You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@ignite.apache.org by "Ksenia Rybakova (JIRA)" <ji...@apache.org> on 2018/02/02 12:42:00 UTC

[jira] [Updated] (IGNITE-7262) Most of clients get stuck with "Failed to wait for partition map ..." during load test

     [ https://issues.apache.org/jira/browse/IGNITE-7262?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Ksenia Rybakova updated IGNITE-7262:
------------------------------------
    Description: 
Most of clients get stuck with "Failed to wait for partition map ..." during load test (preloading phase) with 500 logical caches or 30 physical. Others preload all data and perform operations successfully.

Load configs:

1)
 - CacheRandomOperationBenchmark
 - 14 client nodes, 42 server nodes
 - 50K perloading, 60K key range
 - 26 physical caches
 - 500 logical caches
 - 1 backup

2) 
 - CacheRandomOperationBenchmark
 - 8 client nodes, 40 server nodes
 - 200K perloading, 500K key range
 - 30 physical caches
 - 1 backup

Complete configs and log of one of the clients are attached.

  was:
Most of clients get stuck with "Failed to wait for partition map ..." during load test (preloading phase) with 500 logical caches. Others preload all data and perform operations successfully.

Load config:
- CacheRandomOperationBenchmark
- 14 client nodes, 42 server nodes
- 50K perloading, 60K key range
- 26 physical caches
- 500 logical caches
- 1 backup
Complete configs and log of one of the clients are attached.

        Summary: Most of clients get stuck with "Failed to wait for partition map ..." during load test  (was: Most of clients get stuck with "Failed to wait for partition map ..." during load test with 500 logical caches)

> Most of clients get stuck with "Failed to wait for partition map ..." during load test
> --------------------------------------------------------------------------------------
>
>                 Key: IGNITE-7262
>                 URL: https://issues.apache.org/jira/browse/IGNITE-7262
>             Project: Ignite
>          Issue Type: Bug
>          Components: general
>    Affects Versions: 2.1
>            Reporter: Ksenia Rybakova
>            Priority: Major
>         Attachments: 122310_id1_172.25.1.27_cache-random-benchmark-1-backup.log, ignite-base-load-config.xml, run-load.properties, run-load.xml
>
>
> Most of clients get stuck with "Failed to wait for partition map ..." during load test (preloading phase) with 500 logical caches or 30 physical. Others preload all data and perform operations successfully.
> Load configs:
> 1)
>  - CacheRandomOperationBenchmark
>  - 14 client nodes, 42 server nodes
>  - 50K perloading, 60K key range
>  - 26 physical caches
>  - 500 logical caches
>  - 1 backup
> 2) 
>  - CacheRandomOperationBenchmark
>  - 8 client nodes, 40 server nodes
>  - 200K perloading, 500K key range
>  - 30 physical caches
>  - 1 backup
> Complete configs and log of one of the clients are attached.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)