#general


@zineb.raiiss: Hello friends, I want to test the ThirdEye solution for Pinot anomaly detection, so I followed the documentation , but failed to connect to
  @cyril: any logs with ```docker logs thirdeye -f``` ?
  @zineb.raiiss: I used this command, this is what it gave me as output
  @zineb.raiiss:
@zineb.raiiss: do you have any idea on that?

#troubleshooting


@wentjin: Hi team - we are using `Upsert` to do some use cases, and I found that there will be a large size map `_primaryKeyToRecordLocationMap` in Pinot server, A partition of an instance can reach tens of millions of data. Seems this map can only release some keys when the old segment destroy. Can we consider adding some strategies to expire keys by time, because most of the data will not be updated after a period of time. Just a suggestion.
@anu110195: Hey, Pinot server got crashed unable to see /query data
@anu110195: on server logs i found these exceptions
@anu110195: Exception: java.lang.OutOfMemoryError thrown from the UncaughtExceptionHandler in thread "nioEventLoopGroup-3-1" AsyncLogger error handling event seq=AsyncLogger error handling event seq=AsyncLogger error handling event seq=AsyncLogger error handling event seq=AsyncLogger error handling event seq=AsyncLogger error handling event seq=AsyncLogger error handling event seq=AsyncLogger error handling event seq=10, value='Exception in thread "pqr-0" null': java.lang.OutOfMemoryError: Java heap space java.lang.OutOfMemoryError: Java heap space java.lang.OutOfMemoryError: Java heap space
@anu110195: do we need to increase jvm xmx limit ?
  @dadelcas: Indeed
  @anu110195: No luck after updating, full cluster is down.
  @anu110195: getting this error, 2021/09/15 11:35:54.924 WARN [PinotInstanceRestletResource] [grizzly-http-server-1] Grpc port is not set for instance: Controller_pinot2-controller-0.pinot2-controller-headless.data2.svc.cluster.local_9000 java.lang.NumberFormatException: null at java.lang.Integer.parseInt(Integer.java:542) ~[?:1.8.0_282] at java.lang.Integer.parseInt(Integer.java:615) ~[?:1.8.0_282] at org.apache.pinot.controller.api.resources.PinotInstanceRestletResource.getGrpcPort(PinotInstanceRestletResource.java:119) ~[pinot-all-0.7.0-SNAPSHOT-jar-with-dependencies.jar:0.7.0-SNAPSHOT-7bcbee5058c6a56d02b2d34e173185009ac35146] at org.apache.pinot.controller.api.resources.PinotInstanceRestletResource.getInstance(PinotInstanceRestletResource.java:107) ~[pinot-all-0.7.0-SNAPSHOT-jar-with-dependencies.jar:0.7.0-SNAPSHOT-7bcbee5058c6a56d02b2d34e173185009ac35146]
  @anu110195: @mayanks any clue on this issue?
  @mayanks: Are you using controller to do queries instead of broker?
  @anu110195: Not sure on this, need to check . we have 2 services which makes connection with pinot for making queries but we also do some test queries on pinot dashboard for debugging
  @xiangfu0: Did you set resources required and limit both for all Pinot statefulset?
  @anu110195: no, where exactly to set , can you share some docs for reference
@ebuth: Hey all! We have a metric column that we’re unsure is indexed correctly. Running `SELECT metric_column FROM table ORDER BY metric_column DESC LIMIT 10` results in a full table scan. So far we’ve followed the instructions for setting up a raw value forward index, but it doesn’t seem to help.
  @mayanks: Indexing helps with filtering, your query is essentially sorting all rows of the column so it will have a full table scan
@anu110195: Hey, After running a query in dashboard , zookeeper got crashed
@anu110195: im trying to restart it, but somehow its always getting crashed
@anu110195: in logs i can see storage issues
@anu110195: after debugging i found /data directory in zookeeper pod is getting full
@anu110195: can anyone please help in this ?
@g.kishore: @dlavoie ^^ do we have a values. Yaml for prod
@mayanks: @anu110195 you can config  environment variables : `ZK_SNAP_RETAIN_COUNT=3`  and `ZK_PURGE_INTERVAL=1`  to zk statefulset

#pinot-dev


@atri.sharma: Folks, I need some help. My latest PR adds JMH benchmarks, hence generates relevant classes. However, Pre commit is failing for the same with checkstyle failures. I thought checkstyle is excluded for /target?
  @xiangfu0: have you tried this compatible verifier script from your local?
  @xiangfu0: you can try to exclude those paths from the pom
  @atri.sharma: Yes, and strangely, it passes for me
  @atri.sharma: That. is what is confusing, I thought. it is already excluded
  @xiangfu0: you can add some exclude in config/checkstyle.xml
  @xiangfu0: actually it’s `config/suppressions.xml`
  @atri.sharma: Let me take a loo, thanks!
  @xiangfu0: exclude `target/generated-test-sources`
  @xiangfu0: right now we only exclude ``` <!-- Suppress autogenerated parser files --> <suppress checks=".*" files="target/generated-sources"/>```
  @atri.sharma: Ah yes, my bad. I somehow misread as a complete exclusion of target folder
  @atri.sharma: which, btw, is what we should be doing?
  @xiangfu0: I think it’s fine to exclude `target`
  @atri.sharma: Ok, let me add a separate PR for it, thanks for the help!
--------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@pinot.apache.org For additional commands, e-mail: dev-h...@pinot.apache.org

Reply via email to