[jira] [Updated] (YARN-3669) Attempt-failures validatiy interval should have a global admin configurable lower limit
[ https://issues.apache.org/jira/browse/YARN-3669?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Xuan Gong updated YARN-3669: Attachment: YARN-3669.2.patch > Attempt-failures validatiy interval should have a global admin configurable > lower limit > --- > > Key: YARN-3669 > URL: https://issues.apache.org/jira/browse/YARN-3669 > Project: Hadoop YARN > Issue Type: Sub-task > Components: resourcemanager >Reporter: Vinod Kumar Vavilapalli >Assignee: Xuan Gong > Labels: newbie > Attachments: YARN-3669.1.patch, YARN-3669.2.patch > > > Found this while reviewing YARN-3480. > bq. When 'attemptFailuresValidityInterval'(introduced in YARN-611) is set to > a small value, retried attempts might be very large. So we need to delete > some attempts stored in RMStateStore and RMStateStore. > I think we need to have a lower limit on the failure-validaty interval to > avoid situations like this. > Having this will avoid pardoning too-many failures in too-short a duration. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (YARN-3669) Attempt-failures validatiy interval should have a global admin configurable lower limit
[ https://issues.apache.org/jira/browse/YARN-3669?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Xuan Gong updated YARN-3669: Attachment: YARN-3669.1.patch > Attempt-failures validatiy interval should have a global admin configurable > lower limit > --- > > Key: YARN-3669 > URL: https://issues.apache.org/jira/browse/YARN-3669 > Project: Hadoop YARN > Issue Type: Sub-task > Components: resourcemanager >Reporter: Vinod Kumar Vavilapalli >Assignee: Vinod Kumar Vavilapalli > Labels: newbie > Attachments: YARN-3669.1.patch > > > Found this while reviewing YARN-3480. > bq. When 'attemptFailuresValidityInterval'(introduced in YARN-611) is set to > a small value, retried attempts might be very large. So we need to delete > some attempts stored in RMStateStore and RMStateStore. > I think we need to have a lower limit on the failure-validaty interval to > avoid situations like this. > Having this will avoid pardoning too-many failures in too-short a duration. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (YARN-3669) Attempt-failures validatiy interval should have a global admin configurable lower limit
[ https://issues.apache.org/jira/browse/YARN-3669?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Jian He updated YARN-3669: -- Assignee: Xuan Gong (was: Vinod Kumar Vavilapalli) > Attempt-failures validatiy interval should have a global admin configurable > lower limit > --- > > Key: YARN-3669 > URL: https://issues.apache.org/jira/browse/YARN-3669 > Project: Hadoop YARN > Issue Type: Sub-task > Components: resourcemanager >Reporter: Vinod Kumar Vavilapalli >Assignee: Xuan Gong > Labels: newbie > Attachments: YARN-3669.1.patch > > > Found this while reviewing YARN-3480. > bq. When 'attemptFailuresValidityInterval'(introduced in YARN-611) is set to > a small value, retried attempts might be very large. So we need to delete > some attempts stored in RMStateStore and RMStateStore. > I think we need to have a lower limit on the failure-validaty interval to > avoid situations like this. > Having this will avoid pardoning too-many failures in too-short a duration. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (YARN-3669) Attempt-failures validatiy interval should have a global admin configurable lower limit
[ https://issues.apache.org/jira/browse/YARN-3669?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Xuan Gong updated YARN-3669: Labels: newbie (was: ) Attempt-failures validatiy interval should have a global admin configurable lower limit --- Key: YARN-3669 URL: https://issues.apache.org/jira/browse/YARN-3669 Project: Hadoop YARN Issue Type: Sub-task Components: resourcemanager Reporter: Vinod Kumar Vavilapalli Assignee: Vinod Kumar Vavilapalli Labels: newbie Found this while reviewing YARN-3480. bq. When 'attemptFailuresValidityInterval'(introduced in YARN-611) is set to a small value, retried attempts might be very large. So we need to delete some attempts stored in RMStateStore and RMStateStore. I think we need to have a lower limit on the failure-validaty interval to avoid situations like this. Having this will avoid pardoning too-many failures in too-short a duration. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
