[jira] [Updated] (YARN-6193) FairScheduler might not trigger preemption when using DRF
[ https://issues.apache.org/jira/browse/YARN-6193?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Andrew Wang updated YARN-6193: -- Fix Version/s: 3.0.0-alpha3 > FairScheduler might not trigger preemption when using DRF > - > > Key: YARN-6193 > URL: https://issues.apache.org/jira/browse/YARN-6193 > Project: Hadoop YARN > Issue Type: Sub-task > Components: fairscheduler >Affects Versions: 2.8.0 >Reporter: Karthik Kambatla >Assignee: Karthik Kambatla > Fix For: 2.9.0, 3.0.0-alpha3 > > Attachments: YARN-6193.000.patch, YARN-6193.001.patch, > YARN-6193.002.patch, YARN-6193.003.patch > > > {{FSAppAttempt#canContainerBePreempted}} verifies preempting a container > doesn't lead to new starvation ({{Resources.fitsIn}}). When using DRF, this > leads to verifying both resources instead of just the dominant resource. > Ideally, the check should be based on policy. > Note that current implementation of > {{DominantResourceFairnessPolicy#checkIfUsageOverFairShare}} is broken. -- This message was sent by Atlassian JIRA (v6.3.15#6346) - To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
[jira] [Updated] (YARN-6193) FairScheduler might not trigger preemption when using DRF
[ https://issues.apache.org/jira/browse/YARN-6193?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Karthik Kambatla updated YARN-6193: --- Attachment: YARN-6193.003.patch > FairScheduler might not trigger preemption when using DRF > - > > Key: YARN-6193 > URL: https://issues.apache.org/jira/browse/YARN-6193 > Project: Hadoop YARN > Issue Type: Sub-task > Components: fairscheduler >Affects Versions: 2.8.0 >Reporter: Karthik Kambatla >Assignee: Karthik Kambatla > Attachments: YARN-6193.000.patch, YARN-6193.001.patch, > YARN-6193.002.patch, YARN-6193.003.patch > > > {{FSAppAttempt#canContainerBePreempted}} verifies preempting a container > doesn't lead to new starvation ({{Resources.fitsIn}}). When using DRF, this > leads to verifying both resources instead of just the dominant resource. > Ideally, the check should be based on policy. > Note that current implementation of > {{DominantResourceFairnessPolicy#checkIfUsageOverFairShare}} is broken. -- This message was sent by Atlassian JIRA (v6.3.15#6346) - To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
[jira] [Updated] (YARN-6193) FairScheduler might not trigger preemption when using DRF
[ https://issues.apache.org/jira/browse/YARN-6193?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Karthik Kambatla updated YARN-6193: --- Attachment: YARN-6193.002.patch Most test failures correspond to tests that mock FairScheduler. TestFairSchedulerPreemption is a timeout issue. Fixed them all in the 002.patch. > FairScheduler might not trigger preemption when using DRF > - > > Key: YARN-6193 > URL: https://issues.apache.org/jira/browse/YARN-6193 > Project: Hadoop YARN > Issue Type: Sub-task > Components: fairscheduler >Affects Versions: 2.8.0 >Reporter: Karthik Kambatla >Assignee: Karthik Kambatla > Attachments: YARN-6193.000.patch, YARN-6193.001.patch, > YARN-6193.002.patch > > > {{FSAppAttempt#canContainerBePreempted}} verifies preempting a container > doesn't lead to new starvation ({{Resources.fitsIn}}). When using DRF, this > leads to verifying both resources instead of just the dominant resource. > Ideally, the check should be based on policy. > Note that current implementation of > {{DominantResourceFairnessPolicy#checkIfUsageOverFairShare}} is broken. -- This message was sent by Atlassian JIRA (v6.3.15#6346) - To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
[jira] [Updated] (YARN-6193) FairScheduler might not trigger preemption when using DRF
[ https://issues.apache.org/jira/browse/YARN-6193?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Karthik Kambatla updated YARN-6193: --- Attachment: YARN-6193.001.patch > FairScheduler might not trigger preemption when using DRF > - > > Key: YARN-6193 > URL: https://issues.apache.org/jira/browse/YARN-6193 > Project: Hadoop YARN > Issue Type: Sub-task > Components: fairscheduler >Affects Versions: 2.8.0 >Reporter: Karthik Kambatla >Assignee: Karthik Kambatla > Attachments: YARN-6193.000.patch, YARN-6193.001.patch > > > {{FSAppAttempt#canContainerBePreempted}} verifies preempting a container > doesn't lead to new starvation ({{Resources.fitsIn}}). When using DRF, this > leads to verifying both resources instead of just the dominant resource. > Ideally, the check should be based on policy. > Note that current implementation of > {{DominantResourceFairnessPolicy#checkIfUsageOverFairShare}} is broken. -- This message was sent by Atlassian JIRA (v6.3.15#6346) - To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
[jira] [Updated] (YARN-6193) FairScheduler might not trigger preemption when using DRF
[ https://issues.apache.org/jira/browse/YARN-6193?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Karthik Kambatla updated YARN-6193: --- Issue Type: Sub-task (was: Bug) Parent: YARN-5990 > FairScheduler might not trigger preemption when using DRF > - > > Key: YARN-6193 > URL: https://issues.apache.org/jira/browse/YARN-6193 > Project: Hadoop YARN > Issue Type: Sub-task > Components: fairscheduler >Affects Versions: 2.8.0 >Reporter: Karthik Kambatla >Assignee: Karthik Kambatla > Attachments: YARN-6193.000.patch > > > {{FSAppAttempt#canContainerBePreempted}} verifies preempting a container > doesn't lead to new starvation ({{Resources.fitsIn}}). When using DRF, this > leads to verifying both resources instead of just the dominant resource. > Ideally, the check should be based on policy. > Note that current implementation of > {{DominantResourceFairnessPolicy#checkIfUsageOverFairShare}} is broken. -- This message was sent by Atlassian JIRA (v6.3.15#6346) - To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
[jira] [Updated] (YARN-6193) FairScheduler might not trigger preemption when using DRF
[ https://issues.apache.org/jira/browse/YARN-6193?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Karthik Kambatla updated YARN-6193: --- Attachment: YARN-6193.000.patch > FairScheduler might not trigger preemption when using DRF > - > > Key: YARN-6193 > URL: https://issues.apache.org/jira/browse/YARN-6193 > Project: Hadoop YARN > Issue Type: Bug > Components: fairscheduler >Affects Versions: 2.8.0 >Reporter: Karthik Kambatla >Assignee: Karthik Kambatla > Attachments: YARN-6193.000.patch > > > {{FSAppAttempt#canContainerBePreempted}} verifies preempting a container > doesn't lead to new starvation ({{Resources.fitsIn}}). When using DRF, this > leads to verifying both resources instead of just the dominant resource. > Ideally, the check should be based on policy. > Note that current implementation of > {{DominantResourceFairnessPolicy#checkIfUsageOverFairShare}} is broken. -- This message was sent by Atlassian JIRA (v6.3.15#6346) - To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
[jira] [Updated] (YARN-6193) FairScheduler might not trigger preemption when using DRF
[ https://issues.apache.org/jira/browse/YARN-6193?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Karthik Kambatla updated YARN-6193: --- Description: {{FSAppAttempt#canContainerBePreempted}} verifies preempting a container doesn't lead to new starvation ({{Resources.fitsIn}}). When using DRF, this leads to verifying both resources instead of just the dominant resource. Ideally, the check should be based on policy. Note that current implementation of {{DominantResourceFairnessPolicy#checkIfUsageOverFairShare}} is broken. was:{{DominantResourceFairnessPolicy#checkIfUsageOverFairShare}} considers all resources instead of just dominant resource. Ideally, it should use the DominantResourceCalculator. > FairScheduler might not trigger preemption when using DRF > - > > Key: YARN-6193 > URL: https://issues.apache.org/jira/browse/YARN-6193 > Project: Hadoop YARN > Issue Type: Bug > Components: fairscheduler >Affects Versions: 2.8.0 >Reporter: Karthik Kambatla >Assignee: Karthik Kambatla > > {{FSAppAttempt#canContainerBePreempted}} verifies preempting a container > doesn't lead to new starvation ({{Resources.fitsIn}}). When using DRF, this > leads to verifying both resources instead of just the dominant resource. > Ideally, the check should be based on policy. > Note that current implementation of > {{DominantResourceFairnessPolicy#checkIfUsageOverFairShare}} is broken. -- This message was sent by Atlassian JIRA (v6.3.15#6346) - To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
[jira] [Updated] (YARN-6193) FairScheduler might not trigger preemption when using DRF
[ https://issues.apache.org/jira/browse/YARN-6193?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Karthik Kambatla updated YARN-6193: --- Summary: FairScheduler might not trigger preemption when using DRF (was: FSAppAttempt#canContainerBePreempted considers all resources instead of just dominant resource) > FairScheduler might not trigger preemption when using DRF > - > > Key: YARN-6193 > URL: https://issues.apache.org/jira/browse/YARN-6193 > Project: Hadoop YARN > Issue Type: Bug > Components: fairscheduler >Affects Versions: 2.8.0 >Reporter: Karthik Kambatla >Assignee: Karthik Kambatla > > {{DominantResourceFairnessPolicy#checkIfUsageOverFairShare}} considers all > resources instead of just dominant resource. Ideally, it should use the > DominantResourceCalculator. -- This message was sent by Atlassian JIRA (v6.3.15#6346) - To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org