkCV2_vGLQYiKGfBLXsUt3KGBrxeq6BCTMpPOLROqAvDqBeTwLoGbWXlcReQrdqNl5iOfIqQCCEKhUkT9T0QHngY-2FRqA-2BvNR91zf2E05mAFCvOrGHsXsnURxJqYXbe6SLuSdGGgIV9E66UEC-2FA-2BLbzMMQL4jWxOxXH3eJXnKxLQB6XbvS3tqXim2QaDsHn6PHYdJ9zUAoKTO3aMkGAC8g7jryaeb5Ri22rhe7-2BU3zYZZBqB04-3D><br><strong>@steotia: </strong>has made a change where if aggregateMetrics is turned on<br><strong>@steotia: </strong>the min and max value will not be computed<br><strong>@steotia: </strong>for the metric column<br><strong>@jackie.jxt: </strong>Yes, because the min/max value will be inaccurate<br><strong>@steotia: </strong>so when the consumption is completed and we commit/build the offline segment, how do we get the min/max value in the stats<br><strong>@jackie.jxt: </strong>It won't be there<br><strong>@steotia: </strong>which means we are going to hit NPE<br><strong>@steotia: </strong>```String minValue = columnIndexCreationInfo.getMin().toString(); String maxValue = columnIndexCreationInfo.getMax().toString();```<br><strong>@steotia: </strong>in SegmentColumnarIndexCreator<br><strong>@steotia: </strong>``` 2020/08/13 21:25:30.777 ERROR [LLRealtimeSegmentDataManager] [in_app_notification_persisted_event_additive__0__0__20200813T2117Z] [ pinot-server] [] Could not build segment 2482870 java.lang.NullPointerException: null 2482871 at org.apache.pinot.core.segment.creator.impl.SegmentColumnarIndexCreator.addColumnMetadataInfo(SegmentColumnarIndexCreator.java:535) ~[pinot-core-0.5.0-dev-18120.jar:0.5.0-dev-18120-c9975c5 149e3ee94751df062bed49562185ebd2b] 2482872 at org.apache.pinot.core.segment.creator.impl.SegmentColumnarIndexCreator.writeMetadata(SegmentColumnarIndexCreator.java:489) ~[pinot-core-0.5.0-dev-18120.jar:0.5.0-dev-18120-c9975c5149e3ee9 4751df062bed49562185ebd2b] 2482873 at org.apache.pinot.core.segment.creator.impl.SegmentColumnarIndexCreator.seal(SegmentColumnarIndexCreator.java:399) ~[pinot-core-0.5.0-dev-18120.jar:0.5.0-dev-18120-c9975c5149e3ee94751df062 bed49562185ebd2b] 2482874 at org.apache.pinot.core.segment.creator.impl.SegmentIndexCreationDriverImpl.handlePostCreation(SegmentIndexCreationDriverImpl.java:240) ~[pinot-core-0.5.0-dev-18120.jar:0.5.0-dev-18120-c997 5c5149e3ee94751df062bed49562185ebd2b] 2482875 at org.apache.pinot.core.segment.creator.impl.SegmentIndexCreationDriverImpl.build(SegmentIndexCreationDriverImpl.java:223) ~[pinot-core-0.5.0-dev-18120.jar:0.5.0-dev-18120-c9975c5149e3ee947 51df062bed49562185ebd2b] 2482876 at org.apache.pinot.core.realtime.converter.RealtimeSegmentConverter.build(RealtimeSegmentConverter.java:127) ~[pinot-core-0.5.0-dev-18120.jar:0.5.0-dev-18120-c9975c5149e3ee94751df062bed4956 2185ebd2b] 2482877 at org.apache.pinot.core.data.manager.realtime.LLRealtimeSegmentDataManager.buildSegmentInternal(LLRealtimeSegmentDataManager.java:742) [pinot-core-0.5.0-dev-18120.jar:0.5.0-dev-18120-c9975c 5149e3ee94751df062bed49562185ebd2b] 2482878 at org.apache.pinot.core.data.manager.realtime.LLRealtimeSegmentDataManager.buildSegmentForCommit(LLRealtimeSegmentDataManager.java:693) [pinot-core-0.5.0-dev-18120.jar:0.5.0-dev-18120-c9975 c5149e3ee94751df062bed49562185ebd2b] 2482879 at org.apache.pinot.core.data.manager.realtime.LLRealtimeSegmentDataManager$PartitionConsumer.run(LLRealtimeSegmentDataManager.java:604) [pinot-core-0.5.0-dev-18120.jar:0.5.0-dev-18120-c9975 c5149e3ee94751df062bed49562185ebd2b]```<br><strong>@mayanks: </strong>Where are we seeing this issue?<br><strong>@mayanks: </strong>In ei?<br><strong>@steotia: </strong>I am seeing this NPE for all tables in CORPt have aggregateMetrics as true<br><strong>@mayanks: </strong>In prod?<br><strong>@steotia: </strong>corp<br><strong>@steotia: </strong>this release is not in prod yet<br><strong>@steotia: </strong>it is there in EI<br><strong>@mayanks: </strong>Who needs that in Corp?<br><strong>@jackie.jxt: </strong>Please rollback and add the null check<br><strong>@steotia: </strong>weird name tables -- in_app_notification_persisted_event_additive, message_delivery_event_hourly_additive<br><strong>@jackie.jxt: </strong>The aggregate metric test should be able to catch it<br><strong>@jackie.jxt: </strong>Let me check<br><strong>@mayanks: </strong>Oh man<br><strong>@steotia: </strong>there are several tables<br><strong>@steotia: </strong>corp realtime server log is flooded with NPEs<br><strong>@steotia: </strong>and the segment commit is failing<br><strong>@mayanks: </strong>That is a serious issue<br><strong>@mayanks: </strong>When was the deployment done?<br><strong>@steotia: </strong>we should rollback but I first wanted to check with Jackie about my hypothesis<br><strong>@steotia: </strong>so I am guessing this is the deployment taken to corp last week<br><strong>@steotia: </strong>this week there hasn't been anything<br><strong>@mayanks: </strong>No one complained?<br><strong>@steotia: </strong>One customer complained for realtime table not returning any data and I found this in the logs<br><strong>@steotia: </strong>I checked EI realtime server logs too<br><strong>@steotia: </strong>but don't see this error<br><strong>@mayanks: </strong>Okay, what’s the Kafka consumption rate?<br><strong>@steotia: </strong>pribably because no one is having aggregateMetrics enabled in EI<br><strong>@mayanks: </strong>If low quick fix is to disable the metric aggregate <br><strong>@mayanks: </strong>How many tables?<br><strong>@mayanks: </strong>@g.kishore <br><strong>@g.kishore: </strong>@g.kishore has joined the channel<br><strong>@steotia: </strong>quite a lot<br><strong>@mayanks: </strong>@ssubrama <br><strong>@ssubrama: </strong>@ssubrama has joined the channel<br><strong>@steotia: </strong>I will have to put together a list by going through all corp realtime server logs<br><strong>@jackie.jxt: </strong>Rollback should automatically fix the issue<br><strong>@mayanks: </strong>That is not fix <br><strong>@mayanks: </strong>That is mitigate :grin:<br><strong>@jackie.jxt: </strong>I'll submit a fix soon, but rollback should "mitigate" the issue<br><strong>@mayanks: </strong>Yeah so will disable the featur<br><strong>@mayanks: </strong>I am asking if that feature is needed for this table <br><strong>@mayanks: </strong>@steotia <br><strong>@mayanks: </strong>Because rollback means we have to start again <br><strong>@steotia: </strong>I am checking with the owner of one table that complained initially<br><strong>@steotia: </strong>first I need a list of tables<br><strong>@mayanks: </strong>There are not many RT <br><strong>@mayanks: </strong>We can go with hotfix <br><strong>@jackie.jxt: </strong>IMO, we should rollback the realtime server first<br><strong>@jackie.jxt: </strong>Let it resume consuming<br><strong>@g.kishore: </strong>Let’s add integration test with aggregate metrics turned on<br><strong>@mayanks: </strong>Just fyi I only tagged @g.kishore and @ssubrama so we take this into account for release certifier.<br><strong>@g.kishore: </strong>Thanks, will do<br>
